From wjk at wjk.mv.com  Sat Jan  8 16:13:57 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Sat, 08 Jan 2000 16:13:57 -0500
Subject: does a variable exist
References: <B49BBF60.1373%Dan@Grassi.com> <dQsd4.586$mj7.15972@typhoon.nyroc.rr.com> <38767A50.CFDEC3C2@wjk.mv.com> <Nvxd4.691$mj7.17561@typhoon.nyroc.rr.com>
Message-ID: <3877A894.4019F528@wjk.mv.com>

Thanks -- was just wondering if there was
more overhead in catching/throwing exceptions...
(still learning)       (c;{

Darrell wrote:

> "Wm. King" <wjk at wjk.mv.com> wrote in message
> news:38767A50.CFDEC3C2 at wjk.mv.com...
> > Could you do something like:
> >
> > try:
> >     print x
> > except NameError:
> >    print "No global or local x"
> >
> > Just wondering?
> >
>
> Yulp.
> You see that a lot for imports also.
> try:
>     from _socket import *
> except ImportError:
>     from socket import *
>
> Throwing exceptions takes time. Not a problem for imports but might be a
> problem when accessing variables.
>
> Here's another way to say the same thing.
> >>> myVar=1
> >>> import sys
> >>> globals().get('myVar',None)
> 1
> >>> globals().get('xxxx',None)
> >>>
> >>> class VarError:
> ...     def __init__(self, msg):
> ...             self._msg=msg
> ...     def __str__(self):
> ...             return self._msg
> ...
> >>>
> >>> print globals().get('xxxx',VarError("No global or local xxxx"))
> No global or local xxxx
> >>>
>
> --
> --Darrell



From mchung at mindspring.com  Tue Jan 11 00:17:20 2000
From: mchung at mindspring.com (Mitchell Chung)
Date: Tue, 11 Jan 2000 00:17:20 -0500
Subject: AXScript problem...
Message-ID: <85ehtk$g2b$1@nntp1.atl.mindspring.net>

I'm using Windows NT and IE 5.0 with Python and the AXScript implementation
in win32com.  The demos in win32comext\axscript\demos\client\ie, like
calc.htm, seem to work.  But when I try my own code I'm getting errors with
stuff like "import copy" and "import httplib".  I haven't been able to find
a solution in the docs and the Python web site yet.  Can anyone help?

For example in calc.htm if I have
<SCRIPT LANGUAGE="Python">
import copy
...

"import copy" gives this error message in IE 5.0:
Attribute Error: CodeType

"import httplib" gives this error message:
NameError: socket

I don't have these errors when I use python or jpython by itself.

Mitch




From jblaine at shell2.shore.net  Fri Jan 14 11:08:08 2000
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Fri, 14 Jan 2000 16:08:08 GMT
Subject: A Python bug? (core dump)
References: <slrn87spbq.16i.huaiyu_zhu@localhost.localdomain>
Message-ID: <slrn87uif8.a04.jblaine@shell2.shore.net>

It doesn't coredump under our build of Python 1.5.2 (or 1.5.1) for Solaris
7.  I eventually get RuntimeError: Maximum recursion depth exceeded

On 13 Jan 2000 17:52:52 -0600, Huaiyu Zhu <huaiyu_zhu at moon.localdomain> wrote:
>The following script dumps core when the "print self" line is uncommented,
>because print calls __str__ which calls print which calls ...
>
>##################################
>class Item:
>    def __init__(self, item):
>        self._ = item
>	    
>    def __str__(self):
>#       print self
>        return `self._`
>
>print Item ('aaa')
>##################################
>
>For some other recursive calls I tried that do not involve print and
>__str__, the outcomes are sane:
>
>RuntimeError: Maximum recursion depth exceeded
>
>I've read somewhere that "Python never dumps core".  So this must be a bug.
>The versions are:
>
>Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
>(egcs- on linux-i386


From matthias.huening at univie.ac.at  Sat Jan  8 09:23:47 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Sat, 8 Jan 2000 15:23:47 +0100
Subject: RegExp, Python and strings
Message-ID: <857hhb$3kmi$1@www.univie.ac.at>

Hi,
I'm new to PYTHON and I am trying to find out if PYTHON is the right
language for work with strings (which is what I need most). Therefore
I am playing around with the RE-module.

In PERL I can do things like this (in one line):
$A = "Rossum, Guido van; Harms, Daryl; Python, Franz-Josef";
$A =~ s/([A-Z])[\w]+(?![ \w\-]+,)([ ;-]|$)/$1.$2/g;
This RegExp results in: "Rossum, G. van; Harms, D.; Python, F.-J."

Now I am trying to rebuild this in PYTHON, but I can't get it to work.
Any hints? Should I keep trying? Or should I stick to PERL for those
kinds of string-manipulations?

Matthias

- - - - -
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/





From wtanksle at hawking.armored.net  Thu Jan  6 19:44:27 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 7 Jan 2000 00:44:27 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com> <8537fi$jta$1@nnrp1.deja.com>
Message-ID: <slrn87ae87.c7g.wtanksle@hawking.armored.net>

On Thu, 06 Jan 2000 23:11:55 GMT, Preston Landers wrote:
>In article <14452.45436.641890.197509 at beluga.mojam.com>,
>  skip at mojam.com (Skip Montanaro) wrote:

>> I find nothing wrong with it.  Some people object to using try/except
>> for anything but handling errors however.

>IMHO these people need to be taken out, shown a good time, maybe a
>little wine and dancing.  you know, let off a little steam.

Sounds good.  I didn't used to worry about it, but I guess I'm willing to
change.  ;-)

>I've had at least one situation where it made perfect sense to use
>try/except to handle non-error conditions.  In fact, I had defined a
>bunch of exceptions relating to specific conditions, and it also made
>sence to define a success condition and catch that rather than handle it
>as a special case (different control structure.)

A special case is another word for an exception.  Thus you didn't do
anything out of line.

>|| Preston Landers <prestonlanders at my-deja.com> ||

-- 
-William "Billy" Tanksley, in hoc signo hack


From martinp at mincom.com  Tue Jan  4 22:34:47 2000
From: martinp at mincom.com (Martin Pool)
Date: Wed, 05 Jan 2000 13:34:47 +1000
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <386E7BD4.514764EA@ndh.net> <84nv0u$bu9$1@nntp9.atl.mindspring.net> <3870DB37.761034EA@roguewave.com> <75B5A3A6EB0EDE04.FCE8CED8D9B414D5.95006D8FA0ECEB31@lp.airnews.net> <3870E485.998FF2FD@roguewave.com>
Message-ID: <3872BBD7.16FA16E@mincom.com>

bjorn wrote:
> and (ii) they must remain unchanged for the duration of the container (or more
      ^^^^

You know you were up too late writing Python wrappers if this parses as
"unpack tuple of two ints". ;-)

-- 
Martin

emacs outshines all other editing software in approximately the same
way that the noonday sun does the stars. It is not just bigger and
brighter; it simply makes everything else vanish.
			-- Neal Stephenson


From Vladimir.Marangozov at inrialpes.fr  Wed Jan 26 09:59:48 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Wed, 26 Jan 2000 15:59:48 +0100
Subject: Stackless Python 1.0 + Continuations 0.6
References: <mt2.0-27365-948450827@news.informatik.uni-bonn.de> <86f8uu$3ds$1@kopp.stud.ntnu.no> <388BDEFA.8D57A66F@tismer.com> <388C91E1.500B272A@hassler.de> <388EB066.3DACCDEE@tismer.com>
Message-ID: <388F0BE4.B1F1AE8A@inrialpes.fr>

Christian Tismer wrote:
> 
> Gerald Klix wrote:
> >
> > Christian Tismer wrote:
> >
> > > Magnus Lie Hetland wrote:
> ...
> > > > would it be possible to stop and pickle a thread? This might be very
> > > > interesting for Mobile Agent Technology
> ...
> > Chris, please add this point to your arguments for stackless python!
> 
> I did so. Beneath the PalmIII-opportunity, it seems to be
> a major argument. Thanks a lot!
> 

Easy. Capturing the thread state, i.e. the execution stack + the set of
references within the stack pointing to existing (shared) objects, seems
to be a no brainer, but deducing from here that thread migration is
easier
in stackless, compared to stackfull Python, is a trap.

Assuming that both versions of the interpreter have, at some point in
the
execution of a program, their thread state available for further
manipulation
(be it check-pointing for an eventual future error recovery, or
partial/full
serialization of the objects accessible from that state for an eventual
thread migration), I hardly see any argument why stackless is better,
not
to say any different, than stackfull Python. So thread migration is not
much
of an argument you could rely on... Capturing the full thread *state*
is.

In case you have, in your thread state, a pointer to a dict, which
happens
to contain the sys module, a strawman thread migration scheme would
imply
the serialization of the whole interpreter...

Rest assured -- the 4k Palm argument has still a chance, though ;-)
-- 
       Vladimir MARANGOZOV          | Vladimir.Marangozov at inrialpes.fr
http://sirac.inrialpes.fr/~marangoz | tel:(+33-4)76615277 fax:76615252


From smb at bby.com.au  Mon Jan 10 18:13:50 2000
From: smb at bby.com.au (Sarah Burke)
Date: Tue, 11 Jan 2000 09:13:50 +1000
Subject: Zope and NT databases
Message-ID: <387A67AD.32F2E22B@bby.com.au>

Well, how about if I was using a client/server database, like Microsoft's SQL
server?

Sarah

Jochen Knuth wrote:

> Hi,
>
> Bill Scherer wrote:
> >
> > You want ZODBCDA:
> >
> > http://www.zope.org/Products/DA/ZODBCDA
> >
> > Sarah Burke wrote:
> >
> > > Hello...
> > > I want to connnect up to a NT Access 97 Database through Zope running on
> > > Unix. Is the mxODBC module in an External Method the way to go?
> > > Thanks in advance
> > >
>
> if i remember correctly ZODBCDA only works on Win32 platform.
>
> also Access 97 is not a client/server database, so you cant connect to
> it from unix directly to it.
>
> perhaps the ODBC Socket Server from http://odbc.linuxbox.com/index.html
> will work (haven't tried it).
>
> You have to write a python client for the ODBC Socket Server (if noone
> else has written it, don't know).
>
> Ciao,
> Jochen
> --
> --------------------------------------------------
> Jochen Knuth          WebMaster http://www.ipro.de
> IPRO GmbH             Phone ++49-7152-93330
> Steinbeisstr. 6       Fax ++49-7152-933340
> 71229 Leonberg        EMail: J.Knuth at ipro.de



From news at dorb.com  Thu Jan  6 19:08:24 2000
From: news at dorb.com (Darrell)
Date: Thu, 6 Jan 2000 19:08:24 -0500
Subject: newbie embedding questions
References: <kX6d4.41059$Ke.26557@news21.bellglobal.com> <Fnxpnq.JrD@world.std.com> <cs9d4.42068$Ke.26436@news21.bellglobal.com>
Message-ID: <r5ad4.405$mj7.9263@typhoon.nyroc.rr.com>

I've used SCXX http://starship.python.net/crew/gmcm/scxx.html and it worked
well.

Gordon has some examples and so do I if you want.
http://www.dorb.com/darrell/randomNumber/

--
--Darrell




From jrush at summit-research.com  Mon Jan 17 13:15:30 2000
From: jrush at summit-research.com (Jeff Rush)
Date: 17 Jan 2000 12:15:30 -0600
Subject: If you're still looking for a room for the Python Conference...
Message-ID: <VrDQRddcbbs9-pn2-TpsNnUI9gtvg@Fortress>

I've got one and am looking to share with someone, to reduce the costs.  If
you're
interested, drop me a line and/or come participate in the web-based 
bulletin
board for other travel and hotel accomodations at:

http://starship.python.net/crew/scharf/GroupMind/GroupMind.cgi?req=show&fil
e=ConferenceStuff.gromi

Looks like a great conference!

-Jeff Rush



From tjg at avalongroup.net  Fri Jan 21 12:19:53 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Fri, 21 Jan 2000 09:19:53 -0800
Subject: TKinter keybindings
Message-ID: <38889539.92C49C8C@exceptionalminds.com>

Hi all,

I've got a TK key binding <ISO_Left_Tab> that works great on my Linux
box, but doesn't work at all on a windows box. Is there a windows
equivalent?


-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From fredrik at pythonware.com  Wed Jan 19 11:57:45 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 17:57:45 +0100
Subject: Looking for the FAQ
References: <slrn88bqi2.512.sprawlsr@TacHQ.com>
Message-ID: <009401bf629e$4f9c4760$f29b12c2@secret.pythonware.com>

Robert J. Sprawls <sprawlsr at worldnet.att.net> wrote:
> I'm new to this group. I run Linux and it has Python 1.5. I'm
> an ANSI C programmer, learning C++ at my University, and I'm
> interested in checking python out. Is there a good book that takes a
> person from beginning to at least intermediate level? This is probably
> something covered in the FAQ, so if you just point me to the FAQ, I'll
> take it from there.

http://www.python.org

(also see the "Everything you want" section in the article titled
"Dr. Dobb's Python-URL! - weekly Python news and links" which
was posted earlier today)

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! &smiley; -->




From ullrich at math.okstate.edu  Mon Jan 31 11:42:02 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Mon, 31 Jan 2000 10:42:02 -0600
Subject: Dynamic class construction?
References: <389309F7.4A35371F@math.okstate.edu> <8707c0$ua2$1@nnrp1.deja.com> <38948F0E.36AA2548@math.okstate.edu> <872jt9$gq4$1@nnrp1.deja.com>
Message-ID: <3895BB5A.D94C4823@math.okstate.edu>

Nic Williams wrote:

> [...] Just use lots of comments :-)

    Ok. Yesterday's comment was "thanks, that's more
or less what I thought. I thought I tried setting __name__
and it didn't work, wonder what I did different when I
tried it?"

    Last night's comment was "darn, it still doesn't work
when I try it at home, I wonder what I'm doing different?
I guess I shoulda took the example home with me instead
of just imagining I could remember what one line of code
looked like..."

    This morning's comment is "what the heck, the code
that doesn't work at home works just fine here at the
office. Huh?"

    Thanks again - I'm pretty sure now that things _do_
work more or less the way I thought, all I gotta do
is figure out what's funny about the installation on
that machine at home. Which is definite progress.
Aargh.

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



From gmcm at hypernet.com  Mon Jan 10 16:47:03 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 10 Jan 2000 16:47:03 -0500
Subject: Problem distributing a Gadfly app
In-Reply-To: <85dah2$ip8$1@nnrp1.deja.com>
Message-ID: <1264581670-27467972@hypernet.com>

calishar wrote:
> 
>   I am just in the finishing off stage of an application for work
>   that
> uses the gadfly database package. I am trying to build an
> installable application using installer (from Gordon Macmillan?)
> and running into a problem.
> 
>   The application builds alright, but when I install it on the
>   other
> machine, and go to run it, it fails.
> 
>   After re-building with the debug flag on, use runw off, and
>   support
> on, I saw that it is trying to load the sql.mar file from the c:
> drive. Well, it ain't there. If I try to add the sql.mar file to
> the dependencies line of the cfg file, builder fails, if I add it
> to the includes line it fails.

Try the misc= option of the INSTALL section.

Both options you tried are for the PYZ, which consists only of 
pure Python. (The dependencies option tracks dependencies, 
the include option doesn't).

As I recall, the sql.mar file has to end up in the right place for 
gadfly to find it. You may have to struggle some with gadfly to 
figure out where it expects to find it. Once you've done that, 
you'll need to put it there at installation time. So your install 
script should say something like (assuming you use 
gen_install as a model):

grammarfile = this.extract('sql.mar')[1]
f = open(os.path.join(idir, gadlfypath, 'sql.mar'), 'wb')
f.write(grammarfile)

where "this" and "idir" are things gen_install figured out, and 
"gadflypath" is what you figured out.

- Gordon



From erw at dde.dk  Mon Jan 24 11:29:47 2000
From: erw at dde.dk (Erwin S. Andreasen)
Date: 24 Jan 2000 16:29:47 GMT
Subject: Embedding -- getting error/traceback information not to stderr?
Message-ID: <slrn88oujk.ebi.erw@erw-pc.dde.dk>

In my application using embedded Python, I'd like to print out the error
information as it occurrs (traceback, etc.). This is what PyErr_PrintEx()
(called from PyErr_Print()) does -- unfortunately, it does it to sys.stderr.

Did I miss some other similar function that will return the nicely formatted
exception information in a variable?

Otherwise it looks like my options are: 
 1) copy PyErr_Print(), modify it to append to a file (yuck!)
 2) Modify sys.stderr to append to an internal buffer instead of a file

2) seems to be a better option (I'm not certain it's possible).


PS: Where does the filename in the stack trace messages come from? I can't
quite find it in the code: I create my modules with PyImport_AddModule and
setup a restricted environment then parse the code from a file. I've
__file__, but maybe I do it too late.


-- 
==============================================================================
Erwin Andreasen   Herlev, Denmark <erw at dde.dk>          UNIX System Programmer
<URL:http://www.andreasen.org>              <*>              Goodbye, Richard!
==============================================================================


From joakim at login1.styx.net  Thu Jan 20 18:52:10 2000
From: joakim at login1.styx.net (Joakim Ziegler)
Date: 20 Jan 2000 23:52:10 GMT
Subject: Attributes acting as arrays (in C extensions)?
Message-ID: <slrn88f7le.ke5.joakim@login1.styx.net>

I'm implementing a Python module in C. I've already done object attributes
which invoke C functions to get and set the values, but is it possible to
have object attributes that act as arrays? So that you can do 

object.attribute[n]

in Python, and call a C function with object, the name of the attribute, and
n as arguments? I've not been able to find docs on how to do this.

-- 
Joakim Ziegler - styx research director - joakim at styx.net
FIX sysop - FIXmud admin - FIDEL & Conglomerate developer


From sposhua at my.pc  Fri Jan  7 04:43:49 2000
From: sposhua at my.pc (Sposhua)
Date: Fri, 7 Jan 2000 09:43:49 +0000
Subject: mod python?
Message-ID: <Pine.WNT.4.21.0001070938001.197-100000@wks_3.e-shopwork.com>

I've just heard about an "exciting" thing called mod Perl
( http://perl.apache.org ) - Apache/Perl integration. From what I can work
out, basically it's faster as it doesn't have to start the interpreter a
billion times, and it makes it possible to write Apache modules directly in
Perl. Is is

a) Really exciting?
b) Is there an equivalent for Python?
c) What the heck is an Apache module anyway? (in practical terms)



From gmcm at hypernet.com  Mon Jan 17 08:42:42 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 17 Jan 2000 08:42:42 -0500
Subject: Stopping a thread from the calling process
In-Reply-To: <EF5CE20F7E@nds-3.azg.nl>
References: <1264075595-4560668@hypernet.com>
Message-ID: <1264005932-8750908@hypernet.com>

P.J.W.S. Vrijlandt wrote:

> Gordon wrote:
> 
> > The normal idiom is to ask politely - have the thread 
> > periodically check some variable to see if it should stop.
> 
> How would you do this if the thread-to-be-stopped is waiting for 
> (keyboard-) input?
> 
> e.g. in telnetlib we have:
> 
>     def mt_interact(self):
>         """Multithreaded version of interact()."""
>         import thread
>         thread.start_new_thread(self.listener, ())
>         while 1:
>             line = sys.stdin.readline()
>             if not line:
>                 break
>             self.write(line)
> 
>     def listener(self):
>         """Helper for mt_interact() -- this executes in the other  
>         thread."""         
>         while 1:                        
>             try:
>                 data = self.read_eager()
>             except EOFError:
>                 print '*** Connection closed by remote host ***'
>                 return
>             if data:
>                 sys.stdout.write(data)
>             else:
>                 sys.stdout.flush()
> 
> mt_interact should break out of its loop when EOFError occurs within 
> listener.
> 
> (also note that select on input won't work on win32, and looping on 
> keypressed seems rather processor-unfriendly)

Well, select (if it worked on files on WIn32) would be optimal.

Using busy-wait is somewhat offensive, but
 if not msvcrt.kbhit():
   time.sleep(0.1)
is not going to consume much CPU.

If EOF on the socket is the end of the program, I would 
reverse the purpose of the threads, (have the worker thread 
read stdin).
 
> BTW: Would this be better rewritten (for this reason or another) 
> using threading? 

Glancing at it (I've never used it), it looks like there should 
probably be a condition protecting use of the socket; so 
probably yes.



- Gordon



From daniel at spamfree.clacknet.com  Fri Jan  7 10:25:35 2000
From: daniel at spamfree.clacknet.com (Daniel)
Date: Fri, 07 Jan 2000 15:25:35 +0000
Subject: Python friendly emacs
Message-ID: <3876056F.36E4335D@spamfree.clacknet.com>

I'm running emacs under X and I'm
confused about adding color syntax support
for python.

I know that python comes with a .el file
for emacs but I am unable to make emacs
understand it.

Also, I'd like emacs to default to python mode
when it edits a file with the extension of .cgi

(Yes, this is the python newsgroup but I can't
imagine I'm the only python programer who
uses emacs)

-Daniel

[remove the 'spamfree.' form my address]



From volumeregeling at hotmail.com  Tue Jan 25 10:36:51 2000
From: volumeregeling at hotmail.com (volumeregeling)
Date: Tue, 25 Jan 2000 15:36:51 GMT
Subject: Problems with menuitems
Message-ID: <388dc24e.3978861@122.0.0.250>

Hi,

Can you tell me how to do some configuration things on menuentries
without them being done right away.

I am DISABLING for instance the Save entry.

I an ENABLING the Save entry in the SaveAs option. But when my program
starts it somehow start the saveas function so it enables the save
option at startup which should be impossible.

If you can tell me how to do this please do.

Thanks. Volumeregeling


From david.maslen at iname.com  Tue Jan 25 21:00:07 2000
From: david.maslen at iname.com (David Maslen)
Date: Wed, 26 Jan 2000 02:00:07 GMT
Subject: Gnuplot on two axis, with python-gnuplot
Message-ID: <871z75wqkb.fsf@iname.com>

I'm trying to write a script to pass an csv file of stock prices and
plot the results on a gnuplot graph, rather like yahoo finance does.

The only problem with the attached script is it tries to use the first
X axis for both price AND volume. Since the scale is vastly different
this doesn't work. So I need to put the first dataset on the left y
axis and the second on the right axis. The command in gnuplot is
axes=x1y1 and axes=x1y2 but I can't figure out how to pass that
information to the Gnuplot.Data instance?

If anyone can get me back on track I'd appreciate it.

thanks

-------------- next part --------------
A non-text attachment was scrubbed...
Name: stock-gnuplot.py
Type: application/octet-stream
Size: 2406 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000126/1ef7ab76/attachment.obj>

From Niklas.Blomberg at hassle.se.astra.com  Tue Jan 18 07:17:36 2000
From: Niklas.Blomberg at hassle.se.astra.com (Niklas.Blomberg at hassle.se.astra.com)
Date: Tue, 18 Jan 2000 13:17:36 +0100
Subject: Interacting with external program -pty?
Message-ID: <3DB0284E2B7AD21189A400805F19DB280129E238@se-hsl-mail010.hassle.se.astra.com>

Dear all,

I would appreciate some advice on how to script an external interactive
process(program) using python. In addition to sending commands (which is
straightforward) I need to be able to capture the output (which can be
extensive) for further downstream processing. Since the program expects a
terminal pipes don't work - the output is buffered - I have started to look
at the pty module. This is uncharted territory for me and any advice both on
possible solutions and perhaps availiable example code would be most
appreciated.

I'll be happy to post a summary.

Thanks,
Niklas

-----
Niklas Blomberg
AstraZeneca Structural Chemistry Laboratory
niklas.blomberg at astrazeneca.com





From aahz at netcom.com  Mon Jan 10 12:17:03 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 10 Jan 2000 17:17:03 GMT
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net> <3876B79F.30439283@mdlink.de> <001701bf5b54$e0e110d0$f29b12c2@secret.pythonware.com>
Message-ID: <85d46f$nob$1@nntp3.atl.mindspring.net>

In article <001701bf5b54$e0e110d0$f29b12c2 at secret.pythonware.com>,
Fredrik Lundh <fredrik at pythonware.com> wrote:
>Helge Hess <helge.hess at mdlink.de> wrote:
>>
>> Hm, I wonder why it is simpler to have two separate kinds of methods
>> instead of a consistent behaviour for all methods ?
>
>umm. in my dictionary, "consistent" means "free from variation or
>contradiction".
>
>in the current design, if you call a class object, you *always* get
>an instance of that class. in the current design, if you inherit from
>a class and call the parent's init function, it *always* initializes
>the current instance. in the current design, multiple inheritance is
>*always* supported.
>
>in your design, you've changed "always" to "it depends". hardly
>something that would make Python's behaviour more consistent.

Helge, Fredrik here has done a much better job of illustrating what I
meant by top-view versus bottom-view.  You're focusing on consistency of
method implementation and use (bottom-view); we're focusing on
consistency of the Python model as a whole (top-view).
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From piet at cs.uu.nl  Fri Jan 14 05:47:04 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 14 Jan 2000 11:47:04 +0100
Subject: file width and height
References: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com> <ubqf4.3018$hN5.92483@typhoon.nyroc.rr.com> <004b01bf5e68$92965a40$f29b12c2@secret.pythonware.com>
Message-ID: <uln5tlyl3.fsf@cs.uu.nl>

>>>>> "Fredrik Lundh" <fredrik at pythonware.com> (FL) writes:

FL> Darrell <news at dorb.com> wrote:
>> Sposhua wrote:
>> >Is there a module to find gif/jpeg width and height?
>> 
>> I hacked this out of PIL and it should work for gif and jpg.

FL> """
FL> hacked for PIL by Darrell Gallion
FL> """

FL> where "hacked" means "copied 150 lines of source
FL> code, and removed the copyright notice"...

Isn't that the original meaning of hacking (as in `with an axe`)?
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From aa8vb at yahoo.com  Mon Jan 24 11:23:20 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 24 Jan 2000 11:23:20 -0500
Subject: anydbm - a simple question
In-Reply-To: <3diu0lbz8e.fsf@amarok.cnri.reston.va.us>
References: <20000121110715.A184793@vislab.epa.gov> <3diu0lbz8e.fsf@amarok.cnri.reston.va.us>
Message-ID: <20000124112320.A440571@vislab.epa.gov>

Andrew M. Kuchling:
 |Randall Hopper <aa8vb at yahoo.com> writes:
 |>  File "/home/rhh/software/python-1.5.2/lib/python1.5/dumbdbm.py", line 87,\
 |>    in _addval
 |>    f.write('\0'*(npos-pos))
 |>TypeError: can't multiply sequence with non-int
 |
 |I'll bet you're on Solaris and are picking up the large file support,
 |which makes the .tell() method of file objects return a long integer.
 |But you can't multiply a sequence by a long integer, only a regular
 |integer:
...
 |Short-term, you can patch dumbdbm.py to use int(npos-pos), but this is
 |really an interpreter bug; there's no reason 2L * 'a' should be an
 |error.

Ok, patching around this bug as you suggested, Fredrik's anydbm example
still fails.

   >>> import anydbm
   >>> db = anydbm.open( "database", "c" )
   >>> db[ "1" ] = "one"
   >>> db.close()
   >>> db = anydbm.open( "database", "r" )
   Traceback (innermost last):
     File "<stdin>", line 1, in ?
     File "/home/rhh/software/python-1.5.2/lib/python1.5/anydbm.py", line 80, in open
       raise error, "need 'c' or 'n' flag to open new db"
   anydbm.error: need 'c' or 'n' flag to open new db


It created the database.dir and database.dat files in the write phase,
but the read phase fails on the open.

Thinking it might be extension related I tried changing the file-to-open
from "database" to "database.dat".  But no dice:

   anydbm.error: db type could not be determined

Same thing if I change the path to "database.dir".

Is there another bug at work here?  Or is the example flawed?

-- 
Randall Hopper
aa8vb at yahoo.com



From mlh at vier.idi.ntnu.no  Wed Jan 12 18:32:27 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 13 Jan 2000 00:32:27 +0100
Subject: Text File
References: <cc0q7s8ggiqj62j27mcrtgg5a374qbm94h@4ax.com>
Message-ID: <iacaemalvck.fsf@vier.idi.ntnu.no>

Simon Faulkner <news at titanic.co.uk> writes:

> Hi All,
> 
> I would like to write a CGI program that diskplays a small (<1k) text
> file on a web page and saves any changes made by the browser back to
> the same file.
> 
> Any sugestions to help me get started?
> 
> Specifically
> 
> Which data type should I use and how do I read the text  file into
> it?

Just do:

  file = open("filename.txt")
  text = file.read()
  file.close()

And there it is - as a string. Put that into a text-area in your HTML
output for the user to edit. When you get the result, you'll probably
get it in a manner similar to this:

  data = FieldStorage()
  text = data["text"].value()

  file = open("filename.txt","w")
  file.write(text)
  file.close()

And there you are. Now all you need to do is build the CGI stuff and
HTML around it :)

> 
> Cheers  
> 
> 
> Simon
> 

--

  Magnus
  Lie
  Hetland                                mailto:<anything>@hetland.org


From mal at lemburg.com  Wed Jan 12 13:20:21 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 12 Jan 2000 19:20:21 +0100
Subject: mxODBC
References: <rk17OF5=o=S0QxS2fUYwTLmxxrno@4ax.com>
Message-ID: <387CC5E4.27BB8FC2@lemburg.com>

Anders M Eriksson wrote:
> 
> hELLO!
> 
> I'm trying to use the mxODBC module to access an dBASE 5 database on a
> Windows NT system. I have used this script:
> 
> import ODBC.Windows
> 
> db = ODBC.Windows.connect('myDatabase')         # MyDatabase is a
> local DSN
> <snip>
> db.close()
> 
> When I run this script I get this error on the connect line
> 
> NotSupportedError: ('S1C00', 106, '[Microsoft][ODBC dBase
> Driver]Optional feature not implemented ', 4226)
> 
> If I use the Win32 odbc then it works!
> 
> any idea what is wrong??

As already mentioned in private mail: you have to pass
0 for the optional clear_autocommit parameter of the
connect APIs. File DSNs do not support transactions and thus
always operate in auto commit mode.

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





From paul.robinson at quantisci.co.uk  Tue Jan  4 12:58:33 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Tue, 04 Jan 2000 17:58:33 +0000
Subject: exception arg
References: <slrn8749le.d0.mn@banger.iol.it>
Message-ID: <387234C9.299AC84A@quantisci.co.uk>

Mirko Nasato wrote:
> 
> I can't understand the following:
> 
> >>> 1/0
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ZeroDivisionError: integer division or modulo
> >>> try:
> ...     1/0
> ... except ZeroDivisionError, arg:
> ...     if arg == "integer division or modulo":
> ...             print "This is what I expect."
> ...     else:
> ...             print 'Unexpected. "arg" is "%s".' % arg
> ...
> Unexpected. "arg" is "integer division or modulo".
> >>>
> 
> The same seems to occur with every kind of exception, so maybe
> there's an explanation...

There is: arg is not a string....

For example:
PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> try:
... 	1/0
... except ZeroDivisionError, arg:
... 	print type(arg)
...     print `arg`
...     print arg
...     print str(arg) == 'integer division or modulo'
... 	
<type 'instance'>
<exceptions.ZeroDivisionError instance at 1137370>
integer division or modulo
1
>>> 

arg is an instance of an exception object.

The string that you were comparing it with is the string representation
of that object.

Hope that helps,
	Paul.



From thomas at xs4all.net  Sun Jan  2 19:52:19 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 3 Jan 2000 01:52:19 +0100
Subject: New To Python
In-Reply-To: <1cRb4.10160$zp.149897@news3.mia>; from gnpotts@email.uophx.edu on Sun, Jan 02, 2000 at 06:31:17PM -0500
References: <1cRb4.10160$zp.149897@news3.mia>
Message-ID: <20000103015218.A5669@xs4all.nl>

On Sun, Jan 02, 2000 at 06:31:17PM -0500, Gary N. Potts wrote:

> I'm brand new to Python and really to programming.  I've dabbled a bit by
> programming muds and the like, but nothing real.  I've downloaded Python
> 1.5.2 and all the documentation to go with it.  I've found several help
> sites that are able to teach me the language, however, I have not been able
> to find a decent instructional guide on how to actually use the program
> itself.  Is there any good tutorials for that purpose out there and if so,
> were can they be found?

The Python Tutorial at http://www.python.org/doc/tut/ starts off with some
instructions on how to use the program interactively. (Chapter 2. Using the
Python Interpreter) It goes on about how to use it in scripts, the #!
notation, etc. The Tutorial is an excellent starting place -- it doesn't go
too deep into technicalities, but there's plenty of other reading material
once you've finished it. (See http://www.python.org/doc/)

>From the way you phrase the question, however, I think you may be asking the
wrong thing :) However great fun the python interpreter may be,
interactively, it isn't what python is all about. The interpreter is only
the tool that executes the scripts. You write a script, and you tell a
python interpreter to execute it. So maybe you aren't looking for help on
the python program itself, but are actually looking for an Integrated
DeveLopment Enviroment for python.

Well, if so, you're in luck. Python has (at least) one of those, and it's
named IDLE. It should run on any OS with Tkinter support, which includes
Windows, Macintosh and a lot, if not all, flavors of UNIX. For Windows,
there is also the PythonWin IDE, which comes with Mark Hammond's win32all
PythonWin package (See the Win95 section of the download page on
http://www.python.org/download/ for more details.)

IDLE should be included with the Python 1.5.2 you downloaded. If you're
using some flavor of UNIX and python got installed by your vendor, you might
be missing IDLE and/or Tkinter... You'll either need to download the source
(and compile, if you're missing Tkinter) or, if you're using Linux, check
out the Linux Binaries section of the download page.

IDLE-fingers-ly y'rs

-- 
Thomas Wouters <thomas at xs4all.net>

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



From fredrik at pythonware.com  Sun Jan  2 12:17:58 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 2 Jan 2000 18:17:58 +0100
Subject: Need help with canvas widget in Tkinter please
References: <386EC3E6.FAC5BC0@telebot.com>
Message-ID: <005d01bf5545$51d45ec0$f29b12c2@secret.pythonware.com>

flounder <flounder_pounder at telebot.com> wrote:
> when i run this it gives me this error
> 
> -----------ERROR MESSAGE--------------
> 
> Traceback (innermost last):
>   File ".//test.py", line 11, in ?
>     item = display.create_image(0, 0, anchor=NW, image=photo)
> AttributeError: 'None' object has no attribute 'create_image'
> 
> -----------END OF ERROR---------------
> 
> can anyone show me how this should be done I have looked at the
> documentation and can't figure it out

what documentation?

your problem, and how to fix it, is described here:
http://www.pythonware.com/library/tkinter/introduction/x131-more-on-widget-references.htm

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
(currently number five on fatbrain's bestseller list! ;-) -->




From calishar at *NOSPAM*home.com  Mon Jan 17 06:46:04 2000
From: calishar at *NOSPAM*home.com (Calishar)
Date: Mon, 17 Jan 2000 11:46:04 GMT
Subject: problem with makepy output
References: <4vg48skfrigntmuslpfm40ngea5pdl4r2m@4ax.com> <8itg4.308$iL4.4699@news.rdc1.tx.home.com>
Message-ID: <86068sso7f766ihioe39qj3lcj9041hcdv@4ax.com>

On Mon, 17 Jan 2000 00:23:32 GMT, "Paul E. Bible" <pbible at home.com>
wrote:

>It looks like makepy is grabbing an older version of Excel for some reason.
>Try using makepy from the Excel9.olb (this is Excel's Type Library) from the
>..\Microsoft\Office\ subdirectory.  Then copy the generated Python file to
>your own  import file (eg. Excel9.py) somewhere on your PYTHONPATH.  Once
>you've done that, just import Excel9 and that's it.  I'm running Windows/NT
>(SP4) with Office 2000 and this works fine.
>


  Thanks Paul, I must have been running it on the wrong file, just ran
it on excel9.olb, and it works fine. Thanks

Calishar


From gerrit.holl at pobox.com  Sun Jan 16 11:40:04 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 16 Jan 2000 17:40:04 +0100
Subject: SUGGEST: os.getuser()
Message-ID: <20000116174004.A5995@stopcontact.palga.uucp>

Hello,

may I suggest this function for in the posix module:

def getuser():
    import pwd
    return pwd.getpwuid(getuid())[0]

I don't know how this would be implemented on W/NT.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From michael.stroeder at inka.de  Tue Jan 25 18:04:19 2000
From: michael.stroeder at inka.de (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 26 Jan 2000 00:04:19 +0100
Subject: anydbm - a simple question
References: <20000121110715.A184793@vislab.epa.gov>
  <3diu0lbz8e.fsf@amarok.cnri.reston.va.us>
  <20000124112320.A440571@vislab.epa.gov>
  <388CE28F.D3A100AB@inka.de> <14477.47754.65118.177514@beluga.mojam.com>
Message-ID: <388E2BF3.8CCF5849@inka.de>

Skip Montanaro wrote:
> 
> Michael> The whichdb module fails on determining the DB type.
> 
>     http://www.musi-cal.com/~skip/Python/
> 
> and browse into the Lib directory and grab whichdb.py.

I did not find a "Lib" directory or similar under

  http://www.musi-cal.com/~skip/python/

Can you post the exact URL? Thank you.

Ciao, Michael.


From sameerc1 at yahoo.com  Fri Jan  7 19:33:40 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Fri, 7 Jan 2000 16:33:40 -0800
Subject: python with ASP
Message-ID: <020e01bf5970$041d46e0$556afea9@desktop>

>When you copy-and-pasted my code, it appears you lost the tabbing.
>I've attached the code in case your newsreader isn't displaying the
>tabs happily.
>    -Eugene

Yes, it seems to work now.  But restricting to one line of code, or a
multiline string is kind of limitting.  This kind of limitation doesn't
exist with JScript, or VBScript.  I hope this will be fixed in the future.

__________________________________________
NetZero - Defenders of the Free World
Get your FREE Internet Access and Email at
http://www.netzero.net/download/index.html



From phil at ricochet.net  Tue Jan 18 02:17:52 2000
From: phil at ricochet.net (phil at ricochet.net)
Date: Tue, 18 Jan 2000 07:17:52 GMT
Subject: If you can only buy *one* Python book...
References: <N%rg4.4360$3b6.20735@ozemail.com.au> <20000116182502.A1106@better.net> <t6tg4.305$iL4.4241@news.rdc1.tx.home.com> <20000117204759.B1291@stopcontact.palga.uucp>
Message-ID: <38841039.6530119@news.ricochet.net>

On Mon, 17 Jan 2000 20:47:59 +0100, Gerrit Holl
<gerrit.holl at pobox.com> wrote:

>Paul E. Bible wrote on 948064265:
>> I found 'Learning Python' to be very easy to read and a great book to get
>> started programming in Python...
>
>Agreed.
>
>> The 'Python Essential Reference' is a good addition to the online
>> documentation as well.
>
>Disagreed. I think the online reference is far the best and most
>complete there is. You don't need any other book than 'Learning Python';
>you aren't going to *read* the library reference, are you?
>Maybe the first few chapters on builtin-types and common modules like os,
>string (obsolete in 1.6), sys, and knowing about glob, rexec and
>MimeWriter can also be useful, but *reading* it as a whole... no.

I don't know about 'Python Essential Reference', but i really do agree
that  the online documentation is excellent. Since i use "Widows", the
pythlp.chm works as a very usable reference. 

>> I found 'Programming Python' somewhat difficult to follow myself.
>Agreed.

It seemed a strange book as i thought it started off well for the
first couple of chapters  and then seemed to lose it and wander off
into some GUI hell. Thought it might just be me since i'm really after
Python because of its scripting capabilities.

phil.

The world is divided into two sorts of people:
Those that believe the world is divided into
two sorts of people and those that don't.


From md at i-i-s.co.uk  Mon Jan 24 04:44:09 2000
From: md at i-i-s.co.uk (Matt Dunn)
Date: Mon, 24 Jan 2000 09:44:09 -0000
Subject: python and www
In-Reply-To: <003501bf65c2$d2fe3880$918ffea9@desktop>
Message-ID: <0933F9C2D055D111BE93400001791A050FFE42@iisweb1.i-i-s.co.uk>

Yep, check out http://starship.python.net/crew/aaron_watters/pws.html for
how to use Python with IIS,
or http://starship.python.net/crew/lemburg/mxCGIPython.html for a cgiPython
binary you can use.

Matt Dunn

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of Sameer Chowdhury
Sent: 23 January 2000 16:57
To: python-list at python.org
Subject: python and www


I want to invoke the python interpreter, pass it my script, and direct the
script's output as a response to a request made by a client browser.
Without using Python with ASP, does anyone know how I would go about doing
that (say using IIS or Apache).  Thanks in advance.






From insyte**NOSPAM** at emt-p.org  Tue Jan 25 17:22:14 2000
From: insyte**NOSPAM** at emt-p.org (Ben Beuchler)
Date: Tue, 25 Jan 2000 16:22:14 -0600
Subject: vi and python?
References: <85uc1m$5i8$1@pegasus.tiscalinet.it>
Message-ID: <86l7iq024mg@news2.newsguy.com>

There seems to be a general preference for emacs in the python community.  I am personally much more comfortable working with vim.  Does anyone have any particular tips for python development within vim? Particular shortcuts to test code, autoindentation settings, etc...  I would particularly like to be able to use some of the idle features with, perhaps, gvim...

Ben

-- 
"There is no spoon"
	-- The Matrix



From gmcm at hypernet.com  Wed Jan 19 09:57:37 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 19 Jan 2000 09:57:37 -0500
Subject: no multithreading?
In-Reply-To: <864ih0$5is$1@nnrp1.deja.com>
Message-ID: <1263828637-4444924@hypernet.com>

moonseeker asks:
> 
> I have understood that the python interpreter switches to another thread if a
> fixed amount of python commands is completed. Is this true? 

Yes.

> That means if a
> task hangs on a command (i.e. file open), all tasks are locked? 

No. All core Python (and well-behaved C extensions) release 
the lock when about to attempt a blocking operation; then 
reaquire it when done.

> What happens
> if I use sleep() in a task? 

You release the lock (for the approximate duration of the 
sleep).



- Gordon



From arcege at shore.net  Sat Jan  8 12:25:41 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 08 Jan 2000 17:25:41 GMT
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
References: <20000105082703.A9441@vislab.epa.gov> <1265040530-3892638@hypernet.com> <ONAe+FAsO4c4Ew5h@jessikat.demon.co.uk>
Message-ID: <pqKd4.2780$Ef6.704898@news.shore.net>

Robin Becker <robin at jessikat.demon.co.uk> wrote:
: In article <1265040530-3892638 at hypernet.com>, Gordon McMillan
: <gmcm at hypernet.com> writes
:>Robin Becker wrote:
:> > > |Also under
:>> certain extreme circumstances my code would like to report a >
:>> |total failure eg attempt to divide by error. I would like to ask
:>> more > |experienced pythoneers how arithmetic errors are handled
:>> using built in > |IEEE exceptions and how does one gracefully
:>> crash a python script? (eg > |an abend/abort). 
:>
:>> >raise
:>> ArithmeticError, "Oops" >raise ZeroDivisionError, "Ouch" > >? >
:>
:>> yes that's python, but what's the best way to do it in an
:>> existing C library? -- Robin Becker
:>
:>PyErr_SetString(PyExc_ZeroDivisionError, "Ouch");
:>return NULL;
:>
:>(see bltinmodule.c for the complete list).
:> 
:>> -- 
:>> http://www.python.org/mailman/listinfo/python-list
:>
:>
:>
:>- Gordon
:>

: I guess I could have made my query a bit clearer. When I call my
: extension in C I can pass inputs which may cause unexpected arithmetic
: exceptions. That's an efficiency saving. I could check every arithmetic
: op. Assuming I want speed is there a standard pythonic way eg
: signal+setjmp/longjmp to trap these errors?

You will want to check that Python C/API functions return NULL and
PyErr_Occurred() returns non-NULL (the "pythonic way").

  /* add two numbers and check any exception */
  if ((c = PyNumber_Add(a, b)) == NULL &&
      PyErr_Occurred() != NULL) {  /* exception raised */
    if (PyErr_ExceptionMatches(PyExc_ZeroDivisionError)) {
      PyErr_SetString(PyExc_SystemError,
                      "how can we get a division by zero on an add");
      return NULL;
    } else if (!PyErr_ExceptionMatches(PyExc_ArithmeticError))
      return NULL;  /* propogate the exception */
    else /* arithmetic error, try something else? */
      PyErr_Clear();
    }
  }

<URL: http://www.python.org/doc/current/api/exceptionHandling.html>

:  In the VBA interface I have implemented exception trapping using VC++
: __try blocks (I seem to remember this was because of Excel somehow).

: The current C uses a pair of global functions to handle I/O to
: stdout/err. What's a good way to provide pythonic output for a C
: extension module?

Which pair of global functions are you talking about?

If you mean output to sys.stdout and sys.stderr, then you can use
PySys_WriteStdout() and PySys_WriteStderr().

<URL: file:///sources/Python-1.5.2/Python/sysmodule.c>

If you are talking about real file descriptors (the term you've been
using in this thread), then you can create a Python file object wrapped
around it.

  PyObject *opened_file;
  { FILE *fp;
    fp = fdopen(opened_fd, "w+");
    /* the NULL tells PyFile_FromFile not to close implicitly */
    opened_file = PyFile_FromFile(fp, "<internal>", "w+", NULL);
  }
  PyFile_WriteString("This is to some file\n", opened_file);
  Py_DECREF(opened_file); /* leaves file open, but reclaims obj */

<URL: http://www.python.org/doc/current/api/FileObject.html>

: One other thing I am a bit uncertain about is threading. If I call
: python inside my C extension must I worry about threading issues for the
: extension?

There are some things you need to worry about, yes - mostly about
access to global variables in the module itself.  As long as you do
that within the interpreter lock, things should be fine.



From emile at fenx.com  Fri Jan 14 07:32:30 2000
From: emile at fenx.com (Emile van Sebille)
Date: Fri, 14 Jan 2000 04:32:30 -0800
Subject: Threads of ASP with Python script code
Message-ID: <03fe01bf5e8b$72be4340$01ffffc0@worldnet.att.net>

It's hard to tell if this will help, but I had a similar situation
where a table of totals doubled each time it was viewed.  When I
modified the table initialization routine to be _very_ explicit,
the problem went away.

HTH

Emile van Sebille
emile at fenx.com
-------------------


Stefan Radke <Stefan_Radke at compuserve.com> wrote in message
news:<387E8361.29684CF9 at compuserve.com>...
> The following ASP gives me trouble:
>
> [HTML]
> [some initialisation script code]
> [more HTML]
> <%
> import my_python_module
> %>
>
> [more HTML, mixed with script code, creating an instance of class in
> my_python_module:]
> this = my_python_module.myClass()
>
> The my_python_module has a class defnition. From this class I create
an
> instance, an read some file names from a disk into a list, the only
member
> of this class. The list is use dto create an options-list on the page.
>
> The problem now is, that whenever I do a reload of the ASP in IE, the
list
> gets longer, because the same files are searched in the
class-initialisation
> again, but the previous instance still exists.
>
> This even does not change, if I add a "del this" at the end. On
another
> machine the same list can be viewed, although a new session has been
> started. To me this means that both ASP sessions use the same memory
space
> within the python environment :-( ...
>
> Does anyone know how to implement this correctly ?  It would mean to
have
> seperate memory in each session.
>
> Thank you for any help, Stefan.
> --
> http://www.python.org/mailman/listinfo/python-list
>






From dsnider at mindspring.com  Fri Jan 28 14:17:40 2000
From: dsnider at mindspring.com (Dan Snider)
Date: Fri, 28 Jan 2000 14:17:40 -0500
Subject: ObjectDomain status?
References: <86pqmb$d22$1@nnrp1.deja.com>
Message-ID: <86spoc$2dj$1@news.inc.net>

Hi Harri,

> I wonder if the
> company has gone belly-up?

Absoultely, 100% not!! We are doing quite well but extremely busy preparing
a public beta of a new release. This should be posted real soon... You know
the deal about dates :-) Sorry for the delay in some of the pre-sale support
emails. We are trying to hire more people to accomodate the demand but the
new release has us a little behind.

Regards,

Dan


<hpasanen at my-deja.com> wrote in message news:86pqmb$d22$1 at nnrp1.deja.com...
> Hi,
>
> This brushes on Python, so I though someone here might know...
>
> I've recently evaluated ObjectDomain UML modeller for our internal use.
> (http://www.objectdomain.com/)   The unique feature of their product is
> that it contains an embedded JPython as a scripting language, while
> the rest of it resembles Rational Rose a whole lot from the UI
> perspective.  (perhaps too much?)
>
> All in all, ObjectDomain 2.5 seems to work reasonably well, but I can't
> get any response from ObjectDomain to my queries.  I wonder if the
> company has gone belly-up?
>
> Anyone in the know?  Would be a shame to see a prominent Python product
> going down the drain, but perhaps they'll Open-Source it? ;-)
>
> Regards,
>
> Harri
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From james at daa.com.au  Sun Jan 30 03:13:22 2000
From: james at daa.com.au (James Henstridge)
Date: Sun, 30 Jan 2000 08:13:22 GMT
Subject: Subclassing ExtensionClasses in C?
Message-ID: <3893F07D.A44FE06D@daa.com.au>

Hello,

I have a small question about using ExtensionClass with my python
module.

It was suggested to me recently that I may want to look at rewriting my
python extension pygtk to use ExtensionClasses (to increase speed,
decrease mem usage, and a few other benefits).  For those who have not
used pygtk, it is a wrapper for the GTK GUI widget set (see
www.gtk.org).

If I switched over to using ExtensionClass, I would want to create a
heirachy of ExtensionClasses implemented in C as wrappers for the
various widgets.  Looking through the documentation for ExtensionClass,
it has examples of how to create classes in C that can be subclassed in
python, but it does not seem to have any examples of how to write one
extension class that subclasses another extension class.

I assume that this is possible, but I haven't seen any examples.  Is it
possible to do with the public ExtensionClass API, or is it a little
more difficult.  If anyone can help with this, I would appreciate it.

Thanks in advance,

James Henstridge.


From aahz at netcom.com  Sun Jan  2 18:17:30 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Jan 2000 23:17:30 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com> <84nfct$bal$1@news1.xs4all.nl> <umav6ssq7uubv4gpb192lp2fs1jcan6hc2@4ax.com> <84olef$5j1$1@news1.xs4all.nl>
Message-ID: <84omaa$qln$1@nntp9.atl.mindspring.net>

In article <84olef$5j1$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
>
>No, not really ;-). When I was typing this, I suddenly realized that
>I don't know of any database server where it's a good idea to slurp
>thousands of rows over the network to the client. 

Mind expanding on this?  Are you asserting it's also a Bad Idea to slurp
multi-megabyte BLObs to a client?
--
                      --- Aahz (@netcom.com)

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

Ninth Virtual Anniversary -- 364 days and counting


From ivanlan at callware.com  Wed Jan 19 19:01:54 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 19 Jan 2000 17:01:54 -0700
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu> <865ij9$ia4$1@newshost.accu.uu.nl>
Message-ID: <38865072.DBD8B4E5@callware.com>

Hi All--

Martijn Faassen wrote:
> 
> This is nothing new. Back during the last Python conference, me and some
> other intrepid non-conference-attenders had the run of the newsgroup,
> and we redesigned Python into a more sane language already:
> 
> http://x44.deja.com/[ST_rn=ps]/getdoc.xp?AN=410321807&search=thread
> [and followups]
> 
> And the immensely positive reaction by some famous Pythoneers is here:
> 
> http://x23.deja.com/[ST_rn=ps]/viewthread.xp?AN=412546299&search=thread
> 
> this-year-I-will-attend-so-someone-else-can-have-the-run-ly yours,
> 

Well, this year I am *not* attending.  Therefore, I plan to redesign
Python so that you will all have to read and write Mayan glyphs in order
to write any programs.  You'd better hope that Andy Robinson gets to
work on (*really*) internationalizing and double-bytifying Python.

<you-is-a-wall-eyed-oyster-yourself>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From pearu at ioc.ee  Tue Jan 25 14:26:24 2000
From: pearu at ioc.ee (Pearu Peterson)
Date: Tue, 25 Jan 2000 21:26:24 +0200 (EET)
Subject: f2py 1.116: The Fortran to Python Interface Generator
Message-ID: <Pine.HPX.4.05.10001252125100.10822-100000@egoist.ioc.ee>

I am pleased to announce the first public release of f2py 1.116:

Writing Python C/API wrappers for Fortran routines can be a very
tedious task, especially if a Fortran routine takes more than 20
arguments but only few of them are relevant for the problems that they
solve.

The Fortran to Python Interface Generator, or FPIG for short, is a
command line tool (f2py) for generating Python C/API modules for
wrapping Fortran 77 routines, accessing common blocks from Python, and
calling Python functions from Fortran (call-backs).

The tool can be downloaded from

    http://cens.ioc.ee/projects/f2py2e/

where you can find also information about f2py features and its User's
Guide.

f2py is released under the LGPL license.

With regards,
     Pearu Peterson <pearu at ioc.ee>

<P><A HREF="http://cens.ioc.ee/projects/f2py2e/">f2py 1.116</A> - The
Fortran to Python Interface Generator (25-Jan-00)





From phd at phd.russ.ru  Wed Jan 12 04:32:08 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 12 Jan 2000 09:32:08 +0000 (GMT)
Subject: Python BUG?
In-Reply-To: <85hg1q$2nli$1@gavrilo.mtu.ru>
Message-ID: <Pine.LNX.4.21.0001120930280.9225-100000@fep132.fep.ru>

On Wed, 12 Jan 2000, Oleg Orlov wrote:
> This script:
> 
> d = [1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L,
>     10L, 9L, 8L, 7L, 6L, 5L, 4L, 3L, 2L, 1L]
> print reduce(lambda x, y: x*x + y*y, d)
> 
> hangs up my python interpreter (on NT4).
> 
> Is it a bug or my mistake?
> How about UNIX?

   My computer worked about 10 minutes and printed a very looong result. :)
Pentium-150, Debian GNU/Linux.

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From fredrik at pythonware.com  Thu Jan 20 03:23:21 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 20 Jan 2000 09:23:21 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse>
Message-ID: <00cc01bf631f$9dda5210$f29b12c2@secret.pythonware.com>

Evguenii Smogailov <sei at sunse.jinr.ru> wrote:
> Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.

hmm.  I'm beginning to think that maybe, just maybe, you
should do a little more research on Python's origins.

</F>

PS: start here: http://www.cwi.nl/~steven/abc/qr.html




From grant at nowhere.  Wed Jan 19 12:28:35 2000
From: grant at nowhere. (Grant Edwards)
Date: Wed, 19 Jan 2000 17:28:35 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <025301bf6298$b6dc38a0$01ffffc0@worldnet.att.net>
Message-ID: <slrn88bt23.4pp.grant@grante.comtrol.com>

>Take a look at http://www.slipstick.com/exchange

[...]

>Good luck!

Thanks!  

It looks like there's some useful stuff there.  Right now, I'm
playing with Python/COM/MAPI.  That looks like it _should_ be
the simplest way to get mail sucked out of Exchange Server.  If
that doesn't work out, I'll have to take a closer look at some
of the other options.

-- 
Grant Edwards                   grante             Yow!  An INK-LING? Sure --
                                  at               TAKE one!! Did you BUY any
                               visi.com            COMMUNIST UNIFORMS??


From sdfsd at fdfs.com  Sat Jan 22 16:13:10 2000
From: sdfsd at fdfs.com (sdfsd at fdfs.com)
Date: Sat, 22 Jan 2000 15:13:10 -0600
Subject: get money for browsin web sites!                -- 22791
Message-ID: <220100151310@fdfs.com>

Everyone likes making easy money.  This place pays you 50 cents for every hour you browse the web!!!
Get even more if you refer people!  I just got a check for $36.00!!!

Sign up here:
http://www.alladvantage.com/go.asp?refid=IMY700








.7_jwRhg)/


From a.eyre at optichrome.com  Fri Jan 21 09:25:23 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Fri, 21 Jan 2000 14:25:23 -0000
Subject: Changing the passed arguments value after returning from a module
In-Reply-To: <002c01bf6418$f8b99d90$3acbd9c2@peridot.optichrome.com>
Message-ID: <002e01bf641b$5a5f8fd0$3acbd9c2@peridot.optichrome.com>

> >>> import mymodule
> >>> i = (0,)
> >>> apply(mymodule.test, i)
> >>> print i
> 101

Slight correction here. Probably obvious, but...

>>> import mymodule
>>> i = (0,)
>>> apply(mymodule.test, i)
>>> print i
(101,)

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From darnold at corecomm.net  Sat Jan 29 14:07:33 2000
From: darnold at corecomm.net (derek a. arnold)
Date: Sat, 29 Jan 2000 14:07:33 -0500
Subject: python and college/university programs
Message-ID: <38933A74.F50F352A@corecomm.net>

Are there any schools that are using Python as their primary language?
By primary, I mean it's the course that teachers use to teach computer
science to students at the intro level.  E-mail me and let me know.

Derek Arnold
Python Newbie



From bradh at mediaone.net  Fri Jan 21 23:26:01 2000
From: bradh at mediaone.net (Brad Howes)
Date: Sat, 22 Jan 2000 04:26:01 GMT
Subject: Python dumps core with this.. how come?
References: <38891316.393423@news.erols.com>
Message-ID: <m3snzq7mvo.fsf@h0050047c4794.ne.mediaone.net>

katz at glue.umd.edu (Roey Katz) writes:

<snip>

> When I import this module, the interpreter segfaults. 

<snip>

> class basic_monitor:
> 
>     def __init__( self ):
>         self.xres = []     # list of resource entries

This is innocuous line is the start of your troubles. Below you declare
__setattr__ which tells Python you want to handle all attributes settings for
this instance. So...

>     def __setattr__( self, name, val ):
>         origval = self.res[ name ]

But now you've asked for the attribute 'res' which has yet to be defined for
this instance. You've declared __getattr__ which tells Python to call it when
an attribute is not available (which it isn't yet)

>     def __getattr__( self, name ):
>         return self.res[ name ]

Here, we try to get the same attribute Python failed to get. So, Python calls
your hook again, and again, and again, until lack of stack tells Python to stop
this silliness and to go get a beer.

Since you've defined both hooks, you're going to have to work with the
instance's attribute dictionary __dict__

  dict = self.__dict__

and do your attribute setting using the usual dictionary methods (you could
also declare "class" attributes with default values) That should keep Python up
a bit longer.

Brad

-- 
Brad Howes                                              C C++ HTML Java Python
Waltham, MA                                                         Programmer

    "Were people this stupid before TV?" -- _White Noise_ by Don DeLillo


From news at dorb.com  Fri Jan  7 22:07:58 2000
From: news at dorb.com (Darrell)
Date: Fri, 7 Jan 2000 22:07:58 -0500
Subject: Newbie Win32 Questions
References: <855mj9$ckc$1@nnrp1.deja.com>
Message-ID: <LPxd4.695$mj7.17622@typhoon.nyroc.rr.com>

<tuttled at my-deja.com> wrote in message news:855mj9$ckc$1 at nnrp1.deja.com...
>
>
> Help please. I've got a couple of questions I
> haven't been able to find any answers to.
>
> 1) Can you make a stand alone app? Meaning is
> there a way to compile and distribute one or more
> files without having to install Python on every
> machine you want to run the app on?
>
Yes
http://starship.python.net/crew/gmcm/distribute.html

This is worth a look as well.
ftp://ftp.interet.com/pub/pylib.html

> 2) If not, how do protect source code from prying
> eyes?
>
Welcome to open source.
You could just send out your .pyc files although they aren't secure.

If you wanted to get tricky this might help.
http://www.lyra.org/greg/python/imputil.py
--
--Darrell






From tim_one at email.msn.com  Thu Jan 13 02:43:52 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 13 Jan 2000 02:43:52 -0500
Subject: Python BUG?
In-Reply-To: <Pine.LNX.4.21.0001120930280.9225-100000@fep132.fep.ru>
Message-ID: <000401bf5d99$f00804a0$d62d153f@tim>

[Oleg Orlov]
> This script:
>
> d = [1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L,
>     10L, 9L, 8L, 7L, 6L, 5L, 4L, 3L, 2L, 1L]
> print reduce(lambda x, y: x*x + y*y, d)
>
> hangs up my python interpreter (on NT4).
>
> Is it a bug or my mistake?
> How about UNIX?

[Oleg Broytmann]
>    My computer worked about 10 minutes and printed a very looong
> result. :)
> Pentium-150, Debian GNU/Linux.

This is a cute one!  It's much faster than you think <wink>.

Do x=reduce(...) instead of printing it, and it should take about 30 seconds
on your machine.

The enormous expense here isn't the (unintended) iterated squaring, it's the
conversion of the gigantic result (well over 300,000 bits) into a base-10
string (on the order of 100,000 digits).

Conversion to decimal is a quadratic-time (in the number of bits) process.
Converting to a hex (hex(x)) or octal string instead is almost instantaneous
(in Python 1.5.2; in earlier Pythons conversion to hex was also
quadratic-time).

longs-have-subtleties-too-ly y'rs  - tim





From mwh21 at cam.ac.uk  Sat Jan 15 10:28:08 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 15 Jan 2000 15:28:08 +0000
Subject: Help me improve my Python Skills, please
References: <m2g0vzsrte.fsf@excalibur.peernet> <38808726.751144390@news1.on.sympatico.ca>
Message-ID: <m3iu0vl5h3.fsf@atrus.jesus.cam.ac.uk>

rjroy at takingcontrol.com (Robert Roy) writes:

> On 15 Jan 2000 00:43:41 -0700, tarude at whitestar.localdomain (Trebor A.
> Rude) wrote:
> 
> >
> >I just started picking up Python a few days ago, and am rather
> >enjoying it.  To help me learn it, I converted one of my old TCL
> >scripts into Python, which I've attached below. However, this being my 
> >first serious python script, I'm pretty sure I'm doing some things
> >less efficiently than necessary, through lack of familiarity with the
> >library and language in general. So I'd like to enlist the group's
> >help in improving the code. Logic, style, better ways of doing things, 
> >whatever.  Thanks in advance.
> >
> >A little explanation before the code.  The script recursively descends 
> >a directory structure, building a .jdb (Jukebox database) file for
> >each directory based on the mp3 tag info of the mp3s in that
> >directory and all its subdirectories. It also repairs damaged comment
> >fields caused by a slightly buggy encoder program.
> >
> The os.path.walk method provides an elegant way of recursing a
> directory structure. It might be useful here.

Actually, it's not that helpful in this case; this is because of the
line in the requirements "building a .jdb (Jukebox database) file for
each directory based on the mp3 tag info of the mp3s in that directory
and ALL ITS SUBDIRECTORIES" (my emphasis). os.path.walk does a
pre-order traversal of the tree, which means there'd be faffing around
to do to get the files from the subdirectories indexed into the
database for the root. Sure, it could be done, but rolling your own
post-order traversal has to be easier.

Cheers,
Michael


From skaller at maxtal.com.au  Sun Jan  9 09:51:06 2000
From: skaller at maxtal.com.au (skaller)
Date: Mon, 10 Jan 2000 01:51:06 +1100
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
		<14452.45436.641890.197509@beluga.mojam.com>
		<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
		<14452.60226.853060.116123@beluga.mojam.com>
		<011501bf58a4$7b0db540$0100a8c0@rochester.rr.com>
		<3877513B.73B762F4@maxtal.com.au>
		<004901bf5a08$11a35be0$0100a8c0@rochester.rr.com>
		<387890F6.2D1136F8@maxtal.com.au> <14456.39938.51446.999159@beluga.mojam.com>
Message-ID: <3878A05A.10CB26DF@maxtal.com.au>

Skip Montanaro wrote:
> 
>     >> This is a common thing, does this express the idea ?
>     [ ... out-of-line example snipped ... ]
> 
>     skaller> Yes, that provides the required functionality, but has a
>     skaller> serious problem: cases have to be defined 'out of line', and
>     skaller> therefore do not have access to the current scope.
> 
>     skaller> So I'm looking for a syntax for providing the above
>     skaller> functionality _inline_.
> 
> While not strictly inline, why not define your actions as methods of a
> common class, then share the state they need access to as instance
> attributes?

	Precisely because it isn't inline. Localisation is lost.
Access to local state is lost, in Python, it cannot be recovered
in the case we want to rebind variables.

	put it this way: how would you object, if I said
you should always use a function:

	iterate(f, sequence)

instead of a for loop? The objection is the same.

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From volumeregeling at hotmail.com  Mon Jan 24 05:51:42 2000
From: volumeregeling at hotmail.com (volumeregeling)
Date: Mon, 24 Jan 2000 10:51:42 GMT
Subject: problems with COM objects
Message-ID: <388c2dca.2584926@122.0.0.250>

hi,

I am implementing COM objects in my program. I close the com objects
through their own exit function for instance Editor.exit() where
Editor is the COM object.

when i invoke the Exit function python comes with an OLE Error

the following error appears : 
com_error: (-2146788248, 'OLE error 0x800a9c68', (0, 'ComEditor',
'Application-defined or object-defined error', None, 1000095,
-2146788248), None)

Can anybody telle me if this is a COM object error or is there going
something wrong in my python program.

By the way if you can give me a method to somekind of release the COM
object without needing an exit function from within the comobject
please tell me.


Thanks,

Volumeregeling.


From tim_one at email.msn.com  Tue Jan  4 20:27:36 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 4 Jan 2000 20:27:36 -0500
Subject: Simple cartesian product
In-Reply-To: <slrn8755l1.gvl.neelk@brick.cswv.com>
Message-ID: <000801bf571c$0cf448a0$1a2d153f@tim>

Search in DejaNews for p13x25.

c'mon-folks-new-millennium-new-topics<wink>-ly y'rs  - tim





From steve at acer.spvi.com  Sat Jan  8 21:11:11 2000
From: steve at acer.spvi.com (Steve Spicklemire)
Date: Sat, 8 Jan 2000 21:11:11 -0500 (EST)
Subject: pipe problem on FreeBSD....
In-Reply-To: <858iu6$m75$1@news.kolumbus.fi> (torppa@polykoira.megabaud.fi)
References: <200001082039.PAA12903@acer.spvi.com> <858iu6$m75$1@news.kolumbus.fi>
Message-ID: <200001090211.VAA13772@acer.spvi.com>

Thanks Jarkko! 

       OK... I think I see now how my program was flawed. However
there is still something fishy..... 

when I use *your* program I can do

./jarrkki-tryit | more

and it behaves quite well. I also tried with threads enabled
and without. On the other hand the python version dies with
the broken pipe error (on FreeBSD only..). I'm assuming that
either python is doing something fundamentally different than
the c program, or there is some other library that is being
used by python but not by the c program (that is broken on
FreeBSD, but OK on linux......) whew!

anyway.. thanks for the input.... I've really got to track
this down.. and any insights are most welcome!

-steve

>>>>> "Jarkko" == Jarkko Torppa <torppa at polykoira.megabaud.fi> writes:

    Jarkko> In article <200001082039.PAA12903 at acer.spvi.com>, Steve Spicklemire wrote:
    Jarkko> >   Hmmm.. what I thought started out as a "Zope Bug" isn't. Let me 
    Jarkko> >forgo all the history here and cut to the current analysis. There
    Jarkko> >seems to be a problem on FreeBSD with python sending large amounts of 
    Jarkko> >data through a pipe that blocks.
    Jarkko> ...
    Jarkko> >Here is the reference to that problem report:
    Jarkko> >http://lists.zope.org/pipermail/zope/1999-December/015874.html
    Jarkko> ...
    Jarkko> >I think I've now simplified it to it's most basic form:
    Jarkko> 
    Jarkko> This is quite different matter that what you are seeing in Zope.
    Jarkko> >This python script:
    Jarkko> ...
    Jarkko> >fails on FreeBSD when invoked thusly:
    Jarkko> >
    Jarkko> >pluto.spvi.com> python tryit.py | head -3
    Jarkko> ...
    Jarkko> >IOError: [Errno 32] Broken pipe
    Jarkko> >where this c program does not:
    Jarkko> ..
    Jarkko> 
    Jarkko> Here is c program that does tries more or less match python's 
    Jarkko> behaviour. 
    Jarkko> ----
    Jarkko> #include <stdio.h>
    Jarkko> #include <err.h>
    Jarkko> #include <signal.h>
    Jarkko> 
    Jarkko> int main() {
    Jarkko>   size_t charCount,c;
    Jarkko>   char buffer[1024];
    Jarkko>   FILE *f;
    Jarkko> 
    Jarkko>   signal(SIGPIPE,SIG_IGN);
    Jarkko> 
    Jarkko>   if((f = fopen("blah","r")) != NULL) {
    Jarkko>     do {
    Jarkko>       charCount = fread(buffer,1,sizeof(buffer),f);
    Jarkko>       c=fwrite(buffer,1,charCount,stdout);
    Jarkko>       if ( c != charCount )
    Jarkko> 	err(1,"WriteFailed");
    Jarkko>     } while(charCount == sizeof(buffer));
    Jarkko>     close(f);
    Jarkko>   }
    Jarkko>   else
    Jarkko>     printf("Can't open file... blah\n");
    Jarkko> }
    Jarkko> ---
    Jarkko> tuhnu[2]%./tryit | head -3
    Jarkko> ...
    Jarkko> tryit: WriteFailed: Broken pipe
    Jarkko> 
    Jarkko> >My thinking is that they should behave in the same way. (This also
    Jarkko> >happens with 'more' and other such blocking programs.) Where in python
    Jarkko> >should I look for the difference between tryit.c and tryit.py? Of
    Jarkko> >course on linux both programs *do* behave the same way... I just
    Jarkko> >created the 'c' version to see if it was deep in FreeBSD or somewhere
    Jarkko> >in the python port. The python and c version work correctly in linux.
    Jarkko> ><dang!>
    Jarkko> 
    Jarkko> By default SIGPIPE terminates the program, also your cprogram
    Jarkko> did not check writes return value so it could not have
    Jarkko> caught the error anyhow. 
    Jarkko> 
    Jarkko> Behaviour you are seeing is correct. But this is different from
    Jarkko> blocking it should not give SIGPIPE, writing thread should
    Jarkko> just stall untill there is space in pipe to put the data.
    Jarkko> 
    Jarkko> This python piece does blocking reads on stdin
    Jarkko> 
    Jarkko> import time,sys
    Jarkko> while 1:
    Jarkko>     r=sys.stdin.read(1024)
    Jarkko>     if len(r) == 0:
    Jarkko>         break
    Jarkko>     sys.stdout.write(r)
    Jarkko>     time.sleep(30)
    Jarkko> 
    Jarkko> -- 
    Jarkko>  Jarkko Torppa                torppa at staff.megabaud.fi
    Jarkko>   Megabaud Internet-palvelut
    Jarkko> -- 
    Jarkko> http://www.python.org/mailman/listinfo/python-list
    Jarkko> 
    Jarkko> 



From aahz at netcom.com  Sat Jan  8 11:44:21 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jan 2000 16:44:21 GMT
Subject: Looking for threading tutorial/HOWTO
References: <m3zougva3q.fsf@havenrock.com>
Message-ID: <857ph5$jsh$1@nntp3.atl.mindspring.net>

In article <m3zougva3q.fsf at havenrock.com>,
Matt Gushee  <mgushee at havenrock.com> wrote:
>
>I would like to learn how to program threads in Python. Or maybe I
>should say, I'd like to use threads in a Python program, so I need to
>learn how they work. Can anyone recommend a good online 
>introduction to the subject (I like books but can't afford to buy one
>just now)? Code examples might be good, too, if they are well commented.

This following code example is not well-documented, but should be pretty
easy to follow; it's designed to test a web server with multiple
clients:

import urllib, time
import threading

retrieveDoc = 1
numPings = 20
numThreads = 5

url = 'http://www.foo.com/'



class ping ( threading.Thread ) :
	def __init__ ( self, url, numPings, retrieveDoc=1 ) :
		self.url = url
		self.numPings = numPings
		self.retrieveDoc = retrieveDoc
		threading.Thread.__init__(self)

	def run ( self ) :
		StartTime = time.time()
		for i in range(self.numPings):
			page = urllib.urlopen ( self.url )
			if self.retrieveDoc:
				page.read()
			page.close()
		EndTime = time.time()
		self.TotalTime = EndTime - StartTime



if __name__ == '__main__' :
	threadList = []
	for i in range(numThreads) :
		thread = pingPing ( url, numPings, retrieveDoc )
		threadList.append ( thread )

	StartTime = time.time()
	for thread in threadList :
		thread.start()

	while threading.activeCount() > 1 :
		print ".",
		time.sleep(1)
	EndTime = time.time()
	TotalTime = EndTime - StartTime
	print

	TotalPings = 0
	ThreadTime = 0
	for thread in threadList :
		TotalPings = TotalPings + thread.numPings
		ThreadTime = ThreadTime + thread.TotalTime

	PingAvg = TotalPings / TotalTime
	ResponseAvg = ThreadTime / TotalPings

	print "Threads: ", numThreads
	print "Pings:", TotalPings
	print "Total time:", TotalTime
	print "Pings per second:", PingAvg
	print "Average response time (secs):", ResponseAvg

--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From sp00fD at yahoo.com  Tue Jan 18 15:38:36 2000
From: sp00fD at yahoo.com (sp00fD)
Date: Tue, 18 Jan 2000 20:38:36 GMT
Subject: **kwargs
Message-ID: <862j09$o1k$1@nnrp1.deja.com>

Does **kwargs represent a dictionary?  If so, can I have a method like
such

def foo(self, foo=None, bar=None, **kwargs)

and then do something like

this_dict = {}
..insert stuff into this_dict..

my.foo(foo="something", bar="something else", this_dict)

? How would I do something like this ?


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


From mal at lemburg.com  Mon Jan 24 11:11:22 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 24 Jan 2000 17:11:22 +0100
Subject: ODBC connectivity to SQL7.0 mdb, UNIX - ODBC Bridges
References: <3887F0D5.EBC18C60@bby.com.au>
Message-ID: <388C79AA.CC1C6BD2@lemburg.com>

Peter Larbalestier wrote:
> 
> G'day All,
> 
> A number of queries:
> 
> 1. Windows connection via ODBC to a SQL mdb on NT Server 4.0
>     I have been making a connection to our SQL7.0 mdb via ODBC after
> having set up a 'DSN' in the 'ODBC Data Source Administrator' window..
> Most of the examples I have viewed so far refer to this practice.
>     i.e. >>>import odbc, dbi
>           >>>mydb = odbc.odbc(dsn,uid,pwd)
> 
> It worries me though that this seems to create a "permanent" connection
> to the database as there is an open port listening for any queries.
> 
> Does anyone have any sample code for connecting on the fly to a SQL7.0
> mdb WITHOUT using the 'ODBC Data Source Administrator' to setup a
> "permanent" DSN.
> 
> Having programmed in VB I simplly conect on the fly using ADO to our
> SQ7.0 mdb. Can this be done with Python using mxODBC?
>     I believe so ! but can anyone help me with an example?

Not an example, but you can use the mxODBC ODBC.Windows.DriverConnect()
API to create a connection using all the options you need to specify
for the DSN... perhaps there is a way to tell the ODBC manager
to create a DSN on the fly simply by providing the file name ?!
 
> 2. UNIX connectivity to SQL7.0 mdb on NT Server 4.0
>     It seems that using ODBC - ODBC Bridge is the way to go but this is
> after the NT problem.

You can do this in these ways:

1. Use one of the OpenLink or Merant (InterSolv) multi-tier
   driver sets

2. Use a ODBC-bridge such as the one sold by EasySoft

3. Write a Python server for your app which runs on NT and
   serves a client on the Unix machine (not as hard as you might
   expect -- see SocketServer.py).
 
> I hope that this forum will provide some definitive answers for all us
> 'newbies' in regards to these important issues.

Me too ;-)

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





From jrush at summit-research.com  Fri Jan 14 05:30:41 2000
From: jrush at summit-research.com (Jeff Rush)
Date: 14 Jan 2000 04:30:41 -0600
Subject: mxODBC on Linux
References: <387E02F7.7778E2F2@exceptionalminds.com>
Message-ID: <VrDQRddcbbs9-pn2-uKVxUXVfjmKw@Fortress>

Yes, I have.  I use it all the time here, both standalone and with Zope, 
via the ZODBCDA adapter software.
I made an RPM for it, but I use it with Solid Technology's SQL product, not
PostgreSQL.  Here is the
source RPM, if you want to rebuild it for PostgreSQL or study what I did.

    http://starship.python.net/crew/jrush/RPMs/mxODBC-Solid-1.1.1-2.src.rpm

-Jeff Rush

On Thu, 1 Jan 1970 02:59:59, Timothy Grant <tjg at avalongroup.net> wrote:

> has anybody successfully configured mxODBC on Linux? I'm trying to
> connect to a PostgreSQL database and am being phenomenally unsuccessful.
> Works great on Windows, just not on Linux.



From sjohnson at csnet.net  Wed Jan 26 21:50:34 2000
From: sjohnson at csnet.net (Scott Johnson)
Date: Thu, 27 Jan 2000 02:50:34 GMT
Subject: Another newbie question
References: <000001bf651d$f05ac240$132d153f@tim>
Message-ID: <_nOj4.894$yY6.13534@news.goodnet.com>

I've read the tutorial and I don't get lambda.

----------
"Tim Peters" <tim_one at email.msn.com> wrote...

> [Interrupting Michael Hudson's fine advice to inject a nit]
> 
> [Roey Katz]
> > this brings up another concern. Given
> > the following definition:
> >
> >    class A:
> >
> >       def __getattr__( self, name ):
> >         return -1
> >       def __setattr__( self, name, val ):
> >          self.__dict__[ name ] = val
> >
> > I see that self.__dict__ must be a special variable
> > in that setting it somehow avoids a recursive call
> > to __setattr__()?
> 
> [Michael]
> > Yup. Look in Objects/classobject.c around about line 580
> > if you want the gory details.
> 
> While true, it doesn't apply to Roey's example:  he's not setting __dict__.
> He's binding a key *of* self.__dict__; self.__dict__ itself is merely
> referenced.  So Roey should really be wondering why the reference to
> self.__dict__ isn't invoking __getattr__ (&, of course, that's also due to
> special-case __dict__ magic, but in __getattr__'s implementation).
> 
> I don't mean to be overly picky; it's just that the excruciating details are
> vital when mucking with __getattr__/__setattr__, and *any* small
> misconception about them will eventually bite hard.
> 
> > ...
> > Bizarrely,
> >
> > filter(dest.append,src)
> >
> > is probably more memory efficient (the map form conses up a
> > list of return values, but as append always returns None -
> > which is false - filter `rejects' every element) - but if
> > you write code like this, you deserve to lose.
> 
> Yes, pointing him to the most bizarre endcases of Python behavior will
> surely aid him in "resetting his brain" <wink>.
> 
> Guido gave some excellent semi-exasperated advice (similar in spirit to
> parts of Michael's advice that I've snipped) many moons ago, when the whole
> newsgroup was temporarily filled with msgs of this character:  he invited
> people to just *try* writing a Python program without using any __xxx__
> hooks.  I expect some people were amazed at how much easier their lives got!
> If he were to get that exasperated again today, I bet he'd expand his
> invitation to include avoiding map, filter, reduce and lambda too.
> 
> you're-not-ready-to-use-them-in-python-before-you're-
>     happy-without-them-ly y'rs  - tim
> 
> 
> 



From gerrit.holl at pobox.com  Sat Jan 29 14:12:10 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 29 Jan 2000 20:12:10 +0100
Subject: Imagemagick
In-Reply-To: <wzzotoal2k.fsf@sunshine.cs.uu.nl>; from piet@cs.uu.nl on Sat, Jan 29, 2000 at 05:39:47PM +0100
References: <wzzotoal2k.fsf@sunshine.cs.uu.nl>
Message-ID: <20000129201210.A4953@stopcontact.palga.uucp>

Piet van Oostrum wrote on 949163987:
> Does anybody have a copy of Python/Imagemagick or pymagick or a working link?

Google -> PythonMagick -> I'm feeling lucky
http://starship.skyport.net/crew/zack/pymagick

Dat was makkelijk.
That was easy.
Das waere leicht.
C'etait facile.

doei-bye-gutentag-bounjour-tjau-alegra-ly y'rs - gerrit

-- 
Please correct any bad English, German, French or Dutch you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From a.eyre at optichrome.com  Mon Jan 31 05:19:57 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 10:19:57 -0000
Subject: Dynamic class construction?
In-Reply-To: <8707c0$ua2$1@nnrp1.deja.com>
Message-ID: <000301bf6bd4$b9210150$3acbd9c2@peridot.optichrome.com>

> def newClass(name, ops, superclass):
> 	class NewClass(superclass):
> 		pass
> 	for op in ops.keys():
> 		setattr(NewClass, op, ops[op])
> 	NewClass.__name__ = name
> 	return NewClass
> 
> def anOp(self):
> 	return 'This is an instance of the class %s' % \
> 		self.__class__.__name__
> 
> someOps={'__str__':anOp}
> G=newClass('G', someOps, Group) # Group defined elsewhere
> print G
> <class __main__.G at f92a8>

There's already a module which does this...

import new

def anOp(self):
	return 'This is an instance of the class %s' % \
		self.__class__.__name__

someOps={'__str__':anOp}
G=new.classobj('G', (Group,), someOps) # Group defined elsewhere

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From hpasanen at my-deja.com  Thu Jan 27 11:10:57 2000
From: hpasanen at my-deja.com (hpasanen at my-deja.com)
Date: Thu, 27 Jan 2000 16:10:57 GMT
Subject: ObjectDomain status?
Message-ID: <86pqmb$d22$1@nnrp1.deja.com>

Hi,

This brushes on Python, so I though someone here might know...

I've recently evaluated ObjectDomain UML modeller for our internal use.
(http://www.objectdomain.com/)   The unique feature of their product is
that it contains an embedded JPython as a scripting language, while
the rest of it resembles Rational Rose a whole lot from the UI
perspective.  (perhaps too much?)

All in all, ObjectDomain 2.5 seems to work reasonably well, but I can't
get any response from ObjectDomain to my queries.  I wonder if the
company has gone belly-up?

Anyone in the know?  Would be a shame to see a prominent Python product
going down the drain, but perhaps they'll Open-Source it? ;-)

Regards,

Harri


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


From sjturner at ix.netcom.com  Tue Jan  4 10:25:17 2000
From: sjturner at ix.netcom.com (Stephen J. Turner)
Date: Tue, 04 Jan 2000 10:25:17 -0500
Subject: RHS Python 1.5.2 RPM and RTLD_GLOBAL
Message-ID: <387210DD.E47C8F74@ix.netcom.com>

FYI:

The Python 1.5.2 RPM from Red Hat Software (included with RH Linux 6.1)
apparently still uses the dlopen() RTLD_GLOBAL flag when importing
shared object extensions.  The use of RTLD_GLOBAL -- which made global
symbols exported by dynamically loaded extension modules visible to
other shared objects -- was introduced with Python 1.5b2 but later
removed as of 1.5.2a1.

Red Hat's decision to retain this "feature" has a potentially adverse
effect on any RH 6.1 users of the informixdb DBAPI module.  It so
happens that one of Informix's runtime libraries contains a function
named "initsocket" -- most unfortunate, since this is also the name of
the Python socket module's initialization routine.  If the user imports
socket before informixdb, then when the Informix runtime tries to call
its own initsocket, it winds up in the wrong function.  Bad things
ensue.

Have any other Python extensions been similarly affected?  And does
anyone know why RHS chose to keep the pre-1.5.2 import behavior?

--
Stephen J. Turner <sjturner at ix.netcom.com>


From calishar at my-deja.com  Mon Jan 10 14:05:20 2000
From: calishar at my-deja.com (calishar at my-deja.com)
Date: Mon, 10 Jan 2000 19:05:20 GMT
Subject: Problem distributing a Gadfly app
Message-ID: <85dah2$ip8$1@nnrp1.deja.com>

Hello folks,

  I am just in the finishing off stage of an application for work that
uses the gadfly database package. I am trying to build an installable
application using installer (from Gordon Macmillan?) and running into a
problem.

  The application builds alright, but when I install it on the other
machine, and go to run it, it fails.

  After re-building with the debug flag on, use runw off, and support
on, I saw that it is trying to load the sql.mar file from the c: drive.
Well, it ain't there. If I try to add the sql.mar file to the
dependencies line of the cfg file, builder fails, if I add it to the
includes line it fails.

  Has anyone run into this before?


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


From s.schwarzer at ndh.net  Tue Jan  4 11:28:04 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Tue, 04 Jan 2000 17:28:04 +0100
Subject: makepage-0.1beta.py
References: <mt2.0-29451-946955276@news.informatik.uni-bonn.de>
Message-ID: <38721F94.9DF7B1B9@ndh.net>

Hello,

[on comp.lang.python.announce ...]
Stefan Schwarzer wrote:
> I would like to announce makepage.py 0.1 beta.
> [...]

Two more notes:

- makepage is open source software
- You can get the latest version (currently 0.1.1) as
  http://www.ndh.net/home/sschwarzer/download/makepage.zip

Stefan


From s.schwarzer at ndh.net  Wed Jan 19 20:08:51 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 20 Jan 2000 02:08:51 +0100
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <862of8$57j$1@news.psc.edu>
Message-ID: <38866023.F934AFEB@ndh.net>

Hi John,

John Kochmar schrieb:
> BTW, I'm a dinosour who likes to vi, and hates when an editor sees that
> I've hit a tab and says to itself "gee, he hit a tab, so he must really
> want N spaces, so I'll silently replace it for him..."  If I hit a tab,
> darn it, I MENT a tab.

Isn't there a option in your version of vi to tell it to use always
tabs when you hit tab? At least in Vim, you can easily configure
how you want presses of the tab key to be handled.

Stefan


From news at dorb.com  Tue Jan 18 15:01:18 2000
From: news at dorb.com (Darrell)
Date: Tue, 18 Jan 2000 15:01:18 -0500
Subject: Monitoring file changes in NT
References: <3h2h4.362$B5.8315@news.more.net>
Message-ID: <3D3h4.6233$U5.255821@typhoon.nyroc.rr.com>

http://www.dorb.com/darrell/win32WorkSvr/makeThumbsDG.py

--
--Darrell
"Jeff Kunce" <kuncej at mail.conservation.state.mo.us> wrote in message
news:3h2h4.362$B5.8315 at news.more.net...
> I'd like my python program to know when
> directories and files have changed on
> an NT system.
>
> Could anyone share an example of how to use
> win32file.FindFirstChangeNotification()
> and its relatives?
>
> Thanks.
>
>   --Jeff
>
>
>
> --
> http://www.python.org/mailman/listinfo/python-list




From kputland at servicemagic.com  Fri Jan 28 16:21:30 2000
From: kputland at servicemagic.com (Karl Putland)
Date: Fri, 28 Jan 2000 14:21:30 -0700
Subject: memory
References: <86q8ee$2g7k$1@noao.edu> <20000128132509.A565@stopcontact.palga.uucp>
Message-ID: <00ee01bf69d5$a55ba360$8901010a@w3denver.com>

----- Original Message -----
From: "Gerrit Holl" <gerrit.holl at pobox.com>
> Matt Cheselka wrote on 948999934:
> > Hi,
> >
> > Are there any issues with freeing memory in python?  Say for example I
do a
> >
> > d = array.array('i')
> > for i in range (2000):
> >     d.append(0)
> >
> > Do I need to free 'd' before I quit the program?
>
> Not necissary, but you _can_ do it with the del statement:
> del d
> removed the object and frees the memory.
>

Actually that just removes the binding to the name 'd'.
and the memory of d is not reclamed until later.

If something else is holding on to a reference of d the nothing is gained
save for cleaning up the namespace.

-Karl






From wiktor1325 at my-deja.com  Wed Jan  5 21:49:20 2000
From: wiktor1325 at my-deja.com (wiktor1325 at my-deja.com)
Date: Thu, 06 Jan 2000 02:49:20 GMT
Subject: How to with **kw
Message-ID: <850vrc$v3g$1@nnrp1.deja.com>

Is there a way to pass **kw args
from one function to another,like:

def func1(**kw):
  for key in kw.keys():
     print key

def func2(**kw):
  a_way_to_call_func1_with(kw)

Thanks in advance

Wiktor Sadowski


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


From michelorengo at netscape.com  Wed Jan 26 10:49:49 2000
From: michelorengo at netscape.com (Michel Orengo)
Date: Wed, 26 Jan 2000 15:49:49 GMT
Subject: MAPI and NT Service
References: <4D0A23B3F74DD111ACCD00805F31D8101D8BCC73@RED-MSG-50>
Message-ID: <388F1893.879D0550@netscape.com>

Thanks!
I wrote my program using CDO, when using the session.logon I have a problem.
I used the MAPIInitialize as you indicated and then the MAPILogonEx.
To reuse what I wrote before (using CDO), I wrote:

session = win32com.client.Dispatch("MAPI.Session")
session.MAPIOBJECT = rawSession

The last line should map my session object with the MAPI handle.

BUT when I do:

outbox = session.Outbox

my NT service fails.

I tried it in Interactive mode and it worked (though not with the same flags
for the MAPIInitialize)

Do you have a clue why?

Again thanks for the help.
Michel




From news at dorb.com  Mon Jan  3 16:00:32 2000
From: news at dorb.com (Darrell)
Date: Mon, 3 Jan 2000 16:00:32 -0500
Subject: JPython Projects
References: <386D0703.EB0807AA@prescod.net> <38710CE8.D67E36CD@inforonics.com>
Message-ID: <n38c4.23$hN5.1968@typhoon.nyroc.rr.com>

http://sourceware.cygnus.com/java/

--
--Darrell
"Jeff Macdonald" <jeff.macdonald at inforonics.com> wrote in message
> GJC?
>
>
> Paul Prescod wrote:
> >
> > Here are some interesting JPython projects for someone with a time on
> > their hands:
> >
> >  1. Use the new GJC to compile JPython and CPython into the same binary.





From parkw at better.net  Tue Jan  4 13:39:49 2000
From: parkw at better.net (William Park)
Date: Tue, 4 Jan 2000 13:39:49 -0500
Subject: OT: OpenSource Python Books?
In-Reply-To: <mt2.0-29451-946955341@news.informatik.uni-bonn.de>
References: <mt2.0-29451-946955341@news.informatik.uni-bonn.de>
Message-ID: <20000104133949.A477@better.net>

I think toExcel.com is already publishing Library Reference, Language
Reference, and Tutorial.


On Sun, Feb 10, 2036 at 09:37:17AM +0000, poet at linuxports.com wrote:
> Hello,
> 
> I represent LinuxPorts.Com (http://www.linuxports.com/) and I have been
> approached by a printer to possibly publish OpenSource books. To start
> these books would just be paper versions of the actual documentation.
> In this case, the Python Documentation and we were wondering if this
> would be of interest to the Python Community.
> 
> I spoke with the printer and we both have agreed that if we do this
> that we will be providing a portion of the Gross Profits back to the
> individual projects. In this case the Python project.
> 
> Anyway, if this of interest to you please go to
> http://www.linuxports.com/ and make your vote. It will help us
> determine the demand for such a book.
> 
> Thank you for your time,
> 
> Joshua Drake
> Webmaster Linux Documentation Project
> Owner LinuxPorts.Com
> poet.nospam at linuxports.com
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> 
> -- 
> ----------- 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
> ------------------------------------------------------------
> -- 
> http://www.python.org/mailman/listinfo/python-list



From phil at ricochet.net  Tue Jan 18 03:57:30 2000
From: phil at ricochet.net (phil at ricochet.net)
Date: Tue, 18 Jan 2000 08:57:30 GMT
Subject: PythonWin and Dialogs in DLLs
References: <38839d87$0$22473@senator-bedfellow.mit.edu>
Message-ID: <388418ec.8757803@news.ricochet.net>

On Mon, 17 Jan 2000 17:52:04 -0500, "Robert" <soundhack at hotmail.com>
wrote:

>Hi all,
>    After getting bored creating dialogs by making dialog templates
>programmatically,
>I would instead like to use VC++'s dialog editor to format the dialog.
>Looking at the dialog.Dialog class in PythonWin, apparently
>it accepts a DLL filename to get resource information.
>
>   My question is this: How can I make this DLL that contains
>only the dialog resource? Using the appwizards to make a DLL
>project, I either get a big DLL file (which works, but it has DLLMain() and
>run time libraries linked in that i dont really need) or an error linking,
>saying there is an unresolved external
>reference _DllMainCRTStartup. The ideal situation would be just
>to save the resource file and have pythonwin read that, but if i
>have to use DLLs, I would like to keep unnecessary runtime stuff out of it.

I believe you should be able to do the following:

1. Use the studio wiz to create a win32 dll.

2.  Add a file called say frog.c containing something like
#include <windows.h>
BOOL WINAPI DllMain( HINSTANCE hInstance, DWORD dwReason, LPVOID frog)
{
    return 1;
}
i think the DLL loader needs an entry point called DllMain and the
parameters will allow you to check for stuff, but i don't know why
that would be useful; anyway it doesn't cost much, and using a
resource DLL is a fairly standard thing.

3. Go to project settings/link/input and remove everything from the
object/library modules, you probably don't need any of them.

4. Go to build/set active configuration and select win32 release.
(This minimises your disk image and also improves memory footprint,
and there's nothing to debug).

5. build. This should result in about 30K or so of disk image (depends
on your cluster size or whatever it's called these days). If this is
too big then there are other things you can do, but...

6. Add the resource .rc. The DLL should grow roughly by the size of
your resources. I think that's it.

phil.

The world is divided into two sorts of people:
Those that believe the world is divided into
two sorts of people and those that don't.


From aahz at netcom.com  Fri Jan 21 17:07:19 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 22:07:19 GMT
Subject: comp.lang.python.announce working??
References: <01bf643a$5039a8a0$3e41c3d1@Cyberus.cyberus.ca>
Message-ID: <86alan$hdn$1@nntp9.atl.mindspring.net>

In article <01bf643a$5039a8a0$3e41c3d1 at Cyberus.cyberus.ca>,
Gordon Williams <g_will at cyberus.ca> wrote:
>
>I havent had any news feed from comp.lang.python.announce since the end of
>December.  I contacted my ISP and the news came back from the group that
>supplies the news feed that there havent been any new postings to this
>group since this time.  I find that strange.

There were none until a couple of days ago; a bunch got posted recently.
Bursty behavior is normal for that group.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From Sven.Drescher at dlr.de  Fri Jan 28 05:28:08 2000
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Fri, 28 Jan 2000 11:28:08 +0100
Subject: problems with blocking pipes, threading
Message-ID: <86rqr7$gn2$1@news.go.dlr.de>

Hallo,

I'm on the way to implement a process manager to start and finish specific
processes. I do that with fork and execv under Solaris and with Mark
Hammond's win32 extensions under Windows NT. All starts and finishs correct.
But if I try to redirect the childs stdout with a pipe and read it out
parallel from parent, the read-function blocks until the child is killed.
Here is a short example of my test programm. I don't know what's wrong. I
tried two threads, but still blocking.

The threading modul seems also to be not correct. Threads on my Sun don't
work parallel. Has anybody also such problems or a solution?

The example programm is ones from Mark Hammond's example. I modified it.
The command is only a little program which does a few outputs and delays.

Thanks for hints and help. Can it be, that an other way is more effective
for my problem???

Sven

example:
'''runproc.py

start a process with three inherited pipes.
Try to write to and read from those.
'''

import win32api
import win32pipe
import win32file
import win32process
import win32security
import win32con
import msvcrt
import os

import threading

class Process:
    def run(self, cmdline):
        # security attributes for pipes
        sAttrs = win32security.SECURITY_ATTRIBUTES()
        sAttrs.bInheritHandle = 1

         self.cmdline=cmdline

        # create pipes
        self.hStdout_r, self.hStdout_w = win32pipe.CreatePipe(sAttrs, 0)

        # set the info structure for the new process.
        self.StartupInfo = win32process.STARTUPINFO()
        self.StartupInfo.hStdOutput = self.hStdout_w
        self.StartupInfo.dwFlags = win32process.STARTF_USESTDHANDLES

        # Create new output read handles and the input write handle. Set
        # the inheritance properties to FALSE. Otherwise, the child inherits
        # the these handles; resulting in non-closeable handles to the pipes
        # being created.
        pid = win32api.GetCurrentProcess()

        tmp = win32api.DuplicateHandle(
            pid,
            self.hStdout_r,
            pid,
            0,
            0,     # non-inheritable!
            win32con.DUPLICATE_SAME_ACCESS)
        # Close the inhertible version of the handle
        win32file.CloseHandle(self.hStdout_r)
        self.hStdout_r = tmp

        self.stdout = os.fdopen(msvcrt.open_osfhandle(self.hStdout_r, 0),
"rb", 0)
        threading.Thread(None, self.ausgabe).start()
        threading.Thread(None, self.start).start()

    def start(self):
        # start the process.
        hProcess, hThread, dwPid, dwTid = win32process.CreateProcess(
                None,   # program
                self.cmdline,# command line
                None,   # process security attributes
                None,   # thread attributes
                1,      # inherit handles, or USESTDHANDLES won't work.
                        # creation flags. Don't access the console.
                0,      # Don't need anything here.
                        # If you're in a GUI app, you should use
                        # CREATE_NEW_CONSOLE here, or any subprocesses
                        # might fall victim to the problem described in:
                        # KB article: Q156755, cmd.exe requires
                        # an NT console in order to perform redirection..
                None,   # no new environment
                None,   # current directory (stay where we are)
                self.StartupInfo)
        # normally, we would save the pid etc. here...

        # Child is launched. Close the parents copy of those pipe handles
        # that only the child should have open.
        # You need to make sure that no handles to the write end of the
        # output pipe are maintained in this process or else the pipe will
        # not close when the child process exits and the ReadFile will hang.
        win32file.CloseHandle(self.hStdout_w)

    def ausgabe(self):
        import select
        while 1:
              try:
                  #select.select([self.stdout.fileno()], [], [])
                   print "Read on stdout: ", self.stdout.read(1)
               except:
                   print 'error'

if __name__ == '__main__':
    p = Process()
    exe = win32api.GetModuleFileName(0)
    p.run(exe + ' run_task.py')

# end of runproc.py


--
______________________________________
German Aerospace Research Establishment
e-mail: Sven.Drescher at dlr.de





From digitome at iol.ie  Fri Jan 21 14:18:33 2000
From: digitome at iol.ie (Sean Mc Grath)
Date: Fri, 21 Jan 2000 19:18:33 GMT
Subject: WAP with Python
References: <20000121120558.50068.qmail@hotmail.com>
Message-ID: <3888ab18.1241665@news.iol.ie>

On Fri, 21 Jan 2000 13:05:58 CET, "Ulf Engstr?m" <b2blink at hotmail.com>
wrote:

>Hello..:)
>
>Has anyone tried to do any WAP-things with Python? I'm trying to set up an 
>apache webserver on a windows2000-machine and get it to run with the 
>wap-software. This havn't been all that successful this far, but I'm working 
>on it. The python-part of it seems to be working, since I can see it with my 
>regular browser.
>Anyone has any experience in this area and can give me some hints?
>Brgrds

I use WindowsNT + Apache + Nokia WAP Gateway + Toolkit (Phone
emulator). Works very well.

I cannot say too much just now but if you are interested in WAP stuff
you might want to take a look at http://www.propylon.com.
The majority of the software talked about is written in Python.

Client side software will be made freely available just as soon as
we can . A server side toolkit for WAP application development
(in Python of course) will follow later.

regards,



From gmcm at hypernet.com  Sat Jan  8 12:36:47 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 8 Jan 2000 12:36:47 -0500
Subject: control structures (was "Re: Sins")
In-Reply-To: <38774EBA.40716D1E@maxtal.com.au>
Message-ID: <1264769486-16171098@hypernet.com>

John Skaller wrote:
[snip]
>  Current looping constructions are so bad,
> that the simplest of uses cannot be expressed
> using them : searching a list for an element
> cannot be written correctly:
> 
>  for i in X: 
>   if X[i]=e: goto Found
>  print "Not Found"
>  goto next
>  Found: ...
>  next
> 
> YUKKKKKK!!! There is no natural way to do this.
> We fall through the loop on failure, instead
> of success. (using a flag, or testing anything
> afterwards, is not acceptable as a solution).

That's the reason that Python has the "for 
.. else" construct that William Tanksley 
hates so much.

 for i in X: 
   if X[i]==e: 
     # do the "found" thing
     break
 else:
   print "Not Found"
 
responses-redirected-to-Tanksley-ly y'rs


- Gordon



From aahz at netcom.com  Sat Jan  8 10:05:18 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jan 2000 15:05:18 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com> <38774EBA.40716D1E@maxtal.com.au>
Message-ID: <857jne$kcb$1@nntp8.atl.mindspring.net>

In article <38774EBA.40716D1E at maxtal.com.au>,
skaller  <skaller at maxtal.com.au> wrote:
>
>	Current looping constructions are so bad,
>that the simplest of uses cannot be expressed
>using them : searching a list for an element
>cannot be written correctly:
>
>	for i in X: 
>		if X[i]=e: goto Found
>	print "Not Found"
>	goto next
>	Found: ...
>	next
>
>YUKKKKKK!!! There is no natural way to do this.
>We fall through the loop on failure, instead
>of success. (using a flag, or testing anything
>afterwards, is not acceptable as a solution).

Why is using a flag or testing afterwards not acceptable?
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From scherbi at bam.com  Wed Jan 12 07:54:47 2000
From: scherbi at bam.com (Bill Scherer)
Date: Wed, 12 Jan 2000 07:54:47 -0500
Subject: Python and Oracle
References: <dB_e4.1177$hJ1.22100@news.rdc1.tx.home.com>
Message-ID: <387C7997.623DA440@bam.com>

Paul -

I have succesfully compiled DCOracle on Linux (2.2.x)and Solaris(2.6) against
Python 1.5.x and Oracle 8.0.(4,5).
Succesful compilation with different Oracle versions on different os's
generally requires a not small effort in tweaking the ORACLE_LIBS line in the
makefile.

As for your curiosity of the worth of the effort, it's worth it if you want
to connect to Oracle from Python.  It works very well, and outside of win32,
there are few other options for connecting to Oracle...

Bill Scherer



"Paul E. Bible" wrote:

> Has anyone had success installing the Python Oracle Call Interface (OCI)
> library written by Digital Creations (DCOracle)?  I'm currently having
> compile problems with the C modules, which is a library reference.  I just
> curious as to whether this library is worth the effort to resolve the
> current issues that I'm having with it's installation.
>
> My platform is:
> Python           1.5.2
> Oracle           7.3.4
> OS                Unixware 7.0.1
>
> Thanks in advance for your help,
> Paul
>
> --
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000112/5ce000aa/attachment.vcf>

From xtrm at karoo.co.uk  Tue Jan 11 17:32:02 2000
From: xtrm at karoo.co.uk (xtrm)
Date: Tue, 11 Jan 2000 22:32:02 -0000
Subject: Python Host
References: <bJ3e4.8$78.764@news.uswest.net>
Message-ID: <85gb27$6ab$1@supernews.com>

Andrew Ferk <shawnball at uswest.net> wrote in message
news:bJ3e4.8$78.764 at news.uswest.net...
> Are there any web hosts that support Python?
>
Pair Networks, www.pair.com plus lots of other useful things - perl, php3,
gcc, mysql, SSI, SSL etc. (see also www.support.pair.com)
I've had accounts with them since soon after they started in 1996 and have
yet to have any significant problem!




From nascheme at enme.ucalgary.ca  Sun Jan  9 18:20:03 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Sun, 09 Jan 2000 23:20:03 GMT
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de>
Message-ID: <slrn87i5un.fc5.nascheme@adler.dynodns.net>

Helge Hess <helge.hess at mdlink.de> wrote:
>Hi there,
>
>recently I was heavily wondering why __init__ does not allow return
>values ! Is there any special reason not to do this ?

I would say that it is conceptually wrong.  By the time the
__init__ method is called a new instance has already been
created.  You are proposing that if __init__ returns a value then
this instance should be thrown away.  Any initalizations you do
within __init__ would also be discarded.

A factory function achieves what you want and can do other cool
things too.  If you don't want to use a factory function then
a meta class is probably the best solution (although my head
explodes everything I try to figure them out).

>For example consider this
>  
>  C_singleton = None
>  
>  class C:
>    def __init__(self):
>      if C_singleton is None:
>        c_singleton = self
>      return C_singleton
>
>  c1 = C()
>  c2 = C()
>  'if c1 is c2' will be true


    class Singleton:
        def __init__(self, klass):
            self.klass = klass
            self.instance = None

        def __call__(self, *args, **kw):
            if self.instance is None:
                self.instance = apply(self.klass, args, kw)
            return self.instance

    class RealC:
        pass

    C = Singleton(RealC)

If renaming the class C to something else causes you problems
then you should probably rethink your design.


    Neil

--
"God, root, what is difference?" - Pitr| nascheme at enme.ucalgary.ca
"God is more forgiving." - Dave Aronson| http://www.ucalgary.ca/~nascheme


From grant at nowhere.  Wed Jan 19 14:28:29 2000
From: grant at nowhere. (Grant Edwards)
Date: Wed, 19 Jan 2000 19:28:29 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com>
Message-ID: <slrn88c42t.53m.grant@grante.comtrol.com>

In article <3885D3E0.4F10F28D at netscape.com>, Michel Orengo wrote:

>If you do not have access to the MSDN Library, you can use a
>COM browser: it will give you the properties/methods of the
>objects. Another solution is to make Python aware of the CDO
>object by using makepy (see QuickStartClientCom.html) In the
>file generated by makepy, you will find all the classes and
>attributes.

OK, I did that with msoutlk85.olb.  I'm assuming that a message
object is of type _MailItem.  I've got a big list of properties
and methods that _might_ work, but how do you find out which
ones are implimented by a particular object?

For example, the messages in my inbox implement properties
"Subject" and "Size", but not "Body", "To", or "SenderName".
I guess I could iterate the the properties dictionary and try
all of them...

-- 
Grant Edwards                   grante             Yow!  I want you to
                                  at               MEMORIZE the collected
                               visi.com            poems of EDNA ST VINCENT
                                                   MILLAY... BACKWARDS!!


From seanb at home.com  Fri Jan 21 12:56:03 2000
From: seanb at home.com (seanb at home.com)
Date: Fri, 21 Jan 2000 09:56:03 -0800 (PST)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <20000121135451.B2265@stopcontact.palga.uucp>
Message-ID: <200001211756.JAA18907@seanb.sttln1.wa.home.com>

On 21 Jan, Gerrit Holl wrote:
> Stefan Schwarzer wrote on 948334133:
>> Perhaps the ".." operator is something to consider (if it isn't
>> incompatible with somethings else in the language), but most of
>> the times "range" does it's job very well.
> 
> Unfortunately, not yet. It's quite hard to get a string which
> would be a..k in Perl. Maybe a nice feature for range()?
>>>> string.join(map(lambda i: chr(i), range(ord('a'), ord('k'))), '')
> 'abcdefghij'
> 
> Doesn't look very well, does it?
> 
> regards,
> Gerrit.
> 
As an aside, you do know that that lambda is completely unncecessarry,
right?  The following code is equivalent and (a little) simpler:

>>import string
>>string.join(map(chr, range(ord('a'), ord('k'))), '')

Personally, I would prefer to abstract this into a reusable function:

>>>def srange(start_char, end_char, step=1)
...    '''Create a string range from two given characters.'''
...    start, end = map(ord, (start_char, end_char))
...    chars = map(chr, range(start, end, step))
...    return string.join(chars, '')

-- 
Sean Blakey
(206)297-7123
quine = ['print "quine =",quine,"; exec(quine[0])"']; exec(quine[0])




From fredrik at pythonware.com  Thu Jan 20 03:29:35 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 20 Jan 2000 09:29:35 +0100
Subject: Python Rocks! - ASP problem?
References: <860ftk$bas@news.or.intel.com> <38858F25.84946F4B@bogusaddress.com>
Message-ID: <010101bf6320$7cdb3fb0$f29b12c2@secret.pythonware.com>

Joe Smith <JoeSmith at bogusaddress.com> wrote:
> This conditionally put the html code into the page.  If I remember right,
> python can't do this in ASP because an end marker is needed.  I would say that
> python probably was not designed with ASP in mind.

did ASP really exist back in the late eighties?

</F>




From piet at cs.uu.nl  Sat Jan  8 07:32:46 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 08 Jan 2000 13:32:46 +0100
Subject: bsddb item size?
References: <mrE03AA5nOd4Ew2R@jessikat.demon.co.uk> <wzd7rd4kcd.fsf@sunshine.cs.uu.nl> <1y9ghGAe1nd4Ewk9@jessikat.demon.co.uk>
Message-ID: <wzpuvcn3pt.fsf@sunshine.cs.uu.nl>

>>>>> Robin Becker <robin at jessikat.demon.co.uk> (RB) writes:

RB> thanks that's a great help. How about collisions are they possible?

yes, you can choose to allow duplicate keys.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From thamelry at vub.ac.be  Mon Jan 24 04:19:21 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 24 Jan 2000 09:19:21 GMT
Subject: tabs do WHAT?
References: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net>
Message-ID: <86h5ep$iq$1@mach.vub.ac.be>

Warren Postma <wpostma at earthlink.net> wrote:
[snip]

: If Guido reads this, please note my personal feature request for Python 2.0:
: Make the "whitespace" stuff optional, for people who like weird syntactical
: features only. I'll love you forever. Thanks.

This will not happen. People here (including Guido Himself) get extremely picky 
when the use of indentation is criticised. They do not realize that this feature
is an important obstacle for many people (and since these people have stopped 
using python they are obviously not posting here). Indentation is great, but should 
not be part of the syntax of a language.  

Cheers,

---
Thomas Hamelryck
Institute of Molecular and Structural Biology
Aarhus University
Gustav Wieds Vej 10C
DK-8000 Aarhus C
Denmark                                                                                            



From pinard at iro.umontreal.ca  Thu Jan 20 17:30:07 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 20 Jan 2000 17:30:07 -0500
Subject: Matching a constant string at beginning
In-Reply-To: "Fred L. Drake, Jr."'s message of "Thu, 23 Dec 1999 08:59:01 -0500 (EST)"
References: <oqwvq6wgzq.fsf@titan.progiciels-bpi.ca> <14434.10917.222100.686450@weyr.cnri.reston.va.us>
Message-ID: <oq1z7cgyvk.fsf@titan.progiciels-bpi.ca>

"Fred L. Drake, Jr." <fdrake at acm.org> ?crit:

> Fran?ois Pinard writes:
>  > is rather tedious.  Of course, I could write a very small function to
>  > match a constant string at the beginning of another, but there just must
>  > be some idiom for doing this.

> Fran?ois, Another possibility, if you're willing to use the CVS version(!),
> is to use the string methods:

> 	s = some string...
>         if s.startswith("Simpsons"):
>             do something interesting...

> Or you could write that annoying little function while waiting for 1.6.  ;)

I think I'll do that (write that annoying little function).

"Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> writes:

> Two new methods `startswith' and `endswith' have been added, which will
> fit the bill perfectly: String methods will be one of those new features
> that'll change your life.  :) Wait'll you try s.join().

> -------------------- snip snip --------------------
> >>> ' & '.join(['Guido', 'Tim', 'Gordon', 'David', 'Biff', 'etc.'])
> 'Guido & Tim & Gordon & David & Biff & etc.'
> >>> 
> -------------------- snip snip --------------------

How strange! :-)  I'm curious to see 1.6, indeed...

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





From shayman at uniserve.com  Sun Jan 30 00:40:58 2000
From: shayman at uniserve.com (Mattew)
Date: Sat, 29 Jan 2000 21:40:58 -0800
Subject: GUI
Message-ID: <949210537.666866@neptune.uniserve.ca>

Will Python ever include a GUI in the source like Tcl. Would it give any
noticable speed increase. Could current licensing allow for example Tkinter
to be embedded in the source. ??         Besides the good work in Pysol is
there any basic Pacman type   move an object with arrow keys type Games to
look at .

Making some headway in Python  lately.            Going to take a look at
more Multi Media     type programming with
Director and Lingo .          Any comments  on  Lingo       appreciated

Fergus






From soundhack at hotmail.com  Wed Jan 12 14:52:15 2000
From: soundhack at hotmail.com (Robert)
Date: Wed, 12 Jan 2000 14:52:15 -0500
Subject: PIL 1.0 bugs? (TIFF, Image)
References: <387c6eb2$0$22474@senator-bedfellow.mit.edu> <001401bf5cf9$7a8a9da0$f29b12c2@secret.pythonware.com>
Message-ID: <387cdb1c$0$22473@senator-bedfellow.mit.edu>

Thanks! I removed all traces of PIL and reinstalled and the show() problem
went away.
However the other problem was still there. I was mistaken in the tag name
tho,
it was SAMPLEFORMAT instead of SOURCEFORMAT.

Robert

Fredrik Lundh wrote in message
<001401bf5cf9$7a8a9da0$f29b12c2 at secret.pythonware.com>...
sounds like you're using different versions of
the PIL Python library and the _imaging module.

since there is no thing called SOURCEFORMAT
in the 1.0 release, you're probably using old
python files with a new _imaging module.

better check your installation.






From wtanksle at hawking.armored.net  Sat Jan 29 01:26:11 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 29 Jan 2000 06:26:11 GMT
Subject: statements and blocks
References: <86t0ov$4en$1@news.psc.edu>
Message-ID: <slrn8952ne.gfa.wtanksle@hawking.armored.net>

On 28 Jan 2000 21:13:03 GMT, John Kochmar wrote:
>OK, I haven't seen this in the FAQ (Yeah, I did look ;^), and maybe I 
>just missed it, but someone at the Python conference (sorry, I forget
>which talk) brought up that you couldn't do:

>	print "a"
>	print "b"
>		print "c"
>		print "d"
>		print "e"
>	print "f"

>to visually set off a block of statements.  And ahort of forcing a different 
>scope though a conditional, loop, or some other scope modifier, I haven't 
>been able to create a block of statements in a seperate context.

One way to do that is to use a try: block.  Since many of the cases where
you have a block like that have special entry and exit code, you can
enclose the exit code in a finally: statement.

Oh, and the reason why you won't see any bare-block syntax in Python: that
would make it impossible to detect accidental indentation change.  That,
in turn, would make it harder to select the most virulent
whitespace-eating nanoviruses ;-).

>John

-- 
-William "Billy" Tanksley, in hoc signo hack


From skip at mojam.com  Wed Jan 26 21:17:18 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 26 Jan 2000 20:17:18 -0600 (CST)
Subject: Interesting article on programming languages
In-Reply-To: <slrn88uvvp.935.thantos@brimstone.mecha>
References: <388E7D5B.7840983F@erols.com>
	<slrn88uvvp.935.thantos@brimstone.mecha>
Message-ID: <14479.43694.193271.455972@beluga.mojam.com>

    Alexander> Interestingly, he discards the entire class of functional
    Alexander> programming languages within the first two paragraphs, then
    Alexander> summarises his article with the 'needed facilities' the
    Alexander> industry will need in the next decade ... which, perversely
    Alexander> enough, /already exist and have been used successfully for
    Alexander> years/ in the first group of languages he threw out.

I find nothing particularly troublesome about this aspect of the article.
There needs to be research before ideas can be put into practical usage.
How would we know we needed frameworks if someone hadn't done the research
and demonstrated its usefulness?  He may have ignored Haskell, et al as
research languages.  That doesn't mean he didn't consider the novel ideas
those languages explore useful, just that he didn't think they have the
proper set of non-research features necessary to make them commercially
appealing at this time.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From paul at prescod.net  Sun Jan 23 17:49:05 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jan 2000 16:49:05 -0600
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p>
Message-ID: <388B8561.F049E960@prescod.net>

Edward Welbourne wrote:
> 
> ...
> 
> So, yes, this is a matter of taste: the two-namespace rule is an emetic
> to some, to whom I would commend Algol as a language to use instead of
> python; but to pythoneers, the two-namespace rule is harmonious,
> beautiful and kind to the garbage-collector.  Treat the humblest of thy
> servants kindly, lest they rise up and smite thee with just zeal.

"To some, our current policy towards health care is a immoral and to
those I would suggest Cuba, or perhaps Canada. But to Americans, the
health care if you can afford it policy is harmonius, beautiful and kind
to our budget."

I'm having trouble with your last sentence but then I didn't quite
understand its significance in the original context. Nevertheless I like
it. Do you have a reference for it?

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Earth will soon support only survivor species -- dandelions, roaches, 
lizards, thistles, crows, rats. Not to mention 10 billion humans.
	- Planet of the Weeds, Harper's Magazine, October 1998



From embed at geocities.com  Tue Jan 25 10:23:32 2000
From: embed at geocities.com (Warren Postma)
Date: Tue, 25 Jan 2000 10:23:32 -0500
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <948807929.472741@news.adtranzsig.de>
Message-ID: <Thjj4.6690$S64.195053@nnrp1.uunet.ca>

"Dirk-Ulrich Heise" <hei at adtranzsig.de> wrote in message
news:948807929.472741 at news.adtranzsig.de...
> Again, have a try to convince the C++
> people to use "begin" and "end". If C++ is so evolvable, it should be
> easy going, shouldn't it?

Rather bloody minded of me to point this out, but hey "to each his own"...

#define BEGIN {
#define END     }

Can I do the same somehow in Python?  ;-)

    Warren
    Ministry of Silly Blocks






From ullrich at math.okstate.edu  Sat Jan 22 16:29:53 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Sat, 22 Jan 2000 15:29:53 -0600
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <001a01bf632d$393fbda0$3acbd9c2@peridot.optichrome.com>
Message-ID: <388A2150.D965E4F@math.okstate.edu>


Adrian Eyre wrote:

> > I like the following syntax for readability, and feel the this should be
> > implimented in Python:
> >
> >       sub     ax,bx           # Lets keep the python comment symbol
> >       jne     Mylabel         # We also need labels
> >         rtn                   # Notice all commands are three chars
> > Mylabel
> >       add     00h,ax
> >
>
> I feel this may be too confusing. I'd prefer a syntax like this:
>
> 1000101001011101010010101001010100010010010001000100100101000101
> 1011000001100100100010010010001010011101001001010100100101010010
>
> etc.

    This is the second place I've seen the "etc." operator in this thread.
I can't find it in the docs - what gives?

>
> --------------------------------------------
> Adrian Eyre <a.eyre at optichrome.com>
> http://www.optichrome.com



From skip at mojam.com  Tue Jan  4 13:35:14 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 4 Jan 2000 12:35:14 -0600 (CST)
Subject: NetWare port of Python - not so hot pystone
In-Reply-To: <200001032133.QAA08724@anvil.murkworks.com>
References: <200001032133.QAA08724@anvil.murkworks.com>
Message-ID: <14450.15714.214801.462472@beluga.mojam.com>

    Brad> Still, these numbers are rather crummy. What kinds of things in a
    Brad> CLIB could have a large impact on Python?

    Brad> 1. memory allocation routines
    Brad> 2. string manipulation
    Brad> 3. <other>???

    Brad> I suspect that the memory allocation functions in clib aren't very
    Brad> good, and could be replaced with something better. Does anyone
    Brad> know of a free source of a very good dynamic memory allocator?

Look around the Starship or c.l.py archives for pymalloc.  Also, the GNU
malloc is generally pretty good.

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 bruce_dodson.nospam at bigfoot.com  Wed Jan 12 23:23:56 2000
From: bruce_dodson.nospam at bigfoot.com (Bruce Dodson)
Date: Thu, 13 Jan 2000 00:23:56 -0400
Subject: Python and MapObjects (Was: Python and ArcView GIS)
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca> <Pine.SGI.3.95.1000112172541.19328B-100000@clavey.des.ucdavis.edu>
Message-ID: <Ttcf4.16754$2x3.302645@newscontent-01.sprint.ca>

I have tried that too.  It works fine if you only use it to script the
automation components, e.g. if you just use MO as a way to query and access
spatial data.  But I ran into trouble when I tried to instantiate the MO
control.  I was able to show a map in a window, using PythonWin's MFC-ish
classes, and it worked OK... but when PythonWin closed, Dr. Watson would
complain about access violations or something.  (That is on Windows NT.  On
my home machine, running Windows 95b, it closes without errors.)  I don't
know enough about PythonWin or MFC to track this down.

The other thing with MO is that, since it is a licensed control, it might
not deploy correctly.  I'm not sure how licensing works in COM, but I think
some keycode ends up embedded as a resource within the EXE.  Since a Python
program doesn't have that license information embedded in it, MO might
refuse to load.  I have not tested this; the problem might not even exist.

Attached is a very quick hack of the IE4 example for ActiveX.  Make a few
minor changes in Demo() to point it at a shapefile that exists on your
system, run it from pythonwin, and with any luck an MDI child will pop up
with a map in it.

Bruce

Allan Hollander wrote in message ...
>The ArcView extension sounds very interesting.
>
>Along these lines of interfaces to ESRI software, has anybody done any
>work with getting Python to talk to the MapObjects set of components?  I'd
>love to have some pointers on how to do this.  Right now I'm stuck on
>how to create/access/etc. the Map ActiveX control, which seems basic to
>getting much farther with this.
>
>desperately-trying-to-avoid-visual-basic y'rs --
>
> -- allan
>
>
>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: MapWindow.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20000113/08acc5de/attachment.ksh>

From wlfraed at ix.netcom.com  Tue Jan 18 12:32:29 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Tue, 18 Jan 2000 09:32:29 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <8610or$nsu$1@news.udel.edu> <861jm5$nrj$1@mach.vub.ac.be>
Message-ID: <zqGEOEqNnxHC8gbhL5piinIM1bGb@4ax.com>

On 18 Jan 2000 11:44:05 GMT, Thomas Hamelryck <thamelry at vub.ac.be>
declaimed the following in comp.lang.python:


> 
> Not necessarily. I've been using python for about a year or so and I still
> dislike indentation. Indentation can make your code less readable and leads to

	Better tell that to the Software Practices department at my
place of work then... All the style guides heavily push indentation and
modularization for coding -- even if the language needs begin/end block
statements.

> mysterious errors. I'm sure the strict use of indentation keeps a lot of potential users

	Mysterious errors usually means a mistake in the indentation
applied. And to me, the indentation is easier to pick out then an
"endif" on the wrong line...

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From steve at acer.spvi.com  Sat Jan  8 21:22:51 2000
From: steve at acer.spvi.com (Steve Spicklemire)
Date: Sat, 8 Jan 2000 21:22:51 -0500 (EST)
Subject: pipe problem on FreeBSD....
In-Reply-To: <200001090211.VAA13772@acer.spvi.com> (message from Steve
	Spicklemire on Sat, 8 Jan 2000 21:11:11 -0500 (EST))
References: <200001082039.PAA12903@acer.spvi.com> <858iu6$m75$1@news.kolumbus.fi> <200001090211.VAA13772@acer.spvi.com>
Message-ID: <200001090222.VAA13839@acer.spvi.com>

I suppose I should add that I'm really after the Zope problem. I'm
assuming that these two things are related... but I really don't know
that. I'm curious to know how this is 'supposed' to work and any hints
anyone might have about sorting out the problem.

thanks (again),
-steve

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

Thanks Jarkko! 

       OK... I think I see now how my program was flawed. However
there is still something fishy..... 

[blah blah blah......]




From robin at jessikat.demon.co.uk  Mon Jan 17 14:13:44 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 17 Jan 2000 19:13:44 +0000
Subject: using threads with python and winnt
References: <1264075593-4560755@hypernet.com> <38835f4c.1380925@news.iol.ie>
Message-ID: <mDN0tCAon2g4Ewcx@jessikat.demon.co.uk>

In article <38835f4c.1380925 at news.iol.ie>, Sean Mc Grath
<digitome at iol.ie> writes
>
>>Kreiner Karl or Charly or somebody wrote:
>>
>>>             does anyone have expiriences with using threads and winnt ?
>[...]
><gmcm at hypernet.com> wrote:
>>Threads work dandy on NT
>[...]
>
>Does anybody know how good/bad/indifferent the threading stuff works
>on Windows 95/98? I am considering using threads for an app
>under development but I'm a bit worried (read "terrified") about
>the issues that might pop up in '95 and '98.
>
>regards,
>Sean McGrath
> 
I've been using threaded Zope on NT 4.0 for a while and have a
complicated extension which is thread safe etc etc. No problems at all
with threads so far.
-- 
Robin Becker


From loewis at informatik.hu-berlin.de  Fri Jan 28 15:21:20 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Jan 2000 21:21:20 +0100
Subject: database connectivity
References: <s90ihpl47q511@corp.supernews.com>
Message-ID: <p6qg0vij6bj.fsf@informatik.hu-berlin.de>

"David Metzler" <davidm at softdesigns.com> writes:

> I love what I have seen so far of Python. I am in serious need of
> guidance however. I spend a couple of nights till the wee hours and
> have run into a block. What I want to do is create some python
> database programs that will work on Windows 98/NT and on Linux.

You may want to have a look at http://www.python.org/sigs/db-sig/. The
DB SIG is developing a standard API for Python database drivers, and
also collects database-related information. 

There is a number of DB API compliant database modules available, just
pick one that fits your intended database. Ideally, your code will be
portable even if you switch databases. In addition to traditional DB
systems, you could also consider Gadfly, which is a database written
in Python itself.

If you run into specific problems, and can't get a specific response
in this group, you may want to ask on the SIG list as well.

Hope this helps,
Martin


From fredrik at pythonware.com  Wed Jan 12 04:26:27 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 12 Jan 2000 10:26:27 +0100
Subject: Python BUG?
References: <85hg1q$2nli$1@gavrilo.mtu.ru>
Message-ID: <004d01bf5cdf$1b33df70$f29b12c2@secret.pythonware.com>

Oleg Orlov <orlov at diasoft.ru> wrote:
> This script:
> 
> d = [1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L,
>     10L, 9L, 8L, 7L, 6L, 5L, 4L, 3L, 2L, 1L]
> print reduce(lambda x, y: x*x + y*y, d)
> 
> hangs up my python interpreter (on NT4).
> 
> Is it a bug or my mistake?

neither.  it's just that your computer is not quite
as fast as you think it is...

to see what I mean, replace the lambda with a
function that prints the intermediate result, or
some variation of it.  e.g.

def f(x, y):
    v = x*x + y*y
    print len(str(v))
    return v

print reduce(f, d)

</F>




From kc5tja at armored.net  Wed Jan 19 21:25:26 2000
From: kc5tja at armored.net (Samuel A. Falvo II)
Date: Wed, 19 Jan 2000 18:25:26 -0800
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com>
Message-ID: <38866cbc@fermi.armored.net>

> I suggest you go to other newsgroups and let them know how brilliant
Python
> is because it has an invisible block terminator.

So you're suggesting that we spam other newsgroups?  That's brilliant
advice.  It takes a real rocket scientist to think of this one.  That's just
like saying, "Hey, Mr. President!  Why don't we waltz on over to China and
brag about how many tactical nukes we've got!"  Yeah, real smart.  People
like you are the reason this planet has world wars, you know that?

> Since we are so lazy let's just skip the ending anything and just use
> indentation.

Maybe its this way for readability purposes, and not laziness?  In fact, I
find it makes much MORE effort on my part to use indentation to arrange
Python code than in C.  Yet I prefer Python's indent-oriented block
structure over C.  This is coming from a 12 year C programmer.

> Math:
> Before:
>     (2 + 3) * 5
>
> After Python rule:
>     (2 + 3
>             * 5

No.  Before is as you describe above.  Applying Python's indentation rules
strictly would be as follows:

(
   2+3
*5

> Before:
>     if (x == 5)
>     {
>         printf("blah\n");
>     printf("blah blah");

I'm sorry, you loose.  Unclosed brace error.  Please try again.

> Go ahead and convince these guys to switch do it the "Python Way".
> And Good luck. You will need it.

Why would we want to?  What makes you think that we have the right to
impress our views on other software developers?  Likewise, what gives you
the God-given right to impress your beliefs on us?  Calling us lazy and
attempting to incite people here to start a jihad in other newsgroups?  You
call that professional behavior?  I think not.  We don't NEED to boast about
our language of choice.  The language boasts for itself.

Go ahead and try and convince us to switch our ways.  And good luck.  You
will need it.





From marc_risney at my-deja.com  Sun Jan 30 03:51:09 2000
From: marc_risney at my-deja.com (marc_risney at my-deja.com)
Date: Sun, 30 Jan 2000 08:51:09 GMT
Subject: Announcing Gadfly COM Object
Message-ID: <870u1t$dl4$1@nnrp1.deja.com>

Ok, I just put up 0.51 beta, a little closer, a little smoother,
but still needs some work I need to do a better job with recordset.EOF
and .BOF as far as Looping goes, mebee someone can jump in and help out
here.
I need to clean up a bit and would like to add some sort of DB.help
  Aaron has a whole slew of goodies when
you go into Client/Server mode via gfclient, and it sounds like he is
going to add more with Gadfly 2.0, SHUTDOWN, RESTART, CHECKPOINT,
EXCUTED_PREARED, etc... I need to add the server commands and I will
probably chuck
some VB and Delphi Clients together, and some MS Excel and Access demos
using the object. For now I have  a lowly  VBScript demo...oh well you
get the point


I was inspired by ADO which is incredibly easy to use when scripting
with ASP
and VBScript. but I luv the idea of serving my Gadfly data from a Linux
box
and connecting to it via a port with a pretty VB Form on my NT Box.
Again the idea is to give it similar functionality to ADO, but after
the basics are covered, add the good stuff.

One other thing; a bit on a tangent really, if anyone tells you that
Oracle Lite
is reaally easy, non complicated and small to install is a Liar, and
obviously has never w
orked with Gadfly before, a true multi-platform small footprint RDBMS
soory Larry..yer'
product is sucks

at any rate the  0.51 beta  for  the Gadfly COM  server can be found at
the following: http://www.blazesoft.com/~mrisney/comgadfly.py
and a sample VBScript at
          http://www.blazesoft.com/~mrisney/conn_gadfly.vbs


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


From gmcm at hypernet.com  Wed Jan 19 13:11:14 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 19 Jan 2000 13:11:14 -0500
Subject: DB API
In-Reply-To: <864no9$icc$1@nntp3.atl.mindspring.net>
Message-ID: <1263817020-5143711@hypernet.com>

Aahz wrote:
> Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
> >Aahz Maruch <aahz at netcom.com> wrote:
> >>
> >> Oh, okay.  Let's suppose, for example, that you have a customer
> >> database, and the client wants to view the entire order history for a
> >> customer.  Would you say that it's inappropriate for the database to
> >> return thousands of rows for an active customer?
> >
> >Well, it is if the customer is one of many who expects lightning
> >performance. If the customer wants a lot of data he will have to wait,
> >and he will simultaneously tie up the server for other customers. 
> 
> Uh, if it's a properly constructed database, it shouldn't get tied up
> like that.

Perhaps some bozo code "FOR UPDATE" in the select 
statement? <sound of fingers breaking>. He won't do that 
again.

Actually, if the SQL is coded right (no ORDER BY or GROUP 
BY) and the stuff in the middle is decent, the DB can be 
returning rows before the query has even finished executing. If 
the query is parameterized (pre-parsed) and simple, the 
bottleneck will be the network and the load on the DB will be 
insignificant. Unfortunately, that's a lot of "if"s...
 


- Gordon



From aahz at netcom.com  Thu Jan  6 12:28:20 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 6 Jan 2000 17:28:20 GMT
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de>
Message-ID: <852jbk$9tt$1@nntp9.atl.mindspring.net>

In article <3874CFE5.765A40EA at mdlink.de>,
Helge Hess  <helge.hess at mdlink.de> wrote:
>
>recently I was heavily wondering why __init__ does not allow return
>values ! Is there any special reason not to do this ?

Yup.  The return value *has* to be the object instantiated by Python.
Why do you want to do this?
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From gerrit.holl at pobox.com  Sat Jan 29 08:27:14 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 29 Jan 2000 14:27:14 +0100
Subject: Cutting off trailing zeroes?
In-Reply-To: <3892ad81$1_2@goliath.newsfeeds.com>; from jblazi@netsurf.de on Sat, Jan 29, 2000 at 10:14:10AM +0100
References: <3892ad81$1_2@goliath.newsfeeds.com>
Message-ID: <20000129142714.A2146@stopcontact.palga.uucp>

Janos Blazi wrote on 949137250:
> Id like to format a string like this:
>   s = "(%f)" % x
> 
> Now if x=1.5 then s="1.50000" or something like this. Is there a tirivial
> way of cutting of the zeroes, i.e. getting s="1.5" instead of s="1.50000"?
> 
> Of course it can be programmed very simply. My question is: can I do it with
> specifiers somehow?

s = "(%s)" % str(x)

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From jblazi at netsurf.de  Wed Jan 26 15:52:21 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Wed, 26 Jan 2000 21:52:21 +0100
Subject: To find a point inside a polygon
References: <388F0E36.6CE37135@ust.hk>
Message-ID: <388f5cb1_3@goliath.newsfeeds.com>

You find the algorithm in the book Computational Geometry by
Preparata-Shamos (if I remember correctly the title) and in many other books
on computational geometry. The implementation should be very simple
(actually Python is an extremely friendly language and the method is
simple).

The basic idea is to draw a line through the point and count how often this
line intersects the boundaries of the polygon.

J.B.

David Yeung <dyeung at ust.hk> schrieb in im Newsbeitrag:
388F0E36.6CE37135 at ust.hk...
> Does anyone know if any python implementation for finding whether a point
> is inside a polygon? E.g., by giving a point(x,y) and an array of
polygon's
> points (x0,y0,x1,y1,x2,y2,...), it returns true or false to tell if the
> given (x,y) is inside the polygon.
>
> Thanks
>
> david




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From insyte at emt-p.org  Fri Jan 21 20:46:17 2000
From: insyte at emt-p.org (Ben Beuchler)
Date: Fri, 21 Jan 2000 19:46:17 -0600 (Central Standard Time)
Subject: vi and python?
In-Reply-To: <20000126092430.A20974@better.net>
Message-ID: <Pine.WNT.4.21.0001211945020.-3834393-100000@c166.swdata.com>

On Wed, 26 Jan 2000, William Park wrote:

> 'autocmd' executes its commands when certain condition happens.  Here,
> it executes 'set ...' when entering a new buffer (BufRead) with '*.py'
> suffix.

Cool!  Thanks.  Oddly enough, this was very poorly explained in the O'Reilly.
Now I have to start doing a little modification of my own...

Ben

-- 
Time is an illusion.  Lunchtime doubly so.
		--- Ford Prefect




From roy at popmail.med.nyu.edu  Thu Jan 20 21:17:26 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Thu, 20 Jan 2000 21:17:26 -0500
Subject: regex question
Message-ID: <roy-835F11.21172620012000@netnews.nyu.edu>

This isn't really a python question per-se, but it came up while using 
the re module, so there is a connection :-)

Let's say I've got a pattern "[0-9]", i.e. a 1-digit number.  I want to 
search a string for that pattern, and only return a match if it's found 
preceeded by either a non-digit or the beginning of the string, and 
likewise followed by either a non-digit or the end of the string.  Is 
there a single compact regex I can write to cover that?

For example:

pat = re.compile (pattern)
x = pat.search ("1")
y = pat.search ("12")
z = pat.search ("the number 1, don't you know?")

x and z should be MatchObjects, but y should be None.

The best I can come up with would be something like:

(^[0-9][^0-9])|([^0-9][0-9][^0-9])|([^0-9][0-9]$)|(^[0-9]$)

which is fairly ugly.  This uglyness is exaccerbated by the fact that 
I'm not really looking for 1-digit strings, I'm looking for 12-digit hex 
numbers (i.e. ethernet MAC addresses), where the basic pattern is:

[0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-f
A-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f].


From thomas at xs4all.net  Mon Jan 31 10:07:30 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 31 Jan 2000 16:07:30 +0100
Subject: Case Sensitivity and Learnability
In-Reply-To: <01bf6a99$fafba6c0$LocalHost@dreadzone>; from dheise@debitel.net on Sat, Jan 29, 2000 at 09:48:56PM +0000
References: <B4191059635E81B1.8AFDD386EFACF4A3.6FF161EA0151B5EB@lp.airnews.net> <86u4uf$26lq$1@thoth.cts.com> <01bf6a99$fafba6c0$LocalHost@dreadzone>
Message-ID: <20000131160730.E19725@xs4all.nl>

On Sat, Jan 29, 2000 at 09:48:56PM +0000, Dirk Heise wrote:

> PLEASE DON'T THINK I'M SHOUTING. I'M JUST NOSTALGIC AND
> I WANT MY C64 BACK.

The first thing I used to do on my C64, when I started programming, and the
first thing in the programs I wrote, was changing the case to lowercase. I
was 6 when I started that habit, though, I probably was too dumb to figure
that I was too dumb to use lowercase ;)

(This was all before I sacrificed a couple of pages of RAM to copy the
font-rom and alter it, though ;)

READY-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From fredrik at pythonware.com  Fri Jan 21 10:11:41 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 16:11:41 +0100
Subject: How to commit suicide with a SIGSEGV?
References: <slrn88gs8u.77g.hildeb@stahlw06.stahl.bau.tu-bs.de>
Message-ID: <010801bf6421$d3b97bb0$f29b12c2@secret.pythonware.com>

Ralf Hildebrandt <hildeb at www.stahl.bau.tu-bs.de> wrote:
> I need the python interpreter to "commit suicide" with a special signal. 
> How do I get the pid of the interpreter to be able to use s.th. like
> 
> os.kill(mypid, signal.SIGSEGV)

>>> import os, signal
>>> os.kill(os.getpid(), signal.SIGSEGV)
Segmentation fault (core dumped)

</F>




From darcy at vex.net  Fri Jan 21 11:36:32 2000
From: darcy at vex.net (D'Arcy J.M. Cain)
Date: 21 Jan 2000 16:36:32 GMT
Subject: DBI Commentary
References: <1263695833-12432931@hypernet.com> <869v1t$a2m$1@nntp9.atl.mindspring.net> <3888CD57.1BD59385@starmedia.net>
Message-ID: <86a1ug$gbo$1@news.tht.net>

Andrew Csillag <drew.csillag at starmedia.net> wrote:
> There's also cursor.description which if you do a SELECT * FROM ... will
> tell you the names of the columns so you can reorder them yourself (or
> do as I do and often turn the result list into a list of dictionaries
> keyed by the downcased column name).

This is basically whay PyGreSQL's dictresult() method gives you against
the PostgreSQL database.

-- 
D'Arcy J.M. Cain <darcy at vex.net>           |  Democracy is three wolves
http://www.vex.net/                        |  and a sheep voting on         
+1 416 425 1212     (DoD#0082)    (eNTP)   |  what's for dinner.


From Christian.Schaller at mchp.siemens.de  Tue Jan  4 12:10:09 2000
From: Christian.Schaller at mchp.siemens.de (Christian Schaller)
Date: Tue, 04 Jan 2000 18:10:09 +0100
Subject: py2bat (was: How do I make a Python .bat executable file?)
Message-ID: <38722971.4B036062@mchp.siemens.de>

Hello...

  I was just reading the above thread and didn't find any program to do
this task automatically.  However, there exists a similar tool for perl,
so I took a closer look at that one.

  After some experiments I came up with a solution that suits for my
needs.  Please don't tell me that anyone else has done that already, I
couldn't find anything on python.org.  Any comments, suggestions?

bye
  Chris...

script starts below
----------------------------------------------------------------------
# name: py2bat.py
# version: 0.1(alpha), Jan/04/2000
#
# description: generate .bat files from python files
# usage: py2bat [-h|-?]
#    or: py2bat [files]
#
# at first, create .bat file for py2bat.py:
#   python py2bat.py py2bat.py
#
# warning: dont use extension (.bat) for executing files generated
# by py2bat.
#
# author: Christian Schaller (Christian.Schaller at mchp.siemens.de),
#         inspired by pl2bat

import sys
import os.path
import getopt

_part1 = '''@echo off
rem="""
if "%OS%" == "Windows_NT" goto WinNT
python -x "%0".bat %1 %2 %3 %4 %5 %6 %7 %8 %9
goto endofpython
:WinNT
python -x "%0".bat %*
if NOT "%COMSPEC%" == "%SystemRoot%\system32\cmd.exe" goto endofpython
if %errorlevel% == 9009 echo ERROR: You do not have Python in your PATH.
goto endofpython
rem """

# python script starts here
'''

# for generating py2bat.bat pay attention to :endofpython to ensure the
shell
# jumping to the right label
_part2 = '\n# end of python script\nrem="""\n:endofpython\nrem """'

_usage = '''Usage: py2bat [-h|-?]
   or: py2bat [files]
       -h or -?  show this help'''

try:
    opts, fileargs = getopt.getopt(sys.argv[1:], '?h')
except getopt.error, e:
    print e
    print _usage
    sys.exit()

for switch, value in opts:
    if switch == '-h' or switch == '-?':
        print _usage
        sys.exit()

# mark for stdin, when no filename is given
if fileargs == []:
    fileargs = ['-']

for filearg in fileargs:
    # should script read from stdin?
    if filearg == '-':
        file = sys.stdin
        # write result to stdout
        out = sys.stdout
    else:
        # filename was given, so we can use that file name
        # for creating the .bat file
        file = open(filearg)
        root, ext = os.path.splitext(filearg)
        if ext == '.py':
            outfile = root + '.bat'
        else:
            outfile = root + ext + '.bat'
        print 'generating', outfile
        out = open(outfile, 'w')

    out.write(_part1)

    while 1:
        line = file.readline()
        if line == '':
            break
        out.write(line)

    file.close()
    out.write(_part2)
    out.close()


From frank.sergeant at redneck.net  Sun Jan 16 13:58:28 2000
From: frank.sergeant at redneck.net (Frank Sergeant)
Date: 16 Jan 2000 12:58:28 -0600
Subject: Python not a Very High-Level Language?
References: <000801bf5802$dcca7d60$0aa0143f@tim> <is66x6vng7.fsf@localhost.ps.uni-sb.de> <87u2kljrmz.fsf@den.home.net> <isbt6rhudp.fsf@grizzly.ps.uni-sb.de>
Message-ID: <87wvp93ktn.fsf@den.home.net>

Denys Duchier <Denys.Duchier at ps.uni-sb.de> writes:

> I apologize for the delay in responding to your question, but our
> newsfeed was out to lunch for about a week.

No apology needed!  Thank you for responding at all.  I am in no
rush, with too many other things to do, but I keep an eye out for
the "best" way to do GUI.  My opinion changes daily, but _today_
I lean toward using Python to generate HTML and let the web 
browser worry about the GUI.

> Drat!  Just the thing I know least about; how did you guess?  I am
> forwarding this message to Christian Schulte who was responsible for
 ...
> scheduled tomorrow where the current state of the GTK interface will
> be presented.  I am eager to learn about that :-)

Please, either of you, post the details as and if it is convenient to
do so.

> The Tk interface follows a fairly simple design: the Oz process starts
> a Tk subprocess and communicates with it via a socket.  

I like the idea of this.  It looks like it might be the simpler/better 
way to manage working with Tk from another language.  I gather Python
can't yet move to Tcl/Tk 8.3 (or anything past 8.05) due to the tighter
coupling between Python and Tk.  It looks like the socket coupling
approach might eliminate much of the maintenance effort in keeping up 
with new releases of Tk.

> On the Oz
> side, Tk widgets are represented by Oz objects.  The Oz/Tk interface
> is actually a very thin OO flavored layer.

> To be concrete, here is the functor (a bit like a Python module) for
 ...

Thanks for the example (I think I understood it) and the reference
to further documentation.  When I find the time ...


  -- Frank



From nic at csee.uq.edu.au  Sat Jan 29 21:24:03 2000
From: nic at csee.uq.edu.au (Nic Williams)
Date: Sun, 30 Jan 2000 02:24:03 GMT
Subject: Dynamic class construction?
References: <389309F7.4A35371F@math.okstate.edu>
Message-ID: <8707c0$ua2$1@nnrp1.deja.com>

In article <389309F7.4A35371F at math.okstate.edu>,
  "David C. Ullrich" <ullrich at math.okstate.edu> wrote:

Just for reference, this is Part i.

>     Say I want to make a function that returns a new _class_.
> I can do this:
>
> def NewClass(whatever):
>     class new(baseClass):
>         pass
>     new.SomeAttribute = whatever.whatever
>     return new
>
> and it seems to do the job. Is there another way to
> do this that seems much better for some reason?

and the next question is part ii.
I'll answer part ii first (helps explain part i soln)

>     I guess a related question would be whether
> having distinct classes with the same __name__
> is going to be trouble - does __name__ get used
> in significant ways internally or is it just there
> for the programmer's convenience?

A class's name when the class is first defined is used in two ways:
i) internal information (so you can ask a class object who it is)
ii) binding information within the namespace

A good way to demonstrate this is with the following example:
>>> class FirstName:
...     pass
...
>>> FirstName
<class __main__.FirstName at e6fe0>
>>> FirstName.__name__ = 'SecondName'
>>> FirstName
<class __main__.SecondName at e6fe0>
>>> SecondName
Traceback (innermost last):
  File "<stdin>", line 1, in ?
NameError: SecondName

Note here that when we change the class's internal name, it now thinks
that it's name is SecondName, but the namespace's binding still calls it
FirstName.

See py doc: http://www.python.org/doc/current/ref/execframes.html

So you can multiple classes who think internally they are the same name,
but they must be named differently if they lie in the same namespace.

Ok, back to part i.

With your example above and below we see that the binding name within
the global namespace is determined by the attribute G. Note that there
is also no class called 'new' anywhere at the end of the method (it was
declared within the local namespace which disappeared at the end of the
method).

So we have an object whose internal name is 'new' but its binding is
'G'. If you want to change the class's internal name to that of its
external binding, use something like:


def newClass(name, ops, superclass):
	class NewClass(superclass):
		pass
	for op in ops.keys():
		setattr(NewClass, op, ops[op])
	NewClass.__name__ = name
	return NewClass

def anOp(self):
	return 'This is an instance of the class %s' % \
		self.__class__.__name__

someOps={'__str__':anOp}
G=newClass('G', someOps, Group) # Group defined elsewhere
print G
<class __main__.G at f92a8>

BINGO. It's name binding and internal naming now match.

Hope that answered the question.

Nic


>     (In case someone feels the answer might depend
> on the context: it's mathematical structures. For
> example baseClass might be Group; a subclass
> of Group "is" a group, and an instance of the
> subclass "is" an element of the group. So I might
> want to create a group using
>
> def NewGroup(groupOp):
>     class new(Group):
>         pass
>     new.__mul__ = groupOp
>     return new
>
> so then
>
> G = NewGroup(SomeGroupOp)
> g1 = G(someparams)
>
> makes g1 an element of the group G.)
>
>


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


From goliver at netset.com  Tue Jan 18 17:20:50 2000
From: goliver at netset.com (Guy L. Oliver)
Date: Tue, 18 Jan 2000 17:20:50 -0500
Subject: Python and MS SQL (Microsoft SQL server)
References: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in> <3884C60B.BE820782@psychosis.com>
Message-ID: <862p18$hce$1@athena.netset.com>

im trying to do the same thing. there are commercial odbc drivers/managers
for unix, but the ones I found were quite expensive.  I gave up looking and
just wrote a bridge that would allow my code on the solaris box to remotely
call a server I wrote on a nt box that did the queries (using win32 odbc, or
ado, etc...), and returned the information to the solaris box.

If you (or anyone else reading this) find a good odbc bridge for solaris
that doesnt cost an arm and a leg, let me know.  the mxODBC package has some
odbc drivers available, and pointers to a driver manager for unix, but so
far, I havent found any thing to talk to sql server outside of intersolv's
unix drivers, which were, last time I checked, quite expensive.  and of
course, my project doesnt have much of a budget.  I found the little bridge
to be a passable solution to the problem, thought ymmv.

-G


"Dave Cinege" <dcinege at psychosis.com> wrote in message
news:3884C60B.BE820782 at psychosis.com...
> "George Thom at s" wrote:
> >
> > Hi Dave,
> > The odbc module that comes as part of the win32 extensions bundled with
> > Python 1.5.2 for the various flavours (!) of Windows should do just
fine.
>
> So you're saying python has to be runnign on NT for this to work?
>
> > Additional functionality (with some tweaking ... ymmv ;-) can be
obtained
> > with marc Lemburg's mxODBC module.
>
> What about this? we're looking to run on solaris or linux.
>




From tim_one at email.msn.com  Sat Jan 29 15:36:48 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 29 Jan 2000 15:36:48 -0500
Subject: concurrency in Python
In-Reply-To: <86utmv$1va$1@nnrp1.deja.com>
Message-ID: <000001bf6a98$9142a360$78a0143f@tim>

[posted & mailed; added Python's Thread-SIG]

[Greg Wilson]
> Hi.  I'm interested in hearing from anyone who's ever built a
> concurrency toolkit for Python that used a model other than
> threads + locks --- futures, tuple spaces, active expressions,
> concurrent aggregates, actors, or anything else.

At the 6th Python Conference, Michael Papathomas and Anders Andersen
presented their research on "Concurrent Object-Oriented Programming in
Python with ATOM", a capable experimental framework based (roughly) on
inheritable synchronization constraints.  The full paper is available here:

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

Of course, Java's recent offshoot of Linda's tuple spaces should be
available thru JPython.

Let us know if you turn up anything else!  Most thread activity in the
Python world has been on the commercial side, where threads+locks is "it"
for the usual lowest common denominator reasons.

mix-'em-in-with-floating-point-and-it's-an-unbeatable-
    enabler-of-late-night-debugging<wink>-ly y'rs  - tim





From volumeregeling at hotmail.com  Thu Jan 20 11:34:52 2000
From: volumeregeling at hotmail.com (volumeregeling)
Date: Thu, 20 Jan 2000 16:34:52 GMT
Subject: Exiting program
Message-ID: <38873769.5048399@122.0.0.250>

Hi,

I am not very good in searching through the online documentation for
python. I am trying some things which keep generating errors on
running.

I have the following program :

I am making a menubar which uses functions from a cupple of COM
objects. I want to exit my program through a stop_program function not
from any COM object.

def Exit_program():
    Editor.mnuFile_Exit()
    sys.exit()                        <-- for some reason this also
doesn?t work. but that on problem jet.

Editor in the above is a COM object, with the method from that COM
object to close it.

When the editor.mnuFile_Exit() method is invoked , python crashes with
some OLE error. But when I run the COM object from WScript it exits
fine so the problem is in python.

I hooked the Editor COM object the following way :

Editor= win32com.client.Dispatch("Comeditor.Classeditor")

Can anyone tell me how to "release" the COM object ??
Editor.release() doesn?t work.

I also can?t seem to get a variable COMMAND_Button to the Exit_program
procedure. I need to close the menubar with command_button.quit(), ubt
then it must be invoked from the procedue where command_button is
created


Can anyone tell me how to pass-throug this caommand_button so i can
invoke command_button.quit() from the Exit_program function.


From thamelry at vub.ac.be  Tue Jan 25 03:53:32 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 25 Jan 2000 08:53:32 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim>
Message-ID: <86joac$l2s$1@mach.vub.ac.be>

Tim Peters <tim_one at email.msn.com> wrote:

[snip]
: like-sex-is-great-but-should-not-be-part-of-life-ly y'rs  - tim

Good analogy, actually. Sex is great, but should never be forced
on you.

BTW, people here seem to have adopted the idea that if you don't like
a particular feature of a language, you should abandon it (e.g., "You 
don't like indentation? Use Perl!"). But many languages have evolved 
exactly this way: users criticize, complain, praize and whine, thereby
influencing the shape of the languge in question. You can find a very 
nice description of this process with respect to C++ in "The Evolution 
of C++ : Language Design in the Marketplace of Ideas", by Jim Waldo 
(Editor), James Waldo (Editor).

Cheers,

--- 
Thomas Hamelryck    Institute of Molecular and Structural Biology 
Aarhus University   Gustav Wieds Vej 10C 
DK-8000 Aarhus C    Denmark                   







From quinn at mono.ugcs.caltech.edu  Fri Jan 28 17:02:48 2000
From: quinn at mono.ugcs.caltech.edu (Quinn Dunkan)
Date: 28 Jan 2000 22:02:48 GMT
Subject: lib reference in Unix man format
References: <200001281339.IAA12329@trm.bmo.com> <20000128090950.A2767@better.net> <oqsnzirygy.fsf@titan.progiciels-bpi.ca>
Message-ID: <slrn8944j2.7f4.quinn@mono.ugcs.caltech.edu>

On 28 Jan 2000 10:46:05 -0500, Fran?ois Pinard <pinard at iro.umontreal.ca> wrote:
>William Park <parkw at better.net> writes:
>
>> HTML is essentially killed INFO, in my opinion.
>
>There is still one thing from Info which is sorely lacking in HTML, which
>is the lack of capability to textually search in a whole document.

And don't forget the the ability to run info through TeX and come up with nice
printed docs.  I'm sure similar tools exist for html, but TeX is the standard
for high quality typesetting.

>I have a big deal of Texinfo documentation in both Info and HTML format,
>here, and Info is much more useful that HTML browsers.  In my opinion, Info
>was hurt by die-hard `man' addicts on one side, and more by the popularity
>of HTML than its real virtues.  Info needs you to know the first letter
>of the words Next, Previous, Up and Menu, and to use the SPACE key to
>page forward.  Nothing worth whipping a cat, really.

Despite this, I always had a lot of trouble trying to remember the keystrokes
to the info browser.  It annoyed me so much I would often read info with less.
I don't have any trouble with keystrokes for other programs, but for some
reason I could never grasp info.  I always assumed gnu put a minimal amount of
effort into info because they used emacs and assumed everyone else did too.

The concept of a simple, unified, hyperlink-capable, readable as plain text
but can also be typeset nicely documentation is something I think is a great
idea.  I was never very fond of sgml (or xml) for its complexity.  So I was
unhappy that info wasn't practical for me.  Should have been integrated into
man anyway.  Oh well.


oops, this is getting off-topic...


From aa8vb at yahoo.com  Wed Jan 19 13:47:44 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 19 Jan 2000 13:47:44 -0500
Subject: tkinter and stdout redirection
In-Reply-To: <864iv0$r6o$1@news1.sunrise.ch>
References: <864iv0$r6o$1@news1.sunrise.ch>
Message-ID: <20000119134744.A512978@vislab.epa.gov>

nathan:
 |if I redirect the output stream in a file with
 |
 |sys.stdout = open ("file", "w")
 |
 |direct from the command prompt, I get the output  stream in my file without
 |any problems.
 |
 |But if I do the same from a tkinter application , I get the output stream
 |only when I close the application.
 |
 |I think that there's a buffer between the output stream and my file, but
 |I've not found anywhere how to set it.  Someone can help me?  Thank you

You can execute sys.stdout.flush() every so often.  Or, here's a
slick way to get your file to do it for you:

    http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=531092833

-- 
Randall Hopper
aa8vb at yahoo.com



From kuncej at mail.conservation.state.mo.us  Wed Jan  5 15:52:46 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Wed, 5 Jan 2000 14:52:46 -0600
Subject: http POSTing
References: <8EB19A9E7newshoesmailcom@205.147.63.94> <OOJc4.245$B5.4746@news.more.net> <38739886.EB1A06B1@calibre-dd.com>
Message-ID: <2aOc4.247$B5.4747@news.more.net>

> > For a simple example, see
> >     urlpost.py
> > at
> >     http://starship.python.net/crew/jjkunce/
>
> A very nice example, but perhaps not as simple as it might be?

>From a high-level/black-box point-of-view, it seems prety simple to me:

import urlpost
postdata = {}
postdata['hidd']='en'
postdata['email']='you at yourmail.com'
answer = urlpost.urlpost('www.test.com', postdata)
print answer

And as a bonus, it even works through a firewall :-)

  --Jeff





From fredrik at pythonware.com  Wed Jan 19 04:49:31 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 10:49:31 +0100
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net>
Message-ID: <006701bf6262$7f5b2520$f29b12c2@secret.pythonware.com>

Tom Culliton <culliton at clark.net> wrote:
> BTW - If you want to wave Nick Wirth at us, I think he tends
> toward baby talk in his designs.

maybe "tye4" is Wirth posting as an Anonymous Coward?

after all, it cannot be Knuth; he's known to have said that
"structure from identation" is the future.

</F>




From thomas at xs4all.net  Tue Jan 11 19:59:24 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 12 Jan 2000 01:59:24 +0100
Subject: a small problem
In-Reply-To: <85ggif$u8t$1@nnrp1.deja.com>; from knotwell@my-deja.com on Wed, Jan 12, 2000 at 12:06:50AM +0000
References: <85ggif$u8t$1@nnrp1.deja.com>
Message-ID: <20000112015923.A8450@xs4all.nl>

On Wed, Jan 12, 2000 at 12:06:50AM +0000, knotwell at my-deja.com wrote:

> Is the following a bug?

> >>> q = [(1,'dd'),(2,'ddd'),(3,'ddddd')]

> >>> reduce(lambda x,y: x[0] + y[0],q)

> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 1, in <lambda>
> AttributeError: __getitem__

My version of Python (admittedly, the CVS version) says something slightly
more obvious:

>>> q = [(1,'dd'),(2,'ddd'),(3,'ddddd')]
>>> reduce(lambda x,y: x[0] + y[0],q)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 1, in <lambda>
TypeError: unsubscriptable object

The problem is not in map, but in your lambda. You have to remember what
reduce does :)

For the first run, if not given an 'initial' value, it takes the first two
elements of the list, and gives them to the passed function (the lambda, in
your case.) If you give an initial value, it uses that and the first element
of the list, and passes that to the function.

For every subsequent run, it runs the function on the return value of the
_previous_ run, and the next element of the list.

So, in your example, here's what happens:

reduce takes (1,'dd') and (2,'ddd'), and passes them to the lambda
The lambda takes 1 ((1,'dd')[0]) and 2 ((2,'ddd')[0]) and adds them,
yielding 3.

reduce takes 3 and (3,'ddddd') and passes them to lambda
lambda tries to do '3[0]', and fails.

You need to do something like this, to force lambda to return an indexable
object (of which element 0 is the element you want reduce to use next time.)

>>> reduce(lambda x,y: (x[0] + y[0],),q)
(6,)

or perhaps

>>> reduce(lambda x,y: (x[0] + y[0],),q)[0]
6

depending on what you want. But I think this is roughly the point where
people like Guido(*) start wondering if reduce is a good idea at all. I'd
say the above reduce is almost a good argument for perl ;) Usually, what
you're trying to do so forcefully with just reduce and lambda is much more
elegantly (and readably!) done with map and reduce (and lambda, in this
case):

>>> import operator
>>> tmplist = map(lambda x: x[0], q)
>>> tmplist
[1, 2, 3]
>>> reduce(operator.add, tmplist)
6

(This may seem like nitpicking for the above example, but it makes for
readable code later on, when you write longer and longer and more
complicated map/filter/reduce sequences. The FAQ has a nice example, in the
'oneliner' section ;)

(*) I'm assuming here. I dont actually know so, because i'm not foolish
enough to consider myself knowledgable in all things Guido, but these are
the hints i get from other postings on the subject, by other people.

bytesize-chunks-of-code-make-a-pleasant-meal-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From gmcm at hypernet.com  Tue Jan 11 12:59:28 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 11 Jan 2000 12:59:28 -0500
Subject: [ANN] Multimethod.py -- multimethods for Python
In-Reply-To: <387B6733.2C49FB8A@home.com>
Message-ID: <1264508924-1480880@hypernet.com>

Doug Hellmann wrote:

> How is this different from function overloading, as implemented
> in C++? I'm not a C++  user, but my impression is the effect
> would be the same. Is that the point?

Not at all. C++ (and Java) are limited to figuring out one 
override / overload at a time. If you are dealing with base class 
pointers, you need to bounce the message around for awhile 
to resolve all of them. First, C++ will discover the proper 
derived class (the override), and then the proper overload. 
Unfortunately, this means declaring scads of methods which 
don't do anything except let the language stumble through the 
resolution.

So in order to do a nice simple case:
 widget->process(event)

You need to do something like:
  Widget::process(Event e) // not actually used
  Button::process(Event e) {e->getProcessedBy(this);}
  Text::process(Event e) {e->getProcessedBy(this);}

  SpewEvent::getProcessedBy(Button) ...
  BarfEvent::getProcessedBy(Text)...

but you also need to fill in most of the other squares of the 
cartesian product, or the language gets mad at you.
 
say-what-I-want-you-to-say-not-what-you-mean-ly y'rs

- Gordon



From erno at iki.fi  Wed Jan 19 19:49:08 2000
From: erno at iki.fi (Erno Kuusela)
Date: 20 Jan 2000 02:49:08 +0200
Subject: Realtime scheduling??
References: <38858064.CDA4866C@yahoo.com>
Message-ID: <87ogahtvnf.fsf@fun77.kivikko.hoas.net>

>>>>> "paul" == Paul Winkler <slinkp23 at yahoo.com> writes:

    paul> Hi, I'm working on a module I eventually hope to use to
    paul> generate musical events in realtime. (These events will be
    paul> sent to MIDI devices or to software synthesizers.)

i think there is some midi support in the stock sound drivers.
you might be better off using that.

    paul> I've found that on my Linux system (pretty much stock RedHat
    paul> 5.2), scheduling actions with sched.scheduler is not as
    paul> reliable as I'd like. Events are anywhere from 2 to 20
    paul> milliseconds late.

yeah, linux is not a real-time operating system. using a "real-time"
scheduling class might, or might not, help. they are mostly a way to
tell the scheduler to choose your task over other tasks when it's time
to switch.

    paul> I'd like to improve the reliability, which brings me to my
    paul> next question:

    paul> Is there any way to run a python script, or rather the
    paul> python interpreter itself, with SCHED_FIFO priority on a
    paul> linux system?  Or any other way to guarantee that something
    paul> will happen when you really want it to happen??

there are programs that work like nice(1), one is named rtnice. a web
search should find it for you. then you can say
"rtnice python myprogram.py".

  -- erno


From gerrit.holl at pobox.com  Mon Jan  3 11:17:02 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 3 Jan 2000 17:17:02 +0100
Subject: Doing things in one line?
Message-ID: <20000103171702.A655@stopcontact.palga.uucp>

Hello,

I know there's a FAQ entry called "Can I write obfuscated one-liners
in Python?", but for fun, I'm trying to do it now.

Can I do this in one line:
if a: foo()
elif b: bar()
else: foobar()

Is it possible with an intelligent construct that looks like:
a and foo() or a and bar() or foobar()

This does both foo() and foobar().

And what about while- and forloops?

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From goliver at netset.com  Mon Jan 24 20:00:57 2000
From: goliver at netset.com (Guy L. Oliver)
Date: Mon, 24 Jan 2000 20:00:57 -0500
Subject: using threads with python and winnt
References: <1264075593-4560755@hypernet.com> <38835f4c.1380925@news.iol.ie> <388C1CE3.3391A5CF@zebrasoft.co.nz>
Message-ID: <86ismn$50a$1@athena.netset.com>

Any more details?  I've been using threading package under 98, 95, nt,
linux, and solaris with no problems.  Also, the semaphores, conditions,
queue, etc.  ive never had any problems (thread pool based web server
experiment, with python generated dynamic content)

Id be very interested to hear of any problems you had so I can watch out for
them.

glo

"Matthew Miller" <matthewm at zebrasoft.co.nz> wrote in message
news:388C1CE3.3391A5CF at zebrasoft.co.nz...
> Be afraid, be very afraid! Seriously though, I head-banged with threading
> under Win98 and eventually gave up. The app though works perfectly under
> WinNT. Unless it is REALLY neccessary to support Win9x, spare yourself
from
> the pain. matthew.
>
> Sean Mc Grath wrote:
>
> > >Kreiner Karl or Charly or somebody wrote:
> > >
> > >>             does anyone have expiriences with using threads and winnt
?
> > [...]
> > <gmcm at hypernet.com> wrote:
> > >Threads work dandy on NT
> > [...]
> >
> > Does anybody know how good/bad/indifferent the threading stuff works
> > on Windows 95/98? I am considering using threads for an app
> > under development but I'm a bit worried (read "terrified") about
> > the issues that might pop up in '95 and '98.
> >
> > regards,
> > Sean McGrath
> >
>
> --
> Matthew Miller
> Zebra Software Ltd.
>
> E-mail: matthewm at zebrasoft.co.nz
> Ph: 025-2920625 / FAX: 03-3842397
>
>




From kc5tja at garnet.armored.net  Tue Jan 11 18:12:06 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 11 Jan 2000 23:12:06 GMT
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov> <85fnbk$j6h$1@nntp9.atl.mindspring.net>
Message-ID: <slrn87neil.qla.kc5tja@garnet.armored.net>

In article <85fnbk$j6h$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:
>IOW, similar to C++-style multiple dispatch?

C++ doesn't support multiple dispatch; it supports function overloading,
which isn't the same thing.  Function overloading requires that all types of
all objects be known _precisely_ at compile time, so that the compiler can
find the correct (global) function to call.

Multiple dispatch takes all the types of the supplied arguments and, at _run
time_ decides which method to invoke.  The types of methods need not be
known precisely at compile time.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From paul.magwene at yale.edu  Wed Jan 26 18:53:38 2000
From: paul.magwene at yale.edu (Paul Magwene)
Date: Wed, 26 Jan 2000 17:53:38 -0600
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com> <86np80$jhp$1@harvee.billerica.ma.us>
Message-ID: <86ntlg$2tc$1@news.ycc.yale.edu>

Eric S. Johansson <esj at harvee.billerica.ma.us> wrote in message news:86np80
[snip[
>
> > hey, who the heck is reading code over the phone
> > these days?
>
> anyone trying to program using speech recognition.  Why would they
> do this you might ask?  Probably because they were injured from
> programming using keyboards.
>
> yes, it's not strictly reading code over the phone but there
> are significant similarities...
>

Yes, but isn't it possible to set the speech recognition software so when
you say "indent" it tabs right, and when you say "dedent" it tabs left?

--Paul



From tothfile at freemail.hu  Thu Jan 27 08:09:46 2000
From: tothfile at freemail.hu (Attila =?iso-8859-1?Q?Filet=F3th?=)
Date: Thu, 27 Jan 2000 14:09:46 +0100
Subject: What is winfo_id() under Win32
Message-ID: <3890439A.1C1BA6D2@freemail.hu>

Hi!

I would like to maximize a Toplevel window created from python in Win32.
Since I couldn't find any methods (or options) in Tkinter I've tried
win32xxx imports. I obtained the hwnd from self.winfo_id() and then
tried a win32gui.ShowWindow(hwnd, SW_SHOWMAXIMIZED): It didn't work.
I've checked the hwnd with Spy and I realized the window is composed of
two windows (hwnds): one is for the outer and one for the inner area (as
the outer's child). And winfo_id() returns  hwnd of the inner area.
Is it normal?

So in Windows terminology I have to call a GetParent(hwnd) and for the
hwnd returned call ShowWindow(). 
The other problem is that I couldn't access to GetParent() by importing
win32ui (where this method should live). Why is it?

To get the above text clearer, here is an example code:

[
from Tkinter import *
from win32con import *
import win32gui
import win32ui

class app(Toplevel):
    def __init__(self, master):
        self.master = master
        Toplevel.__init__(self, master)
        self.update_idletasks()
        hwnd = self.winfo_id()
#       hwnd = win32ui.GetParent(hwnd)  # Attribute error: GetParent
        win32gui.ShowWindow(hwnd, SW_SHOWMAXIMIZED)  # It would only
work after the GetParent() call

root = Tk()
root.withdraw()
app(root)
root.mainloop()
]

(I'm using Python 1.5.2)

Thanks for the help in advance.
Best Regards,
  Attila


From skip at mojam.com  Thu Jan  6 20:03:57 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 6 Jan 2000 19:03:57 -0600 (CST)
Subject: Finite State Machines (was "Re: control structures (was "Re: Sins")")
In-Reply-To: <011501bf58a4$7b0db540$0100a8c0@rochester.rr.com>
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
	<14452.45436.641890.197509@beluga.mojam.com>
	<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
	<14452.60226.853060.116123@beluga.mojam.com>
	<011501bf58a4$7b0db540$0100a8c0@rochester.rr.com>
Message-ID: <14453.15229.833652.832323@beluga.mojam.com>

    Darrell> Using an exception as a "switch" statement is a cool idea.  A
    Darrell> dictionary approach comes to mind, with so many states.

Yup...  Check out

    http://www.musi-cal.com/~skip/python/

and scroll down to "Finite State Machine" for a slightly more elaborate way
of doing finite state machines...

Skip



From claird at starbase.neosoft.com  Tue Jan 18 10:03:20 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 18 Jan 2000 09:03:20 -0600
Subject: Can TkInter be frozen?
References: <1263967434-11066381@hypernet.com>
Message-ID: <BF71A72C2C3451AA.44DFB5F8DFE97D59.0BDAEAB17E554E7D@lp.airnews.net>

In article <1263967434-11066381 at hypernet.com>,
Gordon McMillan <gmcm at hypernet.com> wrote:
			.
			.
			.
>If you're on Windows, my Installer will do it for you:
>http://starship.python.net/crew/gmcm
>
>Haven't tried to incorporate that in the Linux version, because 
>of issues with finding the tcl and tk shared libs at runtime.
>
>> Is it expected to change?
>
>My guess would be that if stock Tcl/Tk incorporated a freeze-
>like utility that enabled this, then incorporating it into Python's 
>freeze would follow. But I doubt that anything depending on a 
>Tcl/Tk extension will ever find it's way into the Python 
>distribution.
>
>- Gordon
>

More gossip:  Tcl has at least a half-dozen common ways
to make installations (in principle, they're in <URL:http://
starbase.neosoft.com/~claird/comp.lang.tcl/tcl_compilers.html>,
although that page is uninviting for a person pursuing this one
topic in isolation).  None of them are sufficiently common to
encourage adoption into a Python distribution.

Installation is something our industry doesn't yet come close
to getting right.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From gerrit.holl at pobox.com  Tue Jan  4 16:01:52 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 4 Jan 2000 22:01:52 +0100
Subject: Grail
In-Reply-To: <3872441A.7E86CCED@callware.com>; from ivanlan@callware.com on Tue, Jan 04, 2000 at 12:03:54PM -0700
References: <387237F3.C35B7FA1@callware.com> <14450.17021.111722.710187@weyr.cnri.reston.va.us> <3872441A.7E86CCED@callware.com>
Message-ID: <20000104220152.A3894@stopcontact.palga.uucp>

Ivan Van Laningham wrote on 946983834:
> Hi Fred--
> 
> >  > If this can't be made to run, are there any HTML widgets out there?
> > 
> >   Doesn't wxWindows include one?  Or are you specifically interested
> > in Tk?
> > 
> 
> Um, yeah, I was hoping to get a Tk HTML widget, since the entire rest of
> the application is already written in Tk.  An HTML widget would let me
> use the existing HTML file for help without having to fire up netscape
> or IE. ...

I think pysol uses this, have you read their source yet?

groeten,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From a.eyre at optichrome.com  Tue Jan 25 06:12:32 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 25 Jan 2000 11:12:32 -0000
Subject: newbie question about self and eval
In-Reply-To: <86jjkj$jmu$1@news.inet.co.th>
Message-ID: <000301bf6725$13750430$3acbd9c2@peridot.optichrome.com>

> class myclass:
> 
>     def __init__(self,filename='c:\\test.txt'):
>         self.fee = []
>         self.fi = []
>         self.fo = []
>         self.fum = []
>         f = open(filename,'r')
>         rl = f.readlines()
>         f.close()
>         for lines in rl:
>         token = string.split(lines,'\011')
>         # do something more efficient than this !!!
          # Okay. One way is this...
          try:
              bucket = self.__dict__[token[0]]
          except KeyError:
              bucket = self.fum
          bucket.append(token[1])

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From parkw at better.net  Fri Jan 14 14:23:39 2000
From: parkw at better.net (William Park)
Date: Fri, 14 Jan 2000 14:23:39 -0500
Subject: Database to Web
In-Reply-To: <jmsu7sk17q2hg0dbrf3qi8lobednf72m4a@4ax.com>
References: <jmsu7sk17q2hg0dbrf3qi8lobednf72m4a@4ax.com>
Message-ID: <20000114142339.A468@better.net>

On Fri, Jan 14, 2000 at 07:04:52PM +0000, Simon Faulkner wrote:
> Has anyone got a good sugestion for the easiest way to display the
> results of an sql query on a web page?

It's very difficult to generalize, because each case is different.  For
the code you quoted, do you want a nicer table?  If so, why couldn't you
put some <td> tags?


> This works but it's not pretty!

    <cut>...</cut>
    
>     print "<table border cellspacing=0 cellpadding=5>"
>     for record in range(total):
>       Name =  tuple[record]
>       print "<tr>"
>       print Name
>       print "</tr>"
>     print "</table>"



From ralph at inputplus.demon.co.uk  Sun Jan 23 14:32:38 2000
From: ralph at inputplus.demon.co.uk (Ralph Corderoy)
Date: 23 Jan 2000 19:32:38 -0000
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com>
Message-ID: <86fl0m$gmj$1@inputplus.demon.co.uk>

> > _Python Essential Reference_ is just what it says; a reference.
> > If, like me, you are an experienced (which says nothing about the
> > quality of my code ;^)) coder, you can use it to learn Python.
> 
> Seconded.  If you already know a few languages, and already now OOP,
> then PER is fine to learn from.

OK, I learnt C from the excellent K&R (Old Testament), all the Unix
little languages, and many more besides.  I learnt Perl 4 years ago
purely from the man pages, and Perl 5 at the same time from
eavesdropping on the perl5-porters mailing list.  I like well written,
terse, books.  Given all of that it sounds to me as if I should remove
_Programming Python_ from my shopping basket and add _Python Essential
Reference_.  Your views are sought.


Ralph.



From andrew.henshaw at gtri.gatech.edu  Mon Jan 24 10:55:54 2000
From: andrew.henshaw at gtri.gatech.edu (Andrew Henshaw)
Date: Mon, 24 Jan 2000 10:55:54 -0500
Subject: Whitespace delimiters suck - Seen Occam?
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <86cmom$5pp$1@news-int.gatech.edu> <86h7jm$o7i@newton.cc.rl.ac.uk>
Message-ID: <388C760A.1E66C20@gtri.gatech.edu>

Richard Brodie wrote:
> 
> "Andrew Henshaw" <andrew.henshaw at gtri.gatech.edu> wrote in message
> news:86cmom$5pp$1 at news-int.gatech.edu...
> 
> > By the way, from my long experience with Occam - I can enthusiastically
> > recommend using a folding editor with indentation languages like Python.
> > They're made for each other.
> 
> Curious. Everyone I knew who programmed in Occam had a profound
> dislike for the folding editor - it put me off the idea of using whitespace
> as syntax for years. I've got over it now though ;)

That's interesting.  Everyone here, who has used it for any
reasonably-sized
project, has really liked it.  I've seen opposition from programmer's
who
try it for a day and then immediately abandon it, however.  Have you
tried the Windows version of Origami from Elcom (it adds several
niceties, 
but the default color scheme is horrible)?  There is a new Linux version
of the
free Origami.  I'm getting ready to try that one.

The main problem that I've seen are the non-intuitive key bindings. 
That can 
be fixed -- but I like the old ones.


From Oliphant.Travis at mayo.edu  Tue Jan 25 11:50:41 2000
From: Oliphant.Travis at mayo.edu (Travis Oliphant)
Date: Tue, 25 Jan 2000 10:50:41 -0600 (CST)
Subject: NumPy
In-Reply-To: <388DC152.9624D8F5@obop.com.pl>
References: <388DC152.9624D8F5@obop.com.pl>
Message-ID: <Pine.LNX.4.10.10001251045490.10615-100000@us2.mayo.edu>

The new NumPy requires the distutils available at:

http://www.python.org/sigs/distutils-sig/Distutils-0.1.2.zip


Admittedly, there should put a link somewhere on the page, or a more
helpful error message in setup.py

Sorry for the inconvenience.

Travis Oliphant






From jamarijr at hotmail.com  Mon Jan 10 12:56:27 2000
From: jamarijr at hotmail.com (Arinte)
Date: Mon, 10 Jan 2000 17:56:27 GMT
Subject: Using PyString_AsString? Very Simple.
Message-ID: <85d6g3$f74$1@nnrp1.deja.com>

Should I delete the char* that I get from PyString_AsString()???  Don't
readily see this in docs.


TIA.


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


From aahz at netcom.com  Wed Jan 19 11:11:53 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jan 2000 16:11:53 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com> <84pg9v$rda$1@news1.xs4all.nl> <84qg3m$u0v$1@nntp1.atl.mindspring.net> <84qqs4$dn7$1@news1.xs4all.nl>
Message-ID: <864no9$icc$1@nntp3.atl.mindspring.net>

In article <84qqs4$dn7$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
>Aahz Maruch <aahz at netcom.com> wrote:
>>
>> Oh, okay.  Let's suppose, for example, that you have a customer
>> database, and the client wants to view the entire order history for a
>> customer.  Would you say that it's inappropriate for the database to
>> return thousands of rows for an active customer?
>
>Well, it is if the customer is one of many who expects lightning
>performance. If the customer wants a lot of data he will have to wait,
>and he will simultaneously tie up the server for other customers. 

Uh, if it's a properly constructed database, it shouldn't get tied up
like that.

>If
>he's seriously going to look at thousands of records, well, so be it
>then. But it remains a fact that what a database is best at is selectively
>presenting a bit of data out of a large mass. If your relational database
>is mostly employed serving up most of its contents every query, something
>is wrong, and you will get the blame for its lacklustre performance.

If you have millions of records, a few thousand is very far from "most
of its contents".  Also, from my POV, the client is usually a Python
script, so it *is* interested in every single record.

>Even so, in most cases I've seen, when thousands of rows are retrieved
>there's designers/developers laziness or ignorance behind it - and in
>most of the other cases the customer can be 'educated' in wanting a more
>selective result. Nobody is going to eyeball more than a few hundred
>records, and if they are going to post-process the result it is often
>better to do the post-processing on the server and send out the result.

Often, yes, but far from always.  For example, if one has twenty groups
of records, each group with roughly two hundred items, there's no
efficient SQL to pick up the first thirty items of each group.  Yes, one
could do twenty queries, but it might be more efficient to post-process
a single query.

>> Similarly, suppose you have millions of BLObs that change regularly.
>> How would you organize them on an HTTP server?  Incidentally, HTTP is
>> *not* an efficient transmission mechanism (less true if you're using
>> HTTP 1.1).
>
>If you have millions of large blobs you will have problems no matter
>what - nothing I can say or do can worsen or lighten those ;-). I'm not
>sure that I know the right solution to problems like that. I know of a
>few sites where they regularly work with data like that, but they don't
>store it in a relational database, they use custom solutions, and still
>often store the blobs on the filesystem.

Really?  How do they map the filesystem to the database?

>I know http isn't that efficient, but it allows you to a) offload the
>demand to other servers very easily, and  b) present the textual data
>already while the customer is still waiting for the picture. 

a) is only true if you're replicating the file system or have a really,
really good network file server.  b) doesn't apply if the blob is part
of the data stream being used by the client (the blob is not an image).

>Anyway, while interesting, this is getting away from Python more
>and more...

Sure, but as you can see from the postings here, many people use Python
to access databases, so it isn't completely off-topic -- particularly
when the discussion revolves around whether it's better to do processing
in Python or in the database.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From kochmar at pitfiend.psc.edu  Fri Jan 28 16:13:03 2000
From: kochmar at pitfiend.psc.edu (John Kochmar)
Date: 28 Jan 2000 21:13:03 GMT
Subject: statements and blocks
Message-ID: <86t0ov$4en$1@news.psc.edu>

OK, I haven't seen this in the FAQ (Yeah, I did look ;^), and maybe I 
just missed it, but someone at the Python conference (sorry, I forget
which talk) brought up that you couldn't do:

	print "a"
	print "b"
		print "c"
		print "d"
		print "e"
	print "f"

to visually set off a block of statements.  And ahort of forcing a different 
scope though a conditional, loop, or some other scope modifier, I haven't 
been able to create a block of statements in a seperate context.

I'm sure there was a good reason to do this, maybe it's even been
brought up before (and I missed it, I only scan the newsgroup), but
I've used this in other languages (mostly Java for AWT readability),
so I was wondering what kept it out of Python...

John


From gerrit.holl at pobox.com  Sun Jan 16 15:59:18 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 16 Jan 2000 21:59:18 +0100
Subject: Breaking out of walk()
In-Reply-To: <024601bf605a$1c34f8e0$f29b12c2@secret.pythonware.com>; from fredrik@pythonware.com on Sun, Jan 16, 2000 at 08:44:31PM +0100
References: <38811854.F2BC197E@rl.gov> <3881f20c.844046140@news1.on.sympatico.ca> <20000116195919.A7714@stopcontact.palga.uucp> <024601bf605a$1c34f8e0$f29b12c2@secret.pythonware.com>
Message-ID: <20000116215918.A9305@stopcontact.palga.uucp>

Fredrik Lundh wrote on 948051871:
> Gerrit Holl <gerrit.holl at pobox.com> wrote:
> > Robert Roy wrote on 948036970:
> > > if 'cvs' in names: names.remove('cvs')
> > > 
> > > will skip over directories named 'cvs'
> > 
> > It won't:
> > >>> l = ['foo', 'cvs', 'bar', 'cvs']
> > >>> l.remove('cvs')
> > >>> l
> > ['foo', 'bar', 'cvs']
> 
> it will -- when the walk callback is called, the "names"
> argument only contains the files and subdirectories
> directly under "dirname"...
> 
> (if os.listdir returns duplicates on your computer,
> you should fix your operating system)

Oh yes, of course... stupid me :-)

regards,
Gerrit

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From sam at mitre.org  Mon Jan 31 17:00:12 2000
From: sam at mitre.org (Samuel L. Bayer)
Date: Mon, 31 Jan 2000 17:00:12 -0500
Subject: Tkinter and signals
Message-ID: <389605EC.89687799@mitre.org>

All -

When I set a signal handler (SIGINT, SIGTERM, etc.) in a Tkinter
application and send the signal to it, the application doesn't actually
vanish from the screen until I move the mouse over it. In particular,
the signal isn't even *seen* by the app until I move the mouse over it.
A couple people have reported this previously, but no general solution
has been posted. For instance, Alexandre Ferrieux wrote on 2/11/1999:

"It is a known issue with the Tcl/Tk event loop, that when a signal is
configured to
 *restart* system calls it interrupts, the event loop will not get a
chance to call
 the internal mechanism needed to detect the flags set by the low-level
signal
 handler (AsyncReady). Thus, signals-aware versions of Tcl should
configure the signal handlers *without* the SA_RESTART flag, otherwise
the final script-level callback is only called after the
 next GUI event (mouse...). I don't know the case of TkInter, though."

This comment doesn't apply to my situation, since I'm trying to exit the
application. Furthermore, the wish executable built from the identical
library *does* exit appropriately when the signal is sent. I've tried
this with both 1.5 and 1.5.2 (both using Tk 8.0, I think), and I get the
same behavior in each case.

Does anyone have a solution to this problem? Please CC me directly in
your response, since I'm not reading c.l.p regularly nowadays.

Thanks in advance -
Sam Bayer
The MITRE Corporation
sam at mitre.org



From mmc at obispo.com  Thu Jan 27 13:01:29 2000
From: mmc at obispo.com (Miles Clark)
Date: Thu, 27 Jan 2000 10:01:29 -0800
Subject: CGI: keeping mySQL awake between hits?
References: <86nn6m$tkg$1@nnrp1.deja.com>
Message-ID: <389087F9.931ABC13@obispo.com>

Have a look at FastCGI (http://www.fastcgi.com).  It has a similar API
to CGI, but creates a persistent process that allows you to cache DB
connections, etc.

Here's a *very* simple example of a python FastCGI script that accesses
a PostgreSQL database.  Note that the db connection is not recreated for
every hit (all hits are handled in the while loop):

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

#!/usr/bin/python
#-*- mode: Python; tab-width: 4;-*-
import traceback
import sys
from time import *

try:
        
        import pg
        import fcgi

        conn = pg.DB(dbname='foodb', user='foouser')

        while fcgi.isFCGI():
                req = fcgi.FCGI()

                print "Content-type: text/html\n"
                print "<pre>"

                if not conn:
                        print 'bad connection<p>'
                try:
                        output = conn.query("select * from test")
                        print `output.getresult()`
                except:
                        traceback.print_exc(file=sys.stdout)

                print "</pre>"

                req.Finish()

except:
        traceback.print_exc(file=sys.stdout)
 
------------------------------------------------------------------

jhefferon at my-deja.com wrote:
> 
> I have CGI scripts that want to write to a database.  If I have that
> every hit causes Python to wake up, and then mySQL to wake up, so that
> the script can write to the database (and then they both go back to
> sleep) then the response is very slow.
> 
> I understand that I can put Python into Apache as a module, so that
> Python will always be awake.  I want to also keep mySQL up (maybe
> a half-dozen clients).  I believe that I can do this with Perl's
> Apache::DBI, but I'd much to prefer to go with Python.
> 
> Can I?  (I thought that this is a Python issue because Python needs
> to talk to the open connections -- isn't that why Apache::DBI is a
> Perl module?  I looked a bit in the SIG database archives and did a
> few deja queries but I didn't see any discussion of this; sorry if I
> missed a recent one.)
> 
> Thanks,  Jim jim at joshua.smcvt.edu
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From robin at alldunn.com  Sat Jan 29 16:15:18 2000
From: robin at alldunn.com (Robin Dunn)
Date: Sat, 29 Jan 2000 13:15:18 -0800
Subject: Where is python15_d.dll?
References: <8B06D3D65EAA8FF6.B5390C8B3C382830.3DAD830FA4CDC9B9@lp.airnews.net>
Message-ID: <AMIk4.3$foe.17826018@news.randori.com>

"Chris Stromberger"
<chris.stromberger at mmt.bellhowll.com.cut_here.gov.org.com.edu> wrote in
message
news:8B06D3D65EAA8FF6.B5390C8B3C382830.3DAD830FA4CDC9B9 at lp.airnews.net...
> Newbie attempt to embed python in C++, calling Py_Initialize() and
> PyRun_SimpleFile().  Compiles and links, but cannot find
> "python15_d.dll" when it runs.  If I build the release version (using
> MSVC 6), it runs (uses python15.dll, which is present) but crashes.
> Where can I get the debug dll?
>

Build Python from sources in debug mode.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!





From s.schwarzer at ndh.net  Wed Jan 19 21:08:53 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 20 Jan 2000 03:08:53 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse>
Message-ID: <38866E35.F04D4ADE@ndh.net>

Hi Evguenii

Evguenii Smogailov schrieb:
> Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.

I consider it a bad idea, to add synonyms for things that are
already in the language. So _adding_ "greater" to the language (do you
really find "greater" or just "gt" more readable as ">"?) is not
the same as using "or" _instead_ of "||".

Perhaps the ".." operator is something to consider (if it isn't
incompatible with somethings else in the language), but most of
the times "range" does it's job very well.

Stefan


From che at debian.org  Fri Jan  7 21:16:39 2000
From: che at debian.org (Ben Gertzfield)
Date: 07 Jan 2000 18:16:39 -0800
Subject: Python in Corel Linux
References: <947294639.1002448618@news.islandnet.com>
Message-ID: <yttln61b94o.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "J" == J Dukarm <jdukarm at hydracen.com> writes:

    J> I just installed Corel Linux ("standard" version on CD) on a PC
    J> where I can also boot Win98, and as a Windows refugee I was
    J> very pleased to see how easy it was to access my Windows file
    J> system, read floppies, etc.  Only problem is -- it seems that
    J> Corel left Python out of the distribution!

Ack! Bad Corel! They certainly didn't make it easy for developers, eh?

    J> Has anybody out there installed Python in Corel Linux, and if
    J> so are there any tricks to getting it set up and working?

Since Corel Linux is really just Debian Linux in disguise, it comes
with a wonderful tool called "apt-get" that will automatically install
python things for you.

Just run (as root)

# apt-get update
# apt-get install python

I don't know how Corel Linux comes configured, but this *should* work
out of the box. If it doesn't, make sure your system has a file
/etc/apt/sources.list that looks something like:

deb http://http.us.debian.org/debian stable main contrib non-free
deb http://non-us.debian.org/debian-non-US stable non-US

You can even install pgp and ssh this way, using apt-get!

I believe Corel has a GUI front-end to all this, called System Update
or Package Update or something, if you fear the command-line. Dive
through their menus to find it.

Feel free to email me if you need any more help!

Ben

-- 
Brought to you by the letters Y and S and the number 0.
"Elate means having wings."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/


From arad at fc.hp.com  Wed Jan 12 02:31:35 2000
From: arad at fc.hp.com (Arad Rostampour)
Date: Wed, 12 Jan 2000 00:31:35 -0700
Subject: strange CGI problem
References: <85ga10$p8u$1@nnrp1.deja.com> <14459.47851.670815.455938@beluga.mojam.com>
Message-ID: <85hb86$jjq$1@web1.cup.hp.com>

Do you have a:

print "Content-type: text/html\n"

as your first output?

Enjoy,
Arad
--
Arad Rostampour- arad at fc.hp.com
NT Platform and I/O Team
Software and System Development Lab
Hewlett-Packard

Skip Montanaro wrote in message
<14459.47851.670815.455938 at beluga.mojam.com>...
>
>    >> I'm creating a little web-site; for some reason when I view the page
>    >> in Internet Explorer, it's fine, but Netscape shows me the raw html
>    >> as if it's a text file.  What would cause this?
>
>Sounds like IE is looking at the contents and saying, "hmmm... looks like
>HTML", and displaying it.  Make sure your CGI script is spitting out an
>appropriate Content-type header.
>
>




From aa8vb at yahoo.com  Mon Jan 24 10:56:28 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 24 Jan 2000 10:56:28 -0500
Subject: anydbm - a simple question
In-Reply-To: <3diu0lbz8e.fsf@amarok.cnri.reston.va.us>
References: <20000121110715.A184793@vislab.epa.gov> <3diu0lbz8e.fsf@amarok.cnri.reston.va.us>
Message-ID: <20000124105627.A440284@vislab.epa.gov>

Andrew M. Kuchling:
 |Randall Hopper <aa8vb at yahoo.com> writes:
 |>  File "/home/rhh/software/python-1.5.2/lib/python1.5/dumbdbm.py", line 87,\
 |>    in _addval
 |>    f.write('\0'*(npos-pos))
 |>TypeError: can't multiply sequence with non-int
 |
 |I'll bet you're on Solaris and are picking up the large file support,

Actually, no, I'm on SGI IRIX.

 |which makes the .tell() method of file objects return a long integer.
 |But you can't multiply a sequence by a long integer, only a regular
 |integer:

Yes, that's it.  But apparently it's not just a Solaris issue.

On IRIX:

    >>> open( "/etc/passwd", "r" ).tell()
    0L

 |Short-term, you can patch dumbdbm.py to use int(npos-pos), but this is
 |really an interpreter bug; there's no reason 2L * 'a' should be an
 |error.

Ok, thanks, I'll do that.

On a related not, another rough spot in non-uniform long int support
cropped up in another thread just recently:

    >>> str( 0x80000000L )
    '2147483648L'
    >>> "%ld" % 0x80000000L
    OverflowError: long int too long to convert

Hopefully long int support is on the Python v2 list of wrinkles to smooth.

-- 
Randall Hopper
aa8vb at yahoo.com



From fredrik at pythonware.com  Fri Jan 21 07:17:14 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 13:17:14 +0100
Subject: Screen depth
References: <2B1262E83448D211AE4B00A0C9D61B03BA707A@MSGEURO1>
Message-ID: <003d01bf6409$74785670$f29b12c2@secret.pythonware.com>

Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> I'm working on Windows NT and 95/98.

using Tkinter?  wxPython?  gtk?  PythonWin?

for Tkinter, the answer is here:
http://www.pythonware.com/library/tkinter/introduction/basic-widget-methods.htm
look for winfo_depth under "window related information"

(or you can use an undocumented feature in PIL:
Image.core.display_mode() returns a tuple with mode
and size).

</F>




From news at dorb.com  Mon Jan 17 13:36:26 2000
From: news at dorb.com (Darrell)
Date: Mon, 17 Jan 2000 13:36:26 -0500
Subject: If you're still looking for a room for the Python Conference...
References: <VrDQRddcbbs9-pn2-TpsNnUI9gtvg@Fortress>
Message-ID: <2gJg4.6028$U5.217147@typhoon.nyroc.rr.com>

I also have a room from the Jan24-26th and would split it.
--
--Darrell
"Jeff Rush" <jrush at summit-research.com> wrote in message
news:VrDQRddcbbs9-pn2-TpsNnUI9gtvg at Fortress...
> I've got one and am looking to share with someone, to reduce the costs.
If
> you're
> interested, drop me a line and/or come participate in the web-based
> bulletin
> board for other travel and hotel accomodations at:
>
>
http://starship.python.net/crew/scharf/GroupMind/GroupMind.cgi?req=show&fil
> e=ConferenceStuff.gromi
>
> Looks like a great conference!
>
> -Jeff Rush
>
> --
> http://www.python.org/mailman/listinfo/python-list




From clarence at netlojix.com  Mon Jan 31 14:10:14 2000
From: clarence at netlojix.com (Clarence Gardner)
Date: Mon, 31 Jan 2000 11:10:14 -0800 (PST)
Subject: Why Perl do But Python can't?
In-Reply-To: <001401bf6c16$d5657660$3acbd9c2@peridot.optichrome.com> from "Adrian Eyre" at Jan 31, 2000 06:13:11 PM
Message-ID: <200001311910.LAA11870@liberty.avtel.silcom.com>

> > This is not true, and is a myth people need to get over.  The danger
> > Kelly referred to exists on some Unix-like systems and not on others.
> > On a system where they are safe, there is no reason whatsoever to use
> > a setuid wrapper.
> 
> Just out of interest, what is the reason why this in unsafe. I couldn't
> see the difference between a script and a binary.

It has to do with how the text of the script is provided to the
interpreter (Perl, Python, etc).  The danger comes in if the operating
system, having started the interpreter and changed the processes
credentials according to the setxid bits, then passes it the name of the
script file to run -- that file may have been changed to a different one
by the time the interpreter opens it.

Safe systems (generally SystemV descendants, like Solaris and Irix, but
not Linux), do not pass the name of the actual script file to the
interpreter.  The operating system opens the script file in the context
of the process, then passes it a filename from the file descriptor
filesystem that refers to it (e.g., /dev/fd/6), which cannot have been
tampered with.



From yduchesne at my-deja.com  Fri Jan 21 18:27:04 2000
From: yduchesne at my-deja.com (yduchesne at my-deja.com)
Date: Fri, 21 Jan 2000 23:27:04 GMT
Subject: CGI and Python
Message-ID: <86apvo$qie$1@nnrp1.deja.com>

Hi everyone,

I used to do server-side web programming using java servlets; but know
I've tried Python; I'm just amazed. I never could develop as fast! I
think I'm converted.

There are some points that I would like to clear out before making the
full jump:

a) I know there are a couple of modules out there that make embedding
Python into Apache possible; does this allow for a programmer to have
Python objects residing in the server's process (i.e. as for Java
servlets) - and be accessible to all Python scripts running on the
server? Can a python "servlet" access other python "servlets"?

c) How is embedded (in Apache) Python performance? Does it compare to
mod_perl or Java servlets?

d) I would like someone(s) to come back to me and tell me about what
module they recommend - PCGI, PyApchache...(please, some -even
telegraphic style- explanations would be appreciated)

Thanks a lot-lot-lot!!!!


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


From gawron at obop.com.pl  Wed Jan 26 09:46:54 2000
From: gawron at obop.com.pl (=?iso-8859-2?Q?Przemys=B3aw?= G. =?iso-8859-2?Q?Gawro=F1ski?=)
Date: Wed, 26 Jan 2000 15:46:54 +0100
Subject: vi and python?
References: <85uc1m$5i8$1@pegasus.tiscalinet.it> <86l7iq024mg@news2.newsguy.com> <388ef786.90332811@news.oh.verio.com>
Message-ID: <388F08DE.97432EED@obop.com.pl>

Kevin Dahlhausen wrote:

> Ben Beuchler <insyte**NOSPAM**@emt-p.org> wrote:
>
> >There seems to be a general preference for emacs in the python community.
> >I am personally much more comfortable working with vim.
> >Does anyone have any particular tips for python development within vim?
>
> It looks like Geocities has some issues write now, but when it comes back up
> you can find some scripts for the embedded Python interpreter in VIM at:
>
> http://www.geocities.com/SiliconValley/Peaks/4673/python.html

May be some one has a built gvim for windows with python embedded ?
Thanks Przemek

--
Przemyslaw G. Gawronski UIN:8358522
mailto:gawronskip at usa.net
mailto:gawron at obop.com.pl





From billtut at microsoft.com  Fri Jan 21 15:03:53 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Fri, 21 Jan 2000 12:03:53 -0800
Subject: olemsg32 module used in win32com exchange server example?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BCC50@RED-MSG-50>


> From: Gordon McMillan [mailto:gmcm at hypernet.com]
> Sent: Friday, January 21, 2000 8:16 AM
> To: aahz at netcom.com; python-list at python.org
> Subject: Re: olemsg32 module used in win32com exchange server example?
> 
> 
> Aahz wrote:
> > Grant Edwards <grant at nowhere.> wrote:
> > >In article <38863DB3.7FAEC820 at callware.com>, Ivan Van 
> Laningham wrote:
> > >>
> > >>I suspect that is true, but there's a gotcha.  Unless I'm 
> wrong, both
> > >>NetScape Mail and Outlook use MAPI dll's, but they expect 
> different
> > >>functions to be found there.  So, when I want to use 
> Outlook to retrieve
> > >>mail, I can configure everything else correctly but I 
> have to change the
> > >>MAPI.DLL to the one the Outlook expects or it barfs.  
> Then put it back
> > >>when I want to use NetScape. ...
> > >
> > ><Boggle>
> > 
> > It's even worse than that.  AFAIK, NT *still* only loads 
> one copy of a
> > DLL, so that even if you have DLLs in two different 
> directories, it is
> > impossible simultaneously to run programs that use a 
> same-named DLL if
> > they need even different versions.  
> 
> Wrong about that. Though 2 different versions of a mapi dll 
> loaded at the same time might yield trouble, it's not because 
> you can't load them. I do it all the time with (different) 
> python15.dlls. 
> 

Esp. wrong about that with Windows 2000 and Windows 98.
Win2k and Win98 added features so that you could determine
which specific DLL (even if its a DLL you only create via CoCreateInstance
(aka COM object))
is loaded for any given application.

See the following URLs for more information:
http://msdn.microsoft.com/library/techart/sidebyside.htm
http://msdn.microsoft.com/library/techart/dlldanger1.htm

Bill




From glen at electricorb.com  Sun Jan 23 19:17:47 2000
From: glen at electricorb.com (Glen Starchman)
Date: 23 Jan 2000 18:17:47 -0600
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000119234852.8351A-100000@sunse> <388670D2.2FA842F6@starmedia.net>
Message-ID: <388B6764.8F51F1D4@electricorb.com>

I REALLY have to side with Andrew on this one... there are thousands of
programming languages out there, each with their own little idiosyncracies. The
neat things about Python is that it IS readable yet relatively compact at the
same time.

Andrew Csillag wrote:

> Evguenii Smogailov wrote:
> >
> > On Wed, 19 Jan 2000, Fredrik Lundh wrote:
> > > > Generally for improving readability,
> > > > (and Perl compatibility).
> > >
> > > oh, come on.  since when is "Perl compatibility" a
> > > guiding principle for Python's design?
> >
> > Well, i mean all that 1 million Perl programmers that would like to learn
> > Python someday:)
>
> Just because I know lisp doesn't mean I can't learn python.  I don't
> find it a defect that python doesn't support prefix notation like lisp
> does (wouldn't it be awful if it did?  Yeesh...).  Python doesn't have
> to be Perl for Perl programmers to be able to learn it.  I bet there are
> a bunch of Perl programmers that even know Java.
>
> >
> > But the MAIN reason is READABILITY
> >
>
> You're saying that Perl is readable? HAHAHAHA... Oh, you were serious...
> Ok, 1 eq 2 is easier to read that 1==2?  Hmmmm... I think not.
>
> Cluttering Python's syntax with additional operators like eq, etc. where
> operators like == already exist is just a bad idea.  If you really want
> it, create a module perl.py that has the following:
>
> def eq(a, b):
>     return a==b
> def gt(a, b):
>     return a > b
> ..etc..
> And then, no one will be able to read your code.
>
> --
> print(lambda(q,p):'pmt:$%0.2f'%(q*p/(p-1)))((lambda(a,r,n),t:(a*r/
> t,pow(1+r/t,n*12)))(map(input,('amt:','%rate:','years:')),1200.0))



From ysamama at cybercable.fr  Thu Jan  6 19:02:50 2000
From: ysamama at cybercable.fr (Yann SAMAMA)
Date: Fri, 7 Jan 2000 01:02:50 +0100
Subject: French references?
Message-ID: <853agi$2hh7$1@news5.isdnet.net>

Hi!

As a system administrator in a french medium-sized company, I would like to
install Python on a wide range of computers (NT, AIX, SUNOS, SOLARIS,
LINUX).

Therefore, I have to convince my boss...


Is there any people here in France or french-speaking countries who have
already gone through this and could give me :
- his/her name,
- the name of his/her company
- a small summary of what he/she builds with Python.

This would *REALLY* helps me to convince him, because he doesn't know
anything about "free software"
and would probably distrust it and consider Python unreliable, unless I give
him some well-known french references...

Thanks in advance,

Yann SAMAMA
ysamama at cybercable.fr
yann.samama at seita.fr
+33 (0)1 45 56 68 24 (office)




From meng.engineering at bluewin.ch  Wed Jan 19 01:34:30 2000
From: meng.engineering at bluewin.ch (Markus Meng)
Date: Wed, 19 Jan 2000 07:34:30 +0100
Subject: [Q] What is needed to use other TclTk Widget's within Python ...
Message-ID: <863m5q$3g$1@bw107zhb.bluewin.ch>

Hey and hello,

I just wonder if it is possible to 'import' other TclTk Widgets
within Python like:

-- HTML Widget
-- BLT Widget
-- TkTable
-- TkPrint
-- BWidgets
-- etc...

Do I need for each of these package an additional wrapper like Tkinter?
Is it possible to write the GUI-Client completly in TclTk and then connect
to the Python server through sockets?
Do we have other possibilities for bringing the 'best' of both worlds
together?

markus

--
********************************************************************
** Meng Engineering        Telefon    056 222 44 10               **
** Markus Meng             Natel      079 230 93 86               **
** Bruggerstr. 21          Telefax    056 222 44 10               **
** CH-5400 Baden           Email      meng.engineering at bluewin.ch **
********************************************************************
** Theory may inform, but Practice convinces. -- George Bain      **








From akuchlin at mems-exchange.org  Fri Jan 14 14:44:52 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 14 Jan 2000 14:44:52 -0500
Subject: Desperate: Linux Sound Module
References: <TFSHIQKY@scigen.co.uk> <20000114132931.A1533@stopcontact.palga.uucp>
Message-ID: <3dzou8saiz.fsf@amarok.cnri.reston.va.us>

Timothy Butler had a Python module for the OSS Linux sound driver at
http://net.indra.com/~tim/ossmodule/ .  I don't know if it's still
maintained; perhaps you could take over maintaining it.

It also looks like /F has an experimental release of something called
the PythonWare Sound Toolkit, at
http://www.pythonware.com/products/pst/index.htm .  I'm sure hacking
assistance would be appreciated.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
    "Mulder... this disk wouldn't contain Pentagon access codes, would it?"
    "Actually it's Doom II. You know how the Bureau frowns on computer games."
    -- Scully and Mulder in X-FILES #3




From quinn at xeno.ugcs.caltech.edu  Tue Jan  4 16:42:37 2000
From: quinn at xeno.ugcs.caltech.edu (Quinn Dunkan)
Date: 4 Jan 2000 21:42:37 GMT
Subject: __rcall__???
References: <000601bf531b$42c51d60$3ca2143f@tim>
Message-ID: <slrn874qco.m0p.quinn@xeno.ugcs.caltech.edu>

On Thu, 30 Dec 1999 18:11:52 -0500, Tim Peters <tim_one at email.msn.com> wrote:
>       Your reply to that was one of the first of the 1,000+ msgs waiting in
>my inbox when I returned, and I answered it before reading all the pending
>msgs.

I guess there are some real advantages to being a bot.

reads-1000-msgs-just-like-that-and-still-insists-on-abbrs


From guido at CNRI.Reston.VA.US  Fri Jan 21 14:01:08 2000
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Fri, 21 Jan 2000 14:01:08 -0500
Subject: Python Conference weather
Message-ID: <200001211901.OAA29501@eric.cnri.reston.va.us>

If you're traveling to the Python Conference, be advised that winter
has finally arrived in the Washington, DC area.  We're currently
experiencing *high* temperatures of 22 degrees F (-6 degrees C); with
the wind chill it will feel much colder.  So be sure to pack warm
clothes.

Yesterday, about 6 inches of snow fell, disrupting air travel; more is
expected on Sunday, so expect delays flying into DC.  Local
transportation should be fully operational, but may experience some
delays.

We've placed a weather advisory in the local section of the conference
website:

  http://www.python.org/workshops/2000-01/local.html

Over 250 people have registered for the conference.  See you all there!

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



From pbible at home.com  Tue Jan  4 17:01:04 2000
From: pbible at home.com (Paul E. Bible)
Date: Tue, 04 Jan 2000 22:01:04 GMT
Subject: Python & Apache
Message-ID: <A4uc4.411$hJ1.3104@news.rdc1.tx.home.com>

Has anyone had any success getting Python CGI scripts to run under the
following configuration?

OS:                Windows/NT
Web Server:   Apache 1.3.9
Python:           v1.5.2





From wclodius at lanl.gov  Thu Jan  6 12:16:55 2000
From: wclodius at lanl.gov (William B. Clodius)
Date: Thu, 06 Jan 2000 10:16:55 -0700
Subject: Python not a Very High-Level Language?
References: <000801bf5802$dcca7d60$0aa0143f@tim>
Message-ID: <3874CE06.9A434C20@lanl.gov>


Tim Peters wrote:
> <snip>
> The first validated Ada compiler was written in SETL -- and that's about all
> anyone seems to remember it for these days.  I'm not sure you can even get a
> viable implementation anymore.  
> <snip>

ftp://cs.nyu.edu/pub/languages/setl2/

appeears to be the most likely source


From ivanlan at callware.com  Thu Jan 20 16:44:20 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Thu, 20 Jan 2000 14:44:20 -0700
Subject: read from binary file
References: <38877de8.0@sia.uibk.ac.at>
Message-ID: <388781B4.58743095@callware.com>

Hi All--

Stefan Andreatta wrote:
> 
> I have encountered severe problems when trying to read large amounts of data
> (unsigned short integer) from a binary file under NT. I use something like:
> 
> f = 'C:\\ai.xxx'
> file = open(f,'r')

file=open(f,'rb')

> buffer = file.read()
> 
> But the read() function always returns just some 1000 bytes of data and the
> stops.
> 
> Does anybody know, what could be the reason for that, and how I could read
> the whole file into a string?
> 

You must use the binary flag ('rb' instead of 'r') on Windoze, because
it cares.  Unices don't.

<tell-it-to-an-os-that-cares>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From nospam at bitbucket.com  Sat Jan 29 03:52:11 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Sat, 29 Jan 2000 00:52:11 -0800
Subject: socket.recv() question
References: <slrn893nst.15sv.pds@uberhacker.org>
Message-ID: <DTxk4.701$S3.29074@newsfeed.avtel.net>

Paul Schmidt wrote in message ...
>Hello all,
>      I am trying to parse data from an rwhois server, and it sends back
>lines terminated by \012.  whenever I do a:
>
>      results = sock.recv(1024)
>
>I only get the first line in results, but I want the entire output, so
>I can split it by lines and do other parsing things with the data.
>
>How would I go about doing this?


The server is presumably spitting out a line at a time.  Try
    sock.read()
or
    sock.readlines()
--
Phil Mayes    pmayes AT olivebr DOT com
Olive Branch Software - home of Arranger






From cjw at connection.com  Sun Jan  9 11:35:58 2000
From: cjw at connection.com (Colin J. Williams)
Date: Sun, 09 Jan 2000 11:35:58 -0500
Subject: control structures (was "Re: Sins")
References: <1264769486-16171098@hypernet.com> <38788F7E.76302733@maxtal.com.au>
Message-ID: <3878B8EE.B4939F3A@connection.com>


skaller wrote:
> 
> Gordon McMillan wrote:
> 
> > > YUKKKKKK!!! There is no natural way to do this.
> > > We fall through the loop on failure, instead
> > > of success. (using a flag, or testing anything
> > > afterwards, is not acceptable as a solution).
> >
> > That's the reason that Python has the "for
> > .. else" construct that William Tanksley
> > hates so much.
> >
> >  for i in X:
> >    if X[i]==e:
> >      # do the "found" thing
> >      break
> >  else:
> >    print "Not Found"
> 
> That doesn't work. The 'do the "found" thing'
> is in the wrong place: it must be outside the loop,
> since it is usually the 'normal continuation'
> of the routine.

Aside from that.  It seems to me that the weakness of the
   for ...
   else 
structure is that the else clause is executed whether the 
'for' succeeds or not. The Language Reference Manual has:

	"The expression list is evaluated once; it should yield a     
sequence. The suite is then executed once for each item in the 
sequence, in the order of ascending indices. Each item in turn 
is assigned to the target list using the standard rules for 
assignments, and then the suite is executed. When the items are 
exhausted (which is immediately when the sequence is empty), the 
suite in the else clause, if present, is executed, and the loop 
terminates."

Illustrative code is appended.

More useful would be to execute the else clause only if the sequence is
empty.  This might be expressed as:

    for x in L:
	...
    otherwise ...

Colin W.

t= [1,2]
for i in t:
  print i
else: print 'else'  
t= []
for i in t:
  print i
else: print 'else'




From boud at rempt.xs4all.nl  Sun Jan  2 13:54:07 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 2 Jan 2000 18:54:07 GMT
Subject: MySQL
References: <pj0v6sg2d06cjuih0ofjhnoqjm5lp6q8lr@4ax.com>
Message-ID: <84o6sf$e4j$1@news1.xs4all.nl>

Simon Faulkner <News at titanic.co.uk> wrote:
> Confused - I am!

> I have written a lovely little database stored in MySQL that has a
> Python front end and uses MySQLdb through _mysql on SuSE with MySQL
> 3.27 - Fine.  All on my Sony laptop development machine.

> Now I am trying to put it onto my RH6 production server I try to build
> the MySQLdb-0.1.1 and it reports that it needs _mysql.  Where do I get
> this from?

> If it is part of MySQL then where is it?  I have the client, server,
> benchmarks and devel loaded.

> Is there a simple overall explanation of databases in Python anywhere?

As far as I know, everything is simply in the MySQLdb-0.1.1 tarball -
if you were to unpack it, build it (after checking for local variations,
I had to do some fiddling due to the places SuSE installs stuff in. The
file _mysqlmodule.c defines the _mysql module, really.

Is this database thing of yours available for looking at? I've noticed
a lot of questions about MySQL and Python lately, and having worked on
one or two apps myself, I though about collecting ideas and writing a
short paper or tutorial on the subject - you know, pinching ideas here
and there, digesting and then regurgitating the mess, before presenting
it to the web...

Interesting points in doing stuff with Python and databases are time
and dates (a perennial problem - even when you're simply working with
VB or Delphi and Oracle), numerical formats (ditto, especially when,
after applying the Microsoft millennium patch the number decimal
conversion is broken), and of course decent design. 

-- 

Boudewijn Rempt  | http://denden.conlang.org


From mfletch at tpresence.com  Sun Jan 23 07:12:33 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 23 Jan 2000 07:12:33 -0500
Subject: Win32Installer + Numeric, anyone succeeded?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3096@VRTMAIL>

I've been trying to get a Win32Installer standalone binary created for an
application I've been working on.  However, the application dies with a
memory read failure on what looks like a null pointer when trying to import
multiarray.pyd (which the win32installer has placed in the target
directory).  This is the standard multiarray.pyd from the LLNL distribution
and the "f" release of win32installer.  Has anyone successfully created a
standalone programme which includes Numeric?

Would appreciate any thoughts,
Mike
__________________________________
 Mike C. Fletcher
 Designer, VR Plumber
 http://members.home.com/mcfletch



From tim_one at email.msn.com  Wed Jan 19 23:54:06 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 19 Jan 2000 23:54:06 -0500
Subject: Python Rocks!
In-Reply-To: <slrn88cnpm.7ts.neelk@brick.cswv.com>
Message-ID: <000901bf6302$6134d500$31a2143f@tim>

[Neel Krishnaswami]
> ...
> The only time indentation has been a problem for me is
> when using eval/exec, and encoding programs as strings
> is basically the wrong thing to do anyway. This fact has
> occasionally tempted me to write a PythonGen module along
> the same lines as Robin Friedrich's HTMLgen. It would make
> writing things like code walkers and program transformers
> much more elegant. It seems like an idea obvious enough
> that someone else must have beaten me to the punch, though.

Not that I've seen!  Programs-writing-programs is a twist of mind
conspicuously absent in the Python world.  Long before bytecodehacks,
Michael Hudson & I had some fun exchanging sick little program generators on
c.l.py, with the aim of pouring unreasonable effort into optimization.  You
can get Christian Tismer interested in this by implying he can't do better
<wink>, but nobody else plays this game here (Mike Fletcher gets close, but
stays on the sane side of the line <wink>).

Michael eventually scratched his itch more effectively with bytecodehacks,
which is evolving into a capable system for mucking with the soul of the
semantics (i.e., the Python VM), avoiding the surface syntax entirely.

If you want to pursue it, I bet you could whip up a powerful transformation
system quickly using John Aycock's SPARK framework (which supports tree
transformations via its very general parser -- see his docs for details).

when-programs-make-programs-next-thing-you-know-
    people-will-make-people-ly y'rs  - tim





From aahz at netcom.com  Fri Jan  7 15:30:51 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 7 Jan 2000 20:30:51 GMT
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <852jbk$9tt$1@nntp9.atl.mindspring.net> <3875099A.2CF9333C@mdlink.de>
Message-ID: <855idr$hb3$1@nntp1.atl.mindspring.net>

In article <3875099A.2CF9333C at mdlink.de>, Helge Hess  <helge at mdlink.de> wrote:
>Aahz Maruch wrote:
>> In article <3874CFE5.765A40EA at mdlink.de>,
>> Helge Hess  <helge.hess at mdlink.de> wrote:
>>>
>>>recently I was heavily wondering why __init__ does not allow return
>>>values ! Is there any special reason not to do this ?
>> 
>> Yup.  The return value *has* to be the object instantiated by Python.
>
>No, the return value has to be None, which is the default return value
>as well. The only thing enforcing this from a technical point of view is
>a simple
>
>  if (ret != Py_None)
>    throw exception
>
>in the PyInstance_New() function. The 'fix' is just
>
>  if ((ret == Py_None) || (ret == self))
>    decref(ret), return self
>  else
>    decref(self), return ret
>
>This doesn't break compatibility to any valid Python code.

Not directly, no.  I think it breaks compatibility with Python's
extremely simple object model and I would be surprised if Guido
disagrees with me.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From tilo.schwarz at daimlerchrysler.com  Mon Jan 31 10:45:07 2000
From: tilo.schwarz at daimlerchrysler.com (Tilo Schwarz)
Date: Mon, 31 Jan 2000 16:45:07 +0100
Subject: Compile Problem Imaging-1.0/Tk
Message-ID: <3895AE03.D26FD7CD@daimlerchrysler.com>

Hi all,

(our newsserver was down some days and it seems to me that my first
posting didn't do through to comp.lang.python ... so I try it again.)

I successfully compiled the libImaging subdir of Imaging-1.0, but I
have a problem compiling the Tk subdir (I'm no tk expert). I followed
the install instructions, but I get the error:

> make                                                                          

[snip]

gcc -fpic  -IlibImaging -I/usr/local/include -I/usr/local/include
-DWITH_TKINTER -I/usr/X11R6/include -I/usr/local/include -O2 -m486
-fno-strength-reduce -I/usr/include/python1.5 -I/usr/include/python1.5
-DHAVE_CONFIG_H -c ./Tk/tkImaging.c
./Tk/tkImaging.c: In function `PyImagingPhoto':
./Tk/tkImaging.c:76: warning: passing arg 1 of `Tk_FindPhoto' from
incompatible
pointer type
./Tk/tkImaging.c:76: too many arguments to function `Tk_FindPhoto'
make: *** [tkImaging.o] Error
1                                                 

I couldn't find something about that in an FAQ. 
I have only one tk.h on my system (SUSE 6.2), which is in
/usr/X11R6/include/tk.h and the libs seem to be alright: 
/usr/X11R6/lib/libtk8.0.a
/usr/X11R6/lib/libtk8.0.so       
/usr/lib/libtcl8.0.a
/usr/lib/libtcl8.0.so                                                                         

Now I have in Tk/tkImaging.c:76:

    photo = Tk_FindPhoto(interp, argv[1]);

and in tk.h:

EXTERN Tk_PhotoHandle   Tk_FindPhoto _ANSI_ARGS_((char *imageName));

which obviously doesn't match.


Now in tk8.0.h I found a "good" match:
EXTERN Tk_PhotoHandle   Tk_FindPhoto _ANSI_ARGS_((Tcl_Interp *interp,
                            char *imageName));
but if I include tk8.0.h, I get:

gcc -fpic  -IlibImaging -I/usr/local/include -I/usr/local/include
-DWITH_TKINTER -I/usr/X11R6/include -I/usr/local/include -O2 -m486
-fno-strength-reduce -I/usr/include/python1.5 -I/usr/include/python1.5
-DHAVE_CONFIG_H -c ./Tk/tkImaging.c
In file included from ./Tk/tkImaging.c:42:
/usr/X11R6/include/tk8.0.h:1275: parse error before `Tcl_Obj'
/usr/X11R6/include/tk8.0.h:1468: parse error before `Tcl_Obj'
/usr/X11R6/include/tk8.0.h:1495: parse error before `Tcl_Obj'
/usr/X11R6/include/tk8.0.h:1542: parse error before `Tcl_Obj'
/usr/X11R6/include/tk8.0.h:1551: parse error before `Tcl_Obj'
make: *** [tkImaging.o] Error
1                                                 


Any ideas?

Thanks for any help!


Bye
        Tilo

--------------------
Tilo Schwarz, DaimlerChrysler, Research Center Ulm, FT3/AB
Address:      Wilhelm-Runge-Str.11,  P.O. Box 23 60,  89013 Ulm, 
Germany
Phone:        +49 731 505 2376
Email:        tilo.schwarz at daimlerchrysler.com


From andrew at starmedia.net  Thu Jan  6 15:06:50 2000
From: andrew at starmedia.net (Andrew Csillag)
Date: Thu, 06 Jan 2000 15:06:50 -0500
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com>
Message-ID: <3874F5DA.B73A6039@starmedia.net>

Darrell wrote:
> 
> from filename import *

Actually, that doesn't do it either if the other module wants to
reference vars in the importing namespace.

The way to make it so that the other code can access variables in the
including files namespace is:
exec open('filename').read() 

> 
> Doesn't create another namespace. Avoid this in most cases.
> 
> Use import filename and filename.someVar the "from" was intended to help
> when your running Python interactively.
> 
> --
> --Darrell
> "John Nielsen" <nielsenjf at my-deja.com> wrote in message
> news:8515bs$2vn$1 at nnrp1.deja.com...
> > Hopefully, there's a simple answer to a problem I am having.
> >
> > I am using import to include a file, however, it looks like the module
> > then has another namespace, which means it can't access some variables.
> >
> > Is there a way to simply include the raw text and insert it into your
> > own code, as if it was not in a another file?
> >
> > Thanks for any help,
> >
> > john
> >
> >
> >
> > --
> > nielsenjf at my-Deja.com
> >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> > --
> > http://www.python.org/mailman/listinfo/python-list
> 
> --
> http://www.python.org/mailman/listinfo/python-list

-- 
import string,sys;t=string;f=t.index;j=t.join;print(lambda
s:j(map(lambda
x,s=s:x[:f(x,':')+2]+j(map(lambda y:'%02X'%ord(y),x[f(x,':')+2:]),'
'),map
(lambda i,s=s:("%04X: "%
i)+s[i:i+16],range(0,len(s),16))),'\n'))(open(sys.
argv[1]).read())



From cool at stuff.com  Sat Jan 22 01:22:42 2000
From: cool at stuff.com (cool at stuff.com)
Date: Sat, 22 Jan 2000 00:22:42 -0600
Subject: Make money surfing the web!             12607
Message-ID: <220100002242@stuff.com>

Everyone likes making easy money.  This place pays you 50 cents for every hour you browse the web!!!
Get even more if you refer people!  I just got a check for $36.00!!!

Sign up here:
http://www.alladvantage.com/go.asp?refid=IMY467






0HZ$<&g=ca


From michael.stroeder at inka.de  Mon Jan  3 19:34:35 2000
From: michael.stroeder at inka.de (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Tue, 04 Jan 2000 01:34:35 +0100
Subject: CGI Script using Python
References: <D1bc4.40$hJ1.1161@news.rdc1.tx.home.com>
Message-ID: <3871401B.C6A817A4@inka.de>

"Paul E. Bible" wrote:
> 
>     print "Content-type: text/html\n\n"
                                      ^^
One too much.

> [..]
> The Apache error log indicates a 'Premature end of script headers:
> /usr/local/apache/cgi-bin/test.py'.

The print statement already sends one \n.

Should be

  print "Content-type: text/html\n"

to print exactly one single empty line between HTTP header and HTML
data. Like this:
-----------
Content-type: text/html

<HTML>
-----------

Ciao, Michael.


From janhansen at my-deja.com  Mon Jan 17 06:04:32 2000
From: janhansen at my-deja.com (janhansen at my-deja.com)
Date: Mon, 17 Jan 2000 11:04:32 GMT
Subject: Python DIFF to track changes in HTML pages anyone?
Message-ID: <85usvr$1ps$1@nnrp1.deja.com>

Hello,
I need to track changes in HTML pages and to return
meaningful (to humans) pieces of information.
In html pages there is often added a new piece (or packet) of
information but the "string" concept of common diff sort of has no
meaning. crlf' s in html is not logical connected with "packets" of info
in html.

anyone now of python code for this?
( I know about Peters ndiff)

Best Regards,
Jan



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


From jblazi at netsurf.de  Fri Jan 21 13:18:36 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Fri, 21 Jan 2000 19:18:36 +0100
Subject: function "overloading"
References: <3888787d_2@goliath.newsfeeds.com> <20000121173150.A1498@stopcontact.palga.uucp>
Message-ID: <3888a12e_3@goliath.newsfeeds.com>

This was not exctly what I meant.
These classes are not subclasses of each other. What I mean is: I have
essentially the same function from the mathematical point of you
(intersection) and I have to call it with different arguments. Of course I
could code the type information into the classes, so during runtime
intersection could find out whith which types of arguments it has been
called, but maybe there are more elegant possibilities.

Janos Blazi

Gerrit Holl <gerrit.holl at pobox.com> schrieb in im Newsbeitrag:
20000121173150.A1498 at stopcontact.palga.uucp...
> Janos Blazi wrote on 948468281:
> > I have 3 classes called Tline, Tsegment and Tpoint, repectively. I'd
like to
> > write a function "subset" that cheks if one of the instances of the
above
> > classes  is a subset of another instance (as long as this is sensible).
> >
> > I am afraid I cannot overload the name "subset" and additionally when
subset
> > has control it cannot decide the type of its arguments: Does this mean I
> > have to write several *different* subset functions having different
names?
>
> >>> print issubclass.__doc__
> issubclass(C, B) -> Boolean
>
> Return whether class C is a subclass (i.e., a derived class) of class B.
>
> regards,
> Gerrit.
>
> --
> Please correct any bad English you encounter in my email message!
> -----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
> Version: 3.12
> GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V
PS+ PE?
> Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
> -----END GEEK CODE BLOCK-----
>




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From erwin at andreasen.com  Sun Jan 30 11:44:26 2000
From: erwin at andreasen.com (Erwin S. Andreasen)
Date: Sun, 30 Jan 2000 17:44:26 +0100
Subject: Embedding -- getting error/traceback information not to stderr? 
 (Summary)
References: <slrn88oujk.ebi.erw@erw-pc.dde.dk> <87d7qqrynl.fsf@parus.parus.no>
Message-ID: <38946A6A.A91E6A8A@andreasen.com>

Jon K Hellan wrote:

> erw at dde.dk (Erwin S. Andreasen) writes:
> 
> > In my application using embedded Python, I'd like to print out the error
> > information as it occurrs (traceback, etc.). This is what PyErr_PrintEx()
> > (called from PyErr_Print()) does -- unfortunately, it does it to sys.stderr.

Thanks to everyone for the responses. I ended up replacing sys.stderr with
a cStringIO object for now. At the beginning of the application, sys.stderr
is replaced and a reference to the object is saved in a C variable. When
the C code evaluates some user-entered code and PyRun_String returns
NULL, PyErr_Print() is called, the value within the stream is fetched with
its getvalue method and then reset() and truncate() are called to
make ready for the next error.

I tried running a Python function using the exception module after an
error occurred, but it seems that no further code can be compiled 
after a syntax error exception, until that exception is cleared. 
I guess I could have gotten around this by precompiling the function 
to do traceback.

I may have to end up doing something more complicated anyway, to provide
proper display of the source lines -- the pathnames of the files
as seen by the user are not the same as they are as stored on disk, and
AFAIK, Python does not store the source together with the compiled code
but rather fetches it from disk again when printing out the error.
Plus I plan on doing some preprocessing of the Python code to allow
for some special extensions.

I found after posting that this has been discussed on the newsgroup
in the past 3 years (the topic seems to reappear every 3-4 months 
or so :) -- you will find several discussions on retrieving traceback 
information from C by searching for PyErr_Fetch on Deja.

-- 
=======================================================================
<erwin at andreasen.com>      Herlev, Denmark       UNIX System Programmer
<URL:http://www.andreasen.org/>      <*>              Goodbye, Richard!
=======================================================================


From NuclearZen at home.com  Sun Jan  2 03:55:34 2000
From: NuclearZen at home.com (NuclearZen)
Date: Sun, 02 Jan 2000 08:55:34 GMT
Subject: comparing text strings
Message-ID: <aoEb4.1909$cV.20812@news.rdc1.wa.home.com>

this is probably the stupidest Q's of all time but  Idont know how to
compare text strings in python.  How?  I try to do it like C, or perl but???




From 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net  Sun Jan 23 13:30:50 2000
From: 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net (24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net)
Date: Sunday, 23 Jan 2000 12:30:50 -0600
Subject: DARREN MULLIGAN SEEKING MODELS! 56248
Message-ID: <23010012.3050@adsl-63-197-143-124.dsl.scrm01.pacbell.net>

My name is Darren Mulligan and I am an aspiring photographer from California.
I will get right to the meat and gravy of this post, so as not to waste your time!

+++++WHAT AM I SEEKING?+++++

I am looking for young models (prefer early teen 14-16 year old female) for nude and semi-nude photography. This is a great way for me to update my portfolio, and also a great way for you to get a free portfolio, too! This could lead to employment in the Adult XXX entertainment business. What a great way to make money!

Please respond to one of the following e-mail addresses. I run my very own mail server on an ADSL connection so you can be assured to complete discretion and privacy. If one e-mail address does not work, simply try the next! 

Sorry for the long list! My ADSL connection goes down from time to time, so I posted all the e-mail addresses I can be reached at. If one does not work, just move on to the next!

nanae at adsl-63-197-143-124.dsl.scrm01.pacbell.net
usenet at adsl-63-197-143-124.dsl.scrm01.pacbell.net
24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
news.admin.net-abuse.email at adsl-63-197-143-124.dsl.scrm01.pacbell.net
spamlinkpage at adsl-63-197-143-124.dsl.scrm01.pacbell.net
alt.make.money.fast at nntp.8m.com  nanas at nntp.8m.com   
  hostmaster at FREESERVERS.COM

THIS IS NOT A SCAM! IT IS NOT CHILD PORN!!
There is nothing wrong or illegal about the artistic representation of a young body. If you have a problem with this post, please e-mail me and I will not post to your newsgroup again!

Sincerely,
Darren Mulligan



From khowe at performance-net.com  Wed Jan 12 15:26:55 2000
From: khowe at performance-net.com (Kevin Howe)
Date: Wed, 12 Jan 2000 16:26:55 -0400
Subject: JPython org.python.modules Problem
Message-ID: <9u5f4.203555$5r2.551761@tor-nn1.netcom.ca>

I found the following thread in the Python Mailing List archives (June 99).
It describes the Jpython problem that I as well as serveral others in the
comp.lang.python newsgroup are having trouble with. It includes details on
how to patch jpython's "jar.py" and "main.py" to fix the problem. I haven't
been able to get the patch to work, I am hoping that someone will be able to
figure this out and post the working version.

ciao,
kh

-------- Patching JPython Thread -----------

>From bwarsaw at cnri.reston.va.us Tue, 8 Jun 1999 18:36:56 -0400 (EDT)
Date: Tue, 8 Jun 1999 18:36:56 -0400 (EDT)
From: Barry A. Warsaw bwarsaw at cnri.reston.va.us
Subject: [JPython] Problems with HelloWorld applet under 1.1

>>>>> "DP" == Darren Platt <daz at exelixis.com> writes:

    DP> I installed 1.1 a few days ago and have been trying to get
    DP> basic things working, under NT, and the jdk1.2 (after some
    DP> nasty jre crashes using jre 1.1.8).  To get to the point,
    DP> trying to compile a simple HelloWorld applet;

    DP> Compiling goes well:

    DP> But running the applet under the jdk applet viewer

    DP> Gives a traceback:

    DP> Looking at the java for the core Jpython code, it refers to
    DP> org.python.modules.Setup in org.python.core.PySystemState, and
    DP> that code doesn't get included in the jpython.jar and not into
    DP> the created hw.jar file, and hence doesn't run :(

    DP> Any ideas ?

Yes, that's a bug.  Attached is a fix, although I'm not 100%
comfortable with it.  Perhaps Setup should be moved to
org.python.core?  I'd rather leave it in org.python.modules though...

(I'm still digging into jpythonc so this might change)

-Barry

-------------------- snip snip --------------------
Index: jar.py
===================================================================
RCS file: /projects/cvsroot/jpython/dist/Tools/jpythonc2/jar.py,v
retrieving revision 2.0
diff -c -r2.0 jar.py
*** jar.py 1999/05/19 20:11:59 2.0
--- jar.py 1999/06/08 22:31:00
***************
*** 23,29 ****
--- 23,33 ----
          self.entries = []
          self.packages = packages
          self.manifest = []
+         self.singleclasses = []

+     def addSingleClass(self, class_):
+         self.singleclasses.append(class_)
+
      def addFile(self, rootdir, filename):
          self.files.append( (rootdir, filename) )

***************
*** 72,83 ****
--- 76,102 ----
          for package, skiplist in self.packages:
              self.addPackage(package, skiplist)

+         for class_ in self.singleclasses:
+             self.addOneClass(class_)
+
          self.dumpManifest()

      def dump(self, filename):
          self.zipfile = ZipOutputStream(FileOutputStream(filename))
          self.dumpFiles()
          self.zipfile.close()
+
+     # add just one class from a package
+     def addOneClass(self, pkgclass):
+         parts = pkgclass.split('.')
+         package = string.join(parts[:-1], '.')
+         pkg = lookup(package)
+         filename = os.path.join(pkg.__path__[0], parts[-1]) + '.class'
+         entryname = string.join(parts, '/') + '.class'
+         self.zipfile.putNextEntry(ZipEntry(entryname))
+         instream = FileInputStream(filename)
+         copy(instream, self.zipfile)
+         instream.close()

      # The next three methods handle packages (typically org.python.core,
...)
      def addPackage(self, package, skiplist=[]):
Index: main.py
===================================================================
RCS file: /projects/cvsroot/jpython/dist/Tools/jpythonc2/main.py,v
retrieving revision 2.2
diff -c -r2.2 main.py
*** main.py 1999/06/08 20:14:28 2.2
--- main.py 1999/06/08 22:29:58
***************
*** 158,164 ****
      if options.jar is not None:
          options.deep = 1
      elif options.core or options.all:
!         opts.deep = 1

      if not os.path.isabs(options.workdir):
          options.workdir = os.path.join(".", options.workdir)
--- 158,164 ----
      if options.jar is not None:
          options.deep = 1
      elif options.core or options.all:
!         options.deep = 1

      if not os.path.isabs(options.workdir):
          options.workdir = os.path.join(".", options.workdir)
***************
*** 265,270 ****
--- 265,273 ----

      for dep in comp.trackJavaDependencies():
          ja.addEntry(dep)
+
+     if opts.core or opts.all:
+         ja.addSingleClass('org.python.modules.Setup')

      ja.dump(jarfile)











From backov at nospam.csolve.net  Tue Jan 11 16:20:45 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Tue, 11 Jan 2000 21:20:45 GMT
Subject: Please, someone tell me what's going wrong here. ;>
References: <GzLe4.54805$Ke.42098@news21.bellglobal.com> <87hfgkl388.fsf@baal.winnegan.de>
Message-ID: <N8Ne4.54869$Ke.42764@news21.bellglobal.com>

Siggy Brentrup <bsb at winnegan.de> wrote in message
news:87hfgkl388.fsf at baal.winnegan.de...
> Hi Jason,
>
> your code looks like Borland C++ on WinXX to me, you'd better tell us
which
> environment you're using :)

Yep, Borland C++ Builder 4.0 on Win98.

> "Jason Maskell" <backov at nospam.csolve.net> writes:
>
> > I posted a while back on embedding, and after scouring deja and reading
the
> > demos, I can't even get this simple example to work. It's essentially a
C++
> > Builder'ized version of the simple embedding demo.. PyRun_SimpleString
> > returns -1 and foo is never executed.
> >
>
> [...]
>
> extern "C" {
>
> > void initxyzzy(); /* Forward */

Uh, don't see why it would be necessary.. this function is executing fine,
it's just PyRun_SimpleString is failing. I'll try the extern C bit though.

Thanks,
Jason





From aahz at netcom.com  Wed Jan 26 22:50:14 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 27 Jan 2000 03:50:14 GMT
Subject: Another newbie question
References: <000001bf651d$f05ac240$132d153f@tim> <_nOj4.894$yY6.13534@news.goodnet.com>
Message-ID: <86of9m$i0m$1@nntp1.atl.mindspring.net>

In article <_nOj4.894$yY6.13534 at news.goodnet.com>,
Scott Johnson  <sjohnson at csnet.net> wrote:
>
>I've read the tutorial and I don't get lambda.

So don't use it!  I've been programming for more than twenty years, and
I don't get lambda, either.  (Not that I've tried very hard, mind you;
functional programming doesn't do much for me.)  Even Guido admits that
lambdas may have been a mistake, so I'd strongly suggest that you
concentrate on learning the rest of Python first.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From steve_w at localhost.localdomain  Sun Jan 30 00:14:33 2000
From: steve_w at localhost.localdomain (steve_w at localhost.localdomain)
Date: Sun, 30 Jan 2000 05:14:33 GMT
Subject: [Newbie Problems] AttributeError from CGI script?
Message-ID: <slrn897j95.7pk.steve_w@localhost.localdomain>

I'm trying to use Python in a CGI script for the first time and 
have run into a problem I don't understand.  I get the subject
error message when trying to do a string.replace using one of
the values returned by the form that trigers the script.

The line of code looks like:

line = string.replace(line,"placeholder",form["target"].value)

I get an error message back saying:

Traceback (innermost last):
File "[my script's file name]"
   line = string.replace(line,"placeholder",form["target"].value)
AttributeError: replace

Could someone help me figure out what this error message means?

Thanks in advance,

Steve Wall
steve_w at visi.com


From gerrit.holl at pobox.com  Fri Jan 28 12:08:23 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 28 Jan 2000 18:08:23 +0100
Subject: tabs do WHAT?
In-Reply-To: <949076889.194238@news.adtranzsig.de>; from hei@adtranzsig.de on Fri, Jan 28, 2000 at 04:25:53PM +0100
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <949076889.194238@news.adtranzsig.de>
Message-ID: <20000128180823.A1261@stopcontact.palga.uucp>

Dirk-Ulrich Heise wrote on 949073153:
> What bothers me even more, they don't mind writing functions without the
> function comment headers that shall be used in the project.
> You end up with functions containing not a single char of comment.
> It would be another improvement to Python if it would force you to
> write a doc string for each function of method.

I disagree. If you're writing a quick script, you want to do it
quick. Maybe it's better for large programs, but for small ones, there
will be onle more people who'll think Python sucks.

what-about-obliging-# comments-every-4-lines-ly y'rs - gerrit

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From m.faassen at vet.uu.nl  Wed Jan 19 18:50:01 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Jan 2000 23:50:01 GMT
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu>
Message-ID: <865ij9$ia4$1@newshost.accu.uu.nl>

Paul M <paul.magwene at yale.edu> wrote:
> Dear Pythoneers,

> Inspired by recent postings to this newsgroup I've decided that
> Python, and in particular the entire Python user base, has been going
> about things all wrong.  I've decided I would like to be able to code
> in python something like this:
[snip proposal]

This is nothing new. Back during the last Python conference, me and some
other intrepid non-conference-attenders had the run of the newsgroup,
and we redesigned Python into a more sane language already:

http://x44.deja.com/[ST_rn=ps]/getdoc.xp?AN=410321807&search=thread
[and followups]

And the immensely positive reaction by some famous Pythoneers is here:

http://x23.deja.com/[ST_rn=ps]/viewthread.xp?AN=412546299&search=thread

this-year-I-will-attend-so-someone-else-can-have-the-run-ly yours,

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


From sholden at bellatlantic.net  Fri Jan 14 14:33:58 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Fri, 14 Jan 2000 19:33:58 GMT
Subject: Database to Web
References: <jmsu7sk17q2hg0dbrf3qi8lobednf72m4a@4ax.com>
Message-ID: <387F7A3F.2AC8954F@bellatlantic.net>

Do you mean the HTML renders in an ugly way, or that you don't like
the look of the Python code?  Your program seems an entirely acceptable
way to get the data out, but clearly it won't look too nice to the
user.

For what it's worth, the table rows you are generating aren't well-
formed, since the row contents should be contained between table data
tags.  A minimal improvement would be to replace the

  print Name

statement with

  print "<td>%s</td>" % Name

But a better solution would give each element its own cell:

  for i in range(len(Name)):
    print "<td>%s</td>" % Name[i]

to get a more readable result.  Of course, in any given context you
should know the (fixed) length of the tuples returned for each row,
and thus be able to optimize the len(Name) to a fixed value.

regards
 Steve

Simon Faulkner wrote:
> 
> Has anyone got a good sugestion for the easiest way to display the
> results of an sql query on a web page?
> 
> This works but it's not pretty!
> 
> TIA
> 
> def listdetails():
>   db =MySQLdb.connect(host="localhost",db='BTC',user="root",passwd="")
>   cursor = db.cursor()
>   cursor.execute("SELECT * FROM tblDetail WHERE lngSalesPersonID=30")
>   tuple = cursor.fetchall()
>   total = len(tuple)
>   if total < 1:
>     print "No entries"
>   else:
>     print "Content-type: text/html\n"
>     print "<H3>DETAILS</H3>"
>     print "<P>"
>     print "<table border cellspacing=0 cellpadding=5>"
>     for record in range(total):
>       Name =  tuple[record]
>       print "<tr>"
>       print Name
>       print "</tr>"
>     print "</table>"
>   db.close()
> 
> Simon


From Sara at openhere.com  Fri Jan 21 12:33:56 2000
From: Sara at openhere.com (Sara at openhere.com)
Date: Fri, 21 Jan 2000 12:33:56 -0500
Subject: Your site has been included on OpenHere
Message-ID: <M2000012113091501162@openhere.com>

Hi,

Your site has been included in the OpenHere.com index and search engine.  OpenHere is one of the 10 largest index and search sites on the Internet.  

Your site listing is:
Link:  http://www.python.org/download/Contributed.html#Collections
Title:  Contributed Python Software
Description:  


You might have already received a message similar to this one, as we send a note when ever we add a link to your site in a different category on OpenHere.com.

At OpenHere you can dynamically modify your site's listing at any time.  You can also include your site's listing in other categories on OpenHere.com.  

When you modify your site's listing, it is automatically placed at the top of the category in which it is included, and is placed first in the search engine results for the keywords relating to your site.

To modify, add or delete your listing:

1.  Go to the page on OpenHere where your site is presently listed or you would like it listed.
2.  Select "Suggest a Site".
3.  Follow the instructions for changing your listing.

All of the modifications you submit to OpenHere.com are processed in real time.  As soon as you see the response to your submission, your site listing should be updated.

www.OpenHere.com is frequented by both children and families.  As a result, www.OpenHere.com does not include links to material which is illegal to display to minors.

Sara
www.OpenHere.com
Your key to the Net!



From jam at quark.emich.edu  Sat Jan  8 10:18:08 2000
From: jam at quark.emich.edu (Jeff)
Date: Sat, 8 Jan 2000 10:18:08 -0500
Subject: Error using string.strip
In-Reply-To: <38775227.2CC0D685@worldnet.att.net>; from codeslave@worldnet.att.net on Sat, Jan 08, 2000 at 10:05:11AM -0500
References: <38775227.2CC0D685@worldnet.att.net>
Message-ID: <20000108101808.A11832@quark.emich.edu>

On Sat, Jan 08, 2000 at 10:05:11AM -0500, Eric Hausig wrote:
> Hello there,
> 
> I'm very new to Python.  (In fact, I'm walking through the tuturial
> right now!)
> 
> I'm getting the following message:
> 
> >>> string.strip('str') + 'ing'
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> NameError: string
> >>>
> 
> I'm running Redhat 6.1 and have installed all of the dev rpms off the
> CD.   Is there some library I have to manually include?  If someone
> could give me a little insight, I would be thankful.  (PS - I'm
> currently a VB/SQL programmer trying to make the switch to Linux.  =)
> 
> Sincerely,
> 
> Eric Hausig
> aka Codeslave2k
> 


you need to 'import string' (import the 'string module') before you will
have access to the 'strip' function:

[jam at toast ~]$ python
Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import string
>>> string.strip('str  ')+'ing'
'string'

have fun! ;)

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From mfletch at tpresence.com  Wed Jan 26 10:54:37 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 26 Jan 2000 10:54:37 -0500
Subject: To find a point inside a polygon
Message-ID: <CC59868A0A76D311B1F50090277B321C1F30BA@VRTMAIL>

Hmm, something like:

def inside( point, array ):
	draw line to infinity from point, counting intersections
	return intersections % 2 # odd/even

I think graphic gems has a decent implementation.  You have to figure out
whether on-edge means inside or outside.  I seem to recall that there's some
special casing for edges which are equal to your test ray (I think it's
considered a "hit"), but it's been a long time since I idly browsed that
stuff.

HTH,
Mike

-----Original Message-----
From: David Yeung [mailto:dyeung at ust.hk]
Sent: Wednesday, January 26, 2000 10:10 AM
To: python-list at python.org
Subject: To find a point inside a polygon


Does anyone know if any python implementation for finding whether a point
is inside a polygon? E.g., by giving a point(x,y) and an array of polygon's
points (x0,y0,x1,y1,x2,y2,...), it returns true or false to tell if the 
given (x,y) is inside the polygon.

Thanks

david
-- 
http://www.python.org/mailman/listinfo/python-list



From zakons at objectsbydesign.com  Sun Jan 16 12:16:15 2000
From: zakons at objectsbydesign.com (Stuart Zakon)
Date: Sun, 16 Jan 2000 17:16:15 GMT
Subject: bdb: which class?
References: <20000112164417.4994.qmail@web805.mail.yahoo.com>
Message-ID: <85sucm$mht$1@nnrp1.deja.com>

In article <20000112164417.4994.qmail at web805.mail.yahoo.com>,
  Stuart Zakon <sjz18 at yahoo.com> wrote:
>
> My gosh! I was afraid it was going to be that complicated!
> If I had any spare time I would look into why there couldn't just be
the
> following assignment to get the class:
>
> class_name = frame.f_code.co_class_name
>
> Then you could check to see if it was 'None' to indicate that the
method is not
> on a class but on a module.
>


# from the following it appears that the 'node' n contains the
# type information to give away the parent's identity as a classdef
# (versus a funcdef or a lambdef)
# all that is needed now is the name of the class to pass down to
# the PyCode_New function, which creates the code object.
# the result would be (co_name, co_classname, co_filename), more
# than enough to identify which method in a class is being called.
# is the name of the class readily available from the node??

# from jcompile() (compile.c, line 3409)

        compile_node(&sc, n);
        com_done(&sc);
        if ((TYPE(n) == funcdef || TYPE(n) == lambdef) &&
            sc.c_errors == 0) {
                optimize(&sc);
                sc.c_flags |= CO_NEWLOCALS;
        }
        else if (TYPE(n) == classdef)
                sc.c_flags |= CO_NEWLOCALS;

<snip, snip>

        co = PyCode_New(sc.c_argcount,
                         sc.c_nlocals,
                         sc.c_maxstacklevel,
                         sc.c_flags,
                         sc.c_code,
                         consts,
                         names,
                         varnames,
                         filename,
                         name,
                         sc.c_firstlineno,
                         sc.c_lnotab);



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


From akituc at pk.highway.ne.jp  Tue Jan 25 16:14:53 2000
From: akituc at pk.highway.ne.jp (Akinori Tsuchida)
Date: Wed, 26 Jan 2000 06:14:53 +0900
Subject: Need Preprocessor Module!
Message-ID: <388E124C.B6B9BF9A@pk.highway.ne.jp>

I would like to use preprocessor-like functionality in Python
like we use in C/C++.

Many corresponding functionalities are available in Python, but
 I think preprocessor is still useful.

For example, you can comment multiple lines just by "if 0" and "endif"
with no need to consider indenting.
You don't have to take care of not embracing another comment section, I
mean...

#if 0
#if 0
   not required part.
#endif
#endif

Above is treated as expected but

  """
  """
  not required part.
  """
  """
not the case with above. (unwanted part still remains.)



Well, Any module available ?
like,... input strings containing preprocessor lines
and then output processed strings.

#ifdef YOU_KNOW
#include INFO
Please_Answer(INFO)
#endif

Thanks for reading this article.

Akinori Tsuchida.



From loewis at informatik.hu-berlin.de  Wed Jan 12 15:17:26 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 12 Jan 2000 21:17:26 +0100
Subject: Tricky interpreter initialization question..
References: <ayqe4.51620$Ke.35799@news21.bellglobal.com>
Message-ID: <p6q7lhfhwo9.fsf@informatik.hu-berlin.de>

"Jason Maskell" <backov at nospam.csolve.net> writes:

> Initialize the Python interpreter and then call one or a series of
> init procedures that calls Py_InitModule(), without having used
> "import spam" to create the module object. Can I get this to work?

I'm not sure why you want to do this. What is the purpose of
initialising a module that you are not going to use?

> I figure it should be possible to initialize these modules at
> run-time and thus get around the first two problems. Writing the
> glue code isn't a hardship, but again, the interpreter needs to be
> pointed to already instantiated interfaces, not stuff it has to
> instantiate itself.

If they are extension modules, they do have init* functions, right?
If not, why not?

Just put all these into an array _PyImport_Inittab (or extend the
existing array by PyImport_AppendInittab), and voil?, you have these
modules available to an application.

Regards,
Martin


From dieter at handshake.de  Thu Jan 20 14:34:46 2000
From: dieter at handshake.de (Dieter Maurer)
Date: 20 Jan 2000 20:34:46 +0100
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p>
Message-ID: <x7ogag1qqx.fsf@lindm.dm>

Edward Welbourne <eddyw at lsl.co.uk> writes on Thu, 20 Jan 2000 14:39:31 +0000:
> Personally, I consider the two-namespace rule to be a
> major *strength* of python: it is a major ingredient in pythonicness.
Me and all my colleagues using Python have already been bitten by this
ingredient -- mostly in connection with "lambda", rarely in connection
with methods ("self" forgotten, module variable used).
I would like to live without it.

> I contend that the namespace in which a function executes its suite
> *should* be discardable at the time the function returns.  Lexical
> scoping would mean that we can't discard foo's namespace when foo
> finishes execution.  The implications of this for garbage collection are
> singularly unpleasant - give them some thought.
As you pointed out in the following paragraph (which I deleted),
the compiler could do automatically, what, now, I must do manually
when I need to access the enclosing objects:
	def bar(...., a=a, b=b, ...)
I admit, that this does not give a true lexical scoping semantics, though
(rebindung of "a" is local to "bar").
I think (but am not sure) that I would be satisfied
with this semantics.

> ...
> but to pythoneers, the two-namespace rule is harmonious,
> beautiful and kind to the garbage-collector.  Treat the humblest of thy
> servants kindly, lest they rise up and smite thee with just zeal.
I feel me as a pythoneer, but nevertheless the two-namespace rule is not
beautiful for me, nor is grouping by indentation.


Dieter


From aahz at netcom.com  Wed Jan 19 17:44:06 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jan 2000 22:44:06 GMT
Subject: Mapping Column Names to results using DBI
References: <38863BEB.BCB9E15C@exceptionalminds.com>
Message-ID: <865enm$kkt$1@nntp4.atl.mindspring.net>

In article <38863BEB.BCB9E15C at exceptionalminds.com>,
Timothy Grant  <tjg at avalongroup.net> wrote:
>
>As you know, the results from a database query are returned in a tuple.
>I can access those results by position just fine, is it possible to map
>the column names onto the results, so I can access columns by name
>rather than by number?
>
>While indexing by number works, it is incredibly fragile and will start
>failing the minute the underlying DB changes.

Try "SELECT x, y, z" instead of "SELECT *".
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From neelk at brick.cswv.com  Fri Jan  7 19:54:36 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 8 Jan 2000 00:54:36 GMT
Subject: [ANN] Multimethod.py -- multimethods for Python
Message-ID: <slrn87d2lv.i6r.neelk@brick.cswv.com>

[This is a repost, since the original apparently never made it; if you
see it twice, my apologies.]

I'd like to announce the existence of Multimethod.py, a package that
implements real (eg, CLOS/Cecil/Dylan) multimethods for Python. These
are nifty for oodles of reasons, but some of the more significant are:

  o They make the Visitor and Strategy patterns trivial to implement.
  
  o The whole "binary method/double dispatch" problem goes away, thus 
    promoting loose coupling and other software-engineering buzzwords.
  
  o They resolve the whole contravariance/covariance debate so totally
    it's hard to remember what the problem was.

You can find it on my webpage at:

  <URI:http://www.sff.net/people/neelk/open-source>

It should work out of the box on CPython 1.5.2, and Jpython if Jpython
has lists that support the insert() method. I don't know about Viper,
since Viper apparently has a rather different type system from vanilla
Python, but it should be easy to port, though.

The current version is quite rough, with lame algorithms used in a
couple of places and very substandard documentation. But it all works,
and is quite usable for experimentation and hacking.

(Incidentally, is this the sort of thing that should go on the
types-sig?)


Neel


From alex at somewhere.round.here  Sat Jan  8 23:16:20 2000
From: alex at somewhere.round.here (Alex)
Date: 08 Jan 2000 23:16:20 -0500
Subject: Hope this is the mailing list
References: <005601bf5a52$1a37dcf0$0000fea9@wyvern>
Message-ID: <etdaemfg9rf.fsf@w20-575-109.mit.edu>

> how i can extract this four variable
> gecko
> jing at kirk.com
> toad
> ray at lex.com
> from this two
> "gecko" <jing at kirk.com>
> "toad" <ray at lex.com>
> by using regular expression????

Well, I guess this is using regular expressions somewhere:

>>> import rfc822
>>> b='''"gecko" <jing at kirk.com>
... "toad" <ray at lex.com>'''
>>> a=rfc822.AddressList(b)
>>> a.addresslist
[('gecko', 'jing at kirk.com'), ('toad', 'ray at lex.com')]
>>> reload(rfc822)
<module 'rfc822' from '/usr/lib/python1.5/rfc822.pyc'>
>>> 

If you look in rfc822.py (on my machine, it would be in
/usr/lib/python1.5/rfc822.py) then you can probably work it out.

Alex.


From robin at jessikat.demon.co.uk  Sat Jan  8 08:04:42 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 8 Jan 2000 13:04:42 +0000
Subject: bsddb item size?
References: <mrE03AA5nOd4Ew2R@jessikat.demon.co.uk>
 <wzd7rd4kcd.fsf@sunshine.cs.uu.nl> <1y9ghGAe1nd4Ewk9@jessikat.demon.co.uk>
 <wzpuvcn3pt.fsf@sunshine.cs.uu.nl>
Message-ID: <HnKQ+CAqXzd4EwiV@jessikat.demon.co.uk>

In article <wzpuvcn3pt.fsf at sunshine.cs.uu.nl>, Piet van Oostrum
<piet at cs.uu.nl> writes
>>>>>> Robin Becker <robin at jessikat.demon.co.uk> (RB) writes:
>
>RB> thanks that's a great help. How about collisions are they possible?
>
>yes, you can choose to allow duplicate keys.
thanks very much extremely helpful.
-- 
Robin Becker


From Denys.Duchier at ps.uni-sb.de  Wed Jan 12 16:06:58 2000
From: Denys.Duchier at ps.uni-sb.de (Denys Duchier)
Date: 12 Jan 2000 22:06:58 +0100
Subject: Python not a Very High-Level Language?
References: <000801bf5802$dcca7d60$0aa0143f@tim> <is66x6vng7.fsf@localhost.ps.uni-sb.de> <87u2kljrmz.fsf@den.home.net>
Message-ID: <isbt6rhudp.fsf@grizzly.ps.uni-sb.de>

I apologize for the delay in responding to your question, but our
newsfeed was out to lunch for about a week.

Frank Sergeant <frank.sergeant at redneck.net> writes:

> I think I saw that you have an interface to Tk.  Could you discuss
> GUI building in Mozart a little?

Drat!  Just the thing I know least about; how did you guess?  I am
forwarding this message to Christian Schulte who was responsible for
the development of our Tk interface and who is and has been
supervising a number of students in the general area of GUI
programming in Oz.  According to my agenda, we have a meeting
scheduled tomorrow where the current state of the GTK interface will
be presented.  I am eager to learn about that :-)

The Tk interface follows a fairly simple design: the Oz process starts
a Tk subprocess and communicates with it via a socket.  On the Oz
side, Tk widgets are represented by Oz objects.  The Oz/Tk interface
is actually a very thin OO flavored layer.

To be concrete, here is the functor (a bit like a Python module) for
an application that takes a message argument (using a --msg option)
and displays it in a label together with a quit button:

functor
import Tk Application
define
   Args = {Application.getArgs
           record(msg(single type:string optional:false))}
   proc {Exit} {Application.exit 0} end
   Top = {New Tk.toplevel tkInit}
   Msg = {New Tk.label    tkInit(parent:Top text:Args.msg)}
   But = {New Tk.button   tkInit(parent:Top text:'Quit' action:Exit)}
   {Tk.send pack(Msg But)}
end

Here Top, Msg and But are Oz objects that correspond to the obvious Tk
widgets.

A tutorial for Tk/GUI programming in Oz is available at:

         http://www.mozart-oz.org/documentation/wp/index.html

If you have a question about some more specific aspect of GUI
programming (or anything else for that matter) in Oz, Christian and I
would be delighted to elaborate.

Cheers,

-- 
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 heinkel at dummy.embl-heidelberg.de  Mon Jan 10 06:55:07 2000
From: heinkel at dummy.embl-heidelberg.de (Ralph Heinkel)
Date: 10 Jan 2000 11:55:07 GMT
Subject: purify and python
References: <85cdnt$avq$1@beta.EMBL-Heidelberg.DE>
Message-ID: <85char$baq$1@beta.EMBL-Heidelberg.DE>

ok, after further playing around with it, I recompiled python without
threads. Magically that helped and python is now running correctly.
Strange, but it works.

Ralph



Ralph Heinkel <heinkel at dummy.embl-heidelberg.de> wrote:
> Hi, 

> since I wrote some C, C++, and F77 extensions for python, I am planning
> to use purify to test them for memory leaks and so on.
> The problem is that I cannot get a purified version of python running.
> Python seems to start and immediately exit with error code -1 (255),
> without printing any error message (See below for log). I never get
> to the python command line.

> The way I linked python is shown below. Linking python without purify
> (without any other recompilation) produces a working python version.

> Any clue? 

> I am running python 1.5.2 on IRIX 6.5 (R10000) with purify 4.1. Python
> was configure with  --with-thread  and  --without-gcc

> Ralph

> ---------------------------------------------------------------------

> Modules/Setup is:

> # Pure module.  Cannot be linked dynamically.
> # -DWITH_QUANTIFY, -DWITH_PURIFY, or -DWITH_ALL_PURE
> WHICH_PURE_PRODUCTS=-DWITH_PURIFY
> PURE_INCLS=-I/usr/pure/purify
> PURE_STUBLIBS=-L/usr/pure/purify -lpurify_stubs -lquantify_stubs
> pure puremodule.c $(WHICH_PURE_PRODUCTS) $(PURE_INCLS) $(PURE_STUBLIBS)

> The (python) modules are linked statically, i.e. *shared* not used


> This is how I linked it:

> birdie:src/Python-1.5.2/Modules$ purify -windows=no cc -g  python.o \
>  ../libpython1.5.a -DWITH_PURIFY -I/usr/pure/purify -L/usr/pure/purify \
>  -lpurify_stubs_n32 -lpthread -lmpc -lm  -o python

> ld32: WARNING 84: /usr/lib32/mips4/libmpc.a is not used for resolving 
> any symbol.
> Purify 4.1 IRIX6, Copyright (C) 1992-1997 Rational Software Corp. All 
> rights reserved. 
> .... Processing "/usr/pure/purify/libpurify_stubs_n32.so"
> .... Processing "/usr/lib32/libpthread.so"
> .... Processing "/usr/lib32/libm.so"
> .... Processing "/usr/lib32/libc.so.1"
> .... Processing "python"



> This is the output:

> birdie:src/Python-1.5.2/Modules$ ./python 
> ****  Purify instrumented ./python (pid 63792545 at Mon Jan 10 11:48:44 2000)
>   * Purify 4.1 IRIX6, Copyright (C) 1992-1997 Rational Software Corp. 
> All rights reserved. 
>   * For contact information type: "purify -help"
>   * Command-line: ./python 
>   * Options settings: -purify -windows=no -purify-home=/usr/pure/purify 
>   * Purify licensed to European Molecular Biology La
>   * Purify checking enabled.

> ****  Purify instrumented ./python (pid 63792545)  ****
> Current file descriptors in use: 5
> FIU: file descriptor 0: <stdin>
> FIU: file descriptor 1: <stdout>
> FIU: file descriptor 2: <stderr>
> FIU: file descriptor 26: <reserved for Purify internal use>
> FIU: file descriptor 27: <reserved for Purify internal use>

> ****  Purify instrumented ./python (pid 63792545)  ****
> Purify: Searching for all memory leaks...

> ****  Purify instrumented ./python (pid 63792545)  ****
> SIG: Signal handled:
>   * This is occurring while in:
>         _x_open        [open.s:16]
>         _exit          [syscalls.c:5478]
>         exit_certain   [fork.c:202]
>         _SGIPT_libc_exit [fork.c:253]
>         ptctl          [libcthread.c:149]
>         __exit         [cuexit.c:60]
>   * Received signal 48 (SIGReserved48 - System reserved signal 48)
>   * Handler function: 
>         sched_handler  [vp.c:243]
>   * Signal mask: (SIGReserved48) 
>   * Pending signals: 
> Memory leaked: 0 bytes (0%); potentially leaked: 2048 bytes (1.12%)

> PLK: 2048 bytes potentially leaked at 0x1010c558
>   * This memory was allocated from:
>         _malloc        [malloc.c:886]
>         vp_alloc       [vp.c:1214]

> Purify Heap Analysis (combining suppressed and unsuppressed blocks)
>                          Blocks      Bytes
>               Leaked          0          0
>   Potentially Leaked          1       2048
>               In-Use          4     180236
>   ----------------------------------------
>      Total Allocated          5     182284

> ****  Purify instrumented ./python (pid 63792545)  ****
>   * Program exited with status code -1.
>   * 1 access error, 1 total occurrence.
>   * 0 bytes leaked.
>   * 2048 bytes potentially leaked.
>   * Basic memory usage (including Purify overhead):
>      2179072 code
>       169108 data/bss
>       195436 heap (peak use)
>         4960 stack
>   * Shared library memory usage (including Purify overhead):
>       425984 rtlib_n32.so (shared code)
>        57812 rtlib_n32.so (private data)
>        16384 libpurify_stubs_n32.so.pure (shared code)
>          160 libpurify_stubs_n32.so.pure (private data)
>       212992 libpthread.so.pure.1b7a5a8a8 (shared code)
>         3680 libpthread.so.pure.1b7a5a8a8 (private data)
>       475136 libm.so.pure.18462ee2e (shared code)
>        95696 libm.so.pure.18462ee2e (private data)
>      2883584 libc.so.1.pure.9f4e4221 (shared code)
>        70640 libc.so.1.pure.9f4e4221 (private data)
>   * Memory mapped usage:
>        49152 mmap'd at 0x4888000
>      1425408 mmap'd at 0x48c8000
>       262144 mmap'd at 0x4714000
>        49152 mmap'd at 0x4020000
>      4259840 mmap'd at 0x42f4000
>        20480 mmap'd at 0x48b0000

> -- 
> ------------------------------------------------------
> Ralph Heinkel
> European Molecular Biology Laboratory (EMBL)
> Meyerhofstr. 1
> 69012 Heidelberg
> Tel. +49 6221/387 529
> eMail: heinkel at embl-heidelberg.de

-- 
------------------------------------------------------
Ralph Heinkel
European Molecular Biology Laboratory (EMBL)
Meyerhofstr. 1
69012 Heidelberg
Tel. +49 6221/387 529
eMail: heinkel at embl-heidelberg.de


From nospam at bitbucket.com  Mon Jan 31 02:45:39 2000
From: nospam at bitbucket.com (Phil Mayes)
Date: Sun, 30 Jan 2000 23:45:39 -0800
Subject: Can someone fix my code?
References: <8733rg$ral$1@nnrp1.deja.com>
Message-ID: <S4bl4.777$S3.31118@newsfeed.avtel.net>

bovinesoft at my-deja.com wrote in message <8733rg$ral$1 at nnrp1.deja.com>...
>I am a complete Python newbie and am trying to learn it as my first real
>programming language (I have dabbled in C, Java, and Assembly, but never
>really used them).  I am trying to write a program to get some practice,
>but am getting errors when I try to run it.  If someone could take a
>look at my code and tell me what is wrong with it, I would be greatly
>appreciative.

First: post the error message in future :)
Second: I presume you are merging two files into a third.
You're halfway there, but don't open the second input file!  Instead,
you're treating the output file as the second input:
>      for line in body:
>         body.write(line)
and also iterating over the handle; you need to read the lines, like
temp2.

Third (and getting pickier), I don't like the multiple calls to
ssi_or_exit.  Do this instead:
def ssi_or_exit():
   while 1:
      sorx = raw_input("Would you like to generate HTML <g> or exit <x>? ")
      if sorx == "g":  # if they want to generate a web page
         ssi()
      elif sorx == "x":  # if they want to exit
         sys.exit("Thank you for using Bovine No Server SSI!")

>import sys  # to use the exit() function
>
>def ssi_or_exit():
>   sorx = raw_input("Would you like to generate HTML <g> or exit <x>? ")
>   if sorx == "g":  # if they want to generate a web page
>      ssi()
>   elif sorx == "x":  # if they want to exit
>      sys.exit("Thank you for using Bovine No Server SSI!")
>   else:
>      ssi_or_exit() # try it again
>
>def ssi():
>   body_path = raw_input("Please type the path of the body section: ")
>   body = open(body_path, "w")  # open the text of the web page body
>   temp_path = raw_input("Please type the path of the HTML template: ")
>   temp = open(temp_path, "r")  # open the HTML template
>   temp2 = temp.readlines()
>   where_temp = raw_input("Put the template before <b> or after <a> the
>body? ")
>   if where_temp == "b":  #around here is where my trouble starts
>      for line in temp2:
>         body.write(line)
>      for line in body:
>         body.write(line)
>   else:
>      for line in body:
>         body.write(line)
>      for line in temp2:
>         body.write(line)
>   body.close()
>   temp.close()
>   ssi_or_exit()
>
>ssi_or_exit()

--
Phil Mayes    pmayes AT olivebr DOT com







From gregm at iname.com  Sun Jan 23 07:04:56 2000
From: gregm at iname.com (Greg McFarlane)
Date: Sun, 23 Jan 2000 23:04:56 +1100
Subject: TKinter and focus events
In-Reply-To: <388ABCB4.A51BF817@exceptionalminds.com>; from Timothy Grant on 23 Jan 2000 at 12:32:52AM
References: <388ABCB4.A51BF817@exceptionalminds.com>
Message-ID: <20000123230456.35657@nms.otc.telstra.com.au>

The beep is coming from the line:
    self.EstWt.configure(validate='integer')
This is because, at this stage.  the text is, for example "$1,234",
which is an invalid integer.  What you need to do is move this line
until after you have called setentry.

On 23 Jan, Timothy Grant wrote:
> Hi,
> 
> Thanks to some kind assistance from the list I have been able to make
> Tkinter do a pretty neat trick (at least I think so!)
> 
> Using Pmw.EntryFields, When the field does not have focus it displays
> numbers formatted to a specification (e.g., $1,234.56). When the
> EntryField gets the focus it reformats the number stripping out the
> eye-candy (e.g., 1234.56). To make sure the user enters the correct
> info, I turn on either 'integer' or 'real' validation when the
> EntryField has focus, and turn it off before I set it to the prettified
> entry.
> 
> It all work phenomenally well, and I'm tremendously pleased with how
> easy it was to convert a number in to nicely formatted text with commas.
> However, there is one minor annoyance, every time I move focus to the
> field I get what sounds like an error beep, everything works correctly,
> I just get this beep, I've been trying to track it down but haven't been
> able to figure it out yet. It appears to occur when I do the setentry()
> call with the unprettified number. Does anyone have any ideas what might
> be happening?
> 
>     def FmtEstWt(self, event):
>         # Event type 9 is FocusIn and type 10 is FocusOut
>         if event.type == '9':   #FocusIn
>             print 'EstWt FocusIn'
>             self.EstWt.configure(validate='integer')
>             self.tmp = string.replace(self.EstWt.get(),',', '')
>             self.tmp = string.replace(self.tmp, '$', '')
>             self.EstWt.setentry(self.tmp)	#Beep appears to go off here.
>         else:                   #FocusOut
>             print 'EstWt FocusOut'
>             self.EstWt.configure(validate=None)
>            
> self.EstWt.setentry(commanumber(string.atoi(self.EstWt.get()),0,0))
> 
> OK, I just thought of something just before I hit my send button, so I
> actually setup an "invalidcommand" option for the EntryField. Sure
> enough, the invalid function gets called when I do the setentry. This
> does not stop the setentry from working though. I've dumped the contents
> of self.tmp, and it appears to be a clean string that if I keyed by hand
> would pass the validation no problem.
> 
> -- 
> Stand Fast,
>     tjg.
> 
> Chief Technology Officer              tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.                   (503) 246-3630
> >>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com



From talon at coursen.ifu.net  Wed Jan 19 19:56:45 2000
From: talon at coursen.ifu.net (Stephen Coursen)
Date: Thu, 20 Jan 2000 00:56:45 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com>
Message-ID: <slrn88cnb6.4i6.talon@coursen.ifu.net>

On Wed, 19 Jan 2000 16:43:43 -0800, tye4 <tye4 at yahoo.com> wrote:
>
>Ivan Van Laningham <ivanlan at callware.com> wrote in message >
>
>> Don't code that way to start with.  It's dumb.  Treat yourself to a good
>> editor.
>>
>Better yet. Fix the dumb language.
>
As a confused newbie, what gives you the experience to declare Python to be
a "dumb" language?  Simply because *you* can't figure out the syntax?  It's
not that hard, especially if you have a good editor that handles the
indentation for you.  You're just too blockheaded to figure this simple fact
out.

>
>> You got about as much chance of getting this into Python as you do of
>> getting hit by a meteor.  Less, actually, because the odds for the
>> latter are measurable.  What you're asking for is for Python to not be
>> Python.  You don't like it, design your own language, build the
>> interpreter or compiler, and publish it.  Maybe you can get some people
>> to use it.  One of them won't be me.
>>
>> You're saying to people who know and love Python, "Python is a great
>> language except that it's Python.  Let's fix it."  We don't think it's
>> broken.  Your insistence on swimming upstream on this issue is not only
>> wrong-headed but boring.
>>
>Python is good but could be better.

But not necessarily syntax-wise.  The chosen syntax is clean, workable, and
readable.  Why change it?

>Everything has flaws and nothing is perfect. If we can't fix it then its a
>different
>matter. But in this case the problem can be fixed. It just so happens that
>there
>are a lot of bull-headed close-minded people than I reckoned there would be.
>I'm pretty certain now that this problem will never be fixed.
>

Hopefully, you're correct on this point.  Python don't need none of yer
style of fixin'

>I suggest you go to other newsgroups and let them know how brilliant Python
>is
>because it has an invisible block terminator.
>
>Since we are so lazy let's just skip the ending anything and just use
>indentation.
>
>Math:
>Before:
>    (2 + 3) * 5
>
>After Python rule:
>    (2 + 3
>            * 5
>
>C:
>Before:
>    if (x == 5)
>    {
>        printf("blah\n");
>    printf("blah blah");
>
>HTML:
>Before:
>    <h1> Python indentation: <i>Why it is such a pain</i>
>   </h1>
>    Blah blah blah
>
>After:
>    <h1> Python indentation
>            <i> Why it is such a pain
>    Blah blah blah
>
>Go ahead and convince these guys to switch do it the "Python Way".
>And Good luck. You will need it.
>

Oh, shut up.
<plonk>

Steve

>-tye4
>
>


-- 
--
Stephen Coursen                      talon at coursen.ifu.net
If everything is coming your way then you're in the wrong lane.


From wolftex at attglobal.net  Thu Jan 13 21:35:45 2000
From: wolftex at attglobal.net (Wolfgang Roesner)
Date: Thu, 13 Jan 2000 20:35:45 -0600
Subject: Parser Generator for Python
Message-ID: <387E8B81.23C91488@attglobal.net>

Hi!
I haven't found a good reference so far for a parser generator (a la
ANTLR)
that produces Python code. Is there a robust generator available
anywhere?

Thanks,
Wolfgang Roesner


From eddyw at lsls4p.lsl  Wed Jan 12 10:22:26 2000
From: eddyw at lsls4p.lsl (Edward Welbourne)
Date: Wed, 12 Jan 2000 15:22:26 GMT
Subject: converting tuple to list
Message-ID: <200001121522.PAA15694@lsls4p.lsl>

There are built-ins for this: 

list(sequence) -> yields a list with the same entries as sequence
tuple(sequence) -> yields a tuple, likewise

and map(lambda x: x, sequence) is fine for list (I can believe map would
use this lambda, the identify-function, if given None)

	Eddy.



From skaller at maxtal.com.au  Mon Jan 17 09:36:40 2000
From: skaller at maxtal.com.au (skaller)
Date: Tue, 18 Jan 2000 01:36:40 +1100
Subject: [Types-sig] Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk>
Message-ID: <388328F8.85147F85@maxtal.com.au>

Edward Welbourne wrote:
> 
> I have been thinking about the namespace types.
> That's modules, packages, classes and instances.
> I believe they could profitably be united into *one* type.
> Along the way, I see some constructive uniformisations we could apply.

	Sigh. See Viper. Modules, packages, classes, instance,
raw objects, local scopes .. and some other scopes .. are all
'unified' in two ways:

	1) Support for an 'environment' protocol supporting
unqualified name lookup

	2) Support for qualified lookup [x.y]
 
	In Viper, 'environment' is a fundamental abstraction
of the execution environment: there's an actual class interface
abstraction
for it. Qualified lookup is supported by ad-hoc polymorphism, more or
less
encapsulated in the 'getattr' 'setattr' 'hasattr' functions.

>   * namespaces that you can't modify once they have been initialised
>   * namespaces in which attribute-modification is type-checked

	As you can see, in Viper, there are _two_ dual concepts
of namespaces. In both cases, the fundamental implementation involves
dictionaries, plus extra structure.

>   * hybrids of those with assorted degrees of control
>   * certain tools the functional programmers crave, notably a truly
>     faithful implementation of currie()

	not required: python lambdas subsume what can be done with
	currying.

>   * some of the truly whacky `metaclass' proposals

	There are plenty of them :-)
 
	See Viper. It already supports generic types, eg lists of 
X for any type X. This is done fairly easily by making

	ListOf a class with one attribute 'type'
	ListOf(X) an instance with type = X

and that instance the type of a raw object with a normal list
as it's only attribute. Methods of the ListOf class have
two objects: the type object, (ListOf(X)), and the list instance (self).
These are both bound by the two stage lookup, so that a method
like 

	def append(typeobj, self, value):
		if type(value) is not typeobj.type: 
			raise TypeError
		else:  self.list.append(value)

> I intend to push this proposal during IPC8.  I would greatly appreciate
> constructive comments in advance, if only to temper my enthusiasm ;^}

	The lookup mechanism in Viper is very powerful.
It is not necessarily 'right'. It might be useful to examine
a working implementation? After all, it more or less models
what C Python 1.5.2 does now, and just relaxes some restrictions
on what a type object can be.

	[Viper raw objects can also be metamorphic -- that is,
change type dynamically -- the type is 'just another attribute']

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia voice: 61-2-9660-0850
download: ftp://ftp.cs.usyd.edu/au/jskaller



From neelk at brick.cswv.com  Thu Jan 27 19:13:57 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 28 Jan 2000 00:13:57 GMT
Subject: nesting class definitions?
References: <20000127155551.A2069@stopcontact.palga.uucp>
Message-ID: <slrn891nqh.bt0.neelk@brick.cswv.com>

Gerrit Holl <gerrit.holl at pobox.com> wrote:
> Hello,
> 
> I do see the point in nesting function definitions, namely as an alternative
> on places where one could use lambda. But what's the point in nesting
> class definitions? I know it's possible, but I've never seen it used.

I have used it, though for several reasons it's definitely not a
standard idiom.

(The pragmatic reason: according to Tim, Guido originally allowed
nested class and function definitions because it simplified the
grammar a little bit, and since in Python class and function
definitions are runtime statements rather than compile-time
declarations, there wasn't any additional cost on the interpreter
end. Now for a hopefully more enlightening explanation....)

>From time to time you will see nested function definitions; this is
largely because you can put things into the functions namespace, both
with regular arguments and through the default-argument hack for
faking closures. However, you hardly ever see nested class
definitions, because there really isn't a good way to communicate
information from the outer scope into the inner class's scope. (This
is a small lie, but bear with me.) That is, if you have the function

def class_factory(foo, bar, baz):
    quux = ... # some complicated computation...

    class Inner:
        ...
        # Note, there's no way to involve the contents of quux in
        # the definition of Inner! For example:
	#
	glark = frob(quux) # This won't work!
    ...
    return Inner

As a result, there's no real point to defining an inner class in
Python. If Python were lexically scoped, then quux would be visible
inside Inner's definition body, and there would be some small point to
nested classes. Even if it were possible, I doubt it would get used
much, though -- a class is a type, and most programmers prefer to set
up all their types as part of structuring their program. Generating
new classes at runtime is just hard to think about. Even in Common
Lisp, which has a well-defined facility for hacking the type system
called the metaobject protocol, programmers who make regular use of it
are considered wizardly and strange.

Now, you might notice that at the start of my post I mentioned that I
had used this, but I've spent my whole post explaining why no one
does. This is because of the small lie I mentioned before. Namely,
there *is* one way of communicating with the innards of Inner --
through the list of superclasses. In a class definition like

class Foo(Bar, Thud):
   ...

(Bar, Thud) is just an ordinary Python tuple -- there's nothing magic
about them, and as long as it is tuple of classes Python is perfectly
happy. Now, in languages with multiple inheritance, there is a style
of defining classes called "mixins" -- you create a set of classes,
each containing a separate capability, and then to create a class with
all of those capabilities you define a new class that inherits from
each mixin. (These classes are "mixed in", hence the name.)

So in one application I had, I had a number of mixins, and needed to
define a class based to handle some input data. If I had tried to
define a class for each possible combination, I would have had to
define a *bunch* of classes (I think something like 60), and as a
result I defined a function like this:

def class_factory(*mixins):
    class Anonymous:
        # some various routines
    Anonymous.__bases__ = tuple(mixins)
    return Anonymous

so that I could call class_factory() to create the appropriate class
at runtime.
 

Neel


From michel at digicool.com  Fri Jan 21 18:29:02 2000
From: michel at digicool.com (Michel Pelletier)
Date: Fri, 21 Jan 2000 18:29:02 -0500
Subject: What about REBOL (www.rebol.com)
References: <ngQh4.17853$pb2.1435429@tw11.nn.bcandid.com> <slrn88fmtd.5fj.wtanksle@hawking.armored.net>
Message-ID: <3888EBBE.D293886C@digicool.com>

William Tanksley wrote:

> Heck, not even discovering Joy
> (http://www.latrobe.edu.au/www/philosophy/phimvt/j00syn.html), surely the
> best language ever designed, has not sullied my happiness with Python.

Weird, it looks like the bastard child of a paren-less LISP and an
inverted FORTH.

Cool.

-Michel


From tim_one at email.msn.com  Wed Jan  5 23:59:49 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 5 Jan 2000 23:59:49 -0500
Subject: Python not a Very High-Level Language?
In-Reply-To: <slrn86ur6a.fph.neelk@brick.cswv.com>
Message-ID: <000801bf5802$dcca7d60$0aa0143f@tim>

[Tim]
> In any case, it's quite rare for any given language to evolve
> in really significant ways ... Fortran is ... an exception

[Neel Krishnaswami]
> As possible counterexamples, what about Lisp and C++?

No single version of Lisp ever had enough users for anyone to care, and to
this day 98% of Scheme users do nothing with the language except implement
compilers for their own Scheme variants <wink>.

C++ has certainly been a wild ride, although "the real C" continues to
evolve in conservative ways.

> Common Lisp is quite a different Lisp from basically all of
> the Lisps that preceded it (eg, it's lexically scoped rather
> than dynamically scoped). And on top of that it grew CLOS,
> which is (again) a pretty major change -- it was an backward-
> incompatible unification of the Common Loops and Flavors
< object systems.

Seriously, the Lisp family is unique in many respects.  IIRC, the addition
of CLOS to CL made it the first OO language with an ANSI std!  Lisp rarely
gets the credit it deserves.  Guido honored it by throwing away its syntax
<wink>.

>> Languages ... that should also be mined for inspiration are
>> Prolog, Dylan and SETL

> I know (and quite like) both Prolog and Dylan, but now you've
> given me another language to learn. Thanks!

The first validated Ada compiler was written in SETL -- and that's about all
anyone seems to remember it for these days.  I'm not sure you can even get a
viable implementation anymore.  amazon.com lists the primary book (which is
a good one!) as out of print:

    Programming With Sets: An Introduction to SETL
    (Texts and Monographs in Computer Science)
    Schwartz, Dewar, Dubinsky, Schonberg
    Spring-Verlag

The primary gimmick was, well, sets <wink>. "Even functions" were viewed as
sets of pairs, and high-level mathematical set notation (like construction
via abstraction, and nested existential & universal quantifiers) got
executed directly.  That integers "should be" unbounded was thought so
obvious that the book never even mentions that they are.  The Haskell folk
credit SETL for list comprehensions, and it even had built-in support for
non-deterministic programming (of the flavor Scheme would later model via
"arb").  Way ahead of its time -- heck, likely still ahead!  The Courant
Institute (SETL's home) did much pioneering work in high-level algorithm
transformation, motivated in part by the challenge of making 6 lines of SETL
run to completion before you died <wink>.

> And now a favor for you: have you heard of Mercury? It's kind
> of like a purely declarative Prolog with a higher-order type
> system designed along the lines of Haskell or ML.

I've never used it, but my time for learning new languages now is close to
zero.  It struck me as being Prolog on industrial-strength static
declarations, aiming more at "big system" integration, type safety and speed
than at raising the level of expressiveness.  Someone should tell them that
the "Hello, World!" of the tutorial does not leave a compellingly inviting
first impression <0.5 wink>:

<SNIP>

:- module hello_world.

:- interface.

:- import_module io.

:- pred main(io__state, io__state).
:- mode main(di, uo) is det.

:- implementation.

main -->
	io__write_string("Hello, World!\n").

</SNIP>

That's enough for me <wink>.

If you've got the time, investigate Oz/Mozart for us!

    http://www.mozart-oz.org/

If I had the time, that's the one I'd check out next, because it's the one
I'm most prejudiced against <no wink>:  it seeks to combine all of logic,
constraint-based, functional, and OO approaches on a highly concurrent
substrate.  When I picture those all together, I can only conjure up visions
of extreme pain; but browsing thru the intro material didn't hurt at all,
and someone here last year (Boris?) spoke highly of it.  If not truly
innovative, it's at least bold <wink>.

> Hm. Would a real metaobject protocol (as the offspring of
> type/class unification) count as major change [for Python]?

It would be a big change for the implementation, but not really to the
conceptual bases of the language.  The type/class split today is much more
an artifact of the initial implementation than a design choice.  Indeed,
part of the hangup in healing this split has been that there are so *many*
ways to do it that all "feel Pythonic" (and JPython has already healed it,
in a Java-like way, with some unfortunate hacks to make it behave more like
CPython).

> ...
> Yes. Smalltalk and Common Lisp are two of the most fun
> languages to program in, in large part due to the power
> of their interactive environments.

Almost as good as Model 100 BASIC!

missing-"power off, resume"-ly y'rs  - tim





From wtbridgman at radix.net  Mon Jan 17 17:22:04 2000
From: wtbridgman at radix.net (T. Bridgman)
Date: Mon, 17 Jan 2000 17:22:04 -0500
Subject: Python code for documenting Python classes?
Message-ID: <01HW.B4A9003C005E83C70C98670C@news1.radix.net>

Just so I don't try to re-invent the wheel...

Does anyone know of a Python program for documenting Python classes?

While it's usually easy to identify the methods in a particular class, 
it is often difficult to identify all the data members since they don't 
get listed in header files as in C++.  In particular, I'm looking for 
something that could list out all the classes, methods, and data 
members in a module, perhaps in a format similar to a C++ header:

class 1 (subclass)
 data members
   member 1: list
   member 2: sequence
   member 3: Float32
 methods
   method 1 (parmlist)
   method 2 (parmlist)
   method 3 (parmlist)

This would facilitate documenting the classes.

Naturally, I'd like something that can also recognize classes from 
other modules (particularly NumPy).

Thanks,
Tom



From jstok at bluedog.apana.org.au  Mon Jan 17 21:33:46 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Tue, 18 Jan 2000 02:33:46 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
Message-ID: <eiQg4.5231$3b6.23488@ozemail.com.au>

On Mon, 17 Jan 2000 17:32:02 -0800, tye4 <tye4 at yahoo.com> wrote:

>I'm a python newbie.. but I like Python better than Perl (no ugly @$#@.. in
>syntax).
>
>Python does have one quirk: no end if or end loops

Yes, we know.

[snip]

>Indentation is a confusing and lame way to separate what is inside a block
>and what's outside it.

I'm getting used to it.  I'm familiar with Haskell, which uses
indentation to demarcate scope as well.  I think the problem is that
you're just not used to it.

-- 
Jason Stokes: jstok at bluedog.apana.org.au


From bsb at winnegan.de  Thu Jan 20 21:16:45 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 21 Jan 2000 03:16:45 +0100
Subject: Small `du' reformatter :-)
In-Reply-To: =?ISO-8859-1?Q?Fran=E7ois?= Pinard's message of "20 Jan 2000
 18:11:42 -0500"
References: <oqogagfidt.fsf@titan.progiciels-bpi.ca>
Message-ID: <8766womanm.fsf@baal.winnegan.de>

Fran?ois Pinard <pinard at iro.umontreal.ca> writes:

> [1  <text/plain>]
> Hi, people.

> I knew I wrote a small Python tool to reformat `du' output, but I
> lost it for a while, and unexpectedly stumbled on it today, looking
> for something unrelated.  Here it is.  So if I loose it again, I'll
> ask you! :-)

Quoting Tim Peters from another thread:

Tim> As I mentioned a couple days ago in this thread, 99% of such a
Tim> script ships with Python, in Tools/Scripts/pindent.py.  From the
Tim> top of that file:

s/pindent/dutree/

Regards
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
                  bsb at north.de - http://www.north.de/~bsb/
****** ceterum censeo javascriptum esse restrictam *******



From jbauer at rubic.com  Sat Jan 22 16:01:08 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Sat, 22 Jan 2000 15:01:08 -0600
Subject: ANN: platform.py - Version 0.7.0
References: <3889CFEE.E2C0EF1E@lemburg.com>
Message-ID: <388A1A94.C4B4B462@rubic.com>

"M.-A. Lemburg" wrote:
> Documentation is provided through doc-strings and many
> comments in the source code, which can be found via my
> Python Pages:
> 
>         http://www.lemburg.com/~python/

Actually should be: http://www.lemburg.com/python/
URL without the tilde (~) character.

Jeff Bauer
Rubicon Research



From dworkin at ccs.neu.edu  Sat Jan 29 16:45:39 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 29 Jan 2000 16:45:39 -0500
Subject: Some questions about history...
In-Reply-To: Gerrit Holl's message of "Sat, 29 Jan 2000 20:24:38 +0100"
References: <20000129202438.A5089@stopcontact.palga.uucp>
Message-ID: <vndoga4mu0s.fsf@betelgeuse.ccs.neu.edu>

Gerrit Holl <gerrit.holl at pobox.com> writes:

> I'm creating a Dutch Python page, and want to tell something about
> the history. But I can't find out the following things:
> 
> 	- When did the PSA begin?

I don't know that one, but I know how to find out the others.

> 	- When was the domain www.python.org registered?

Accoring to Network Solutions (*spit*) this domain was registered on
27-Mar-1995.  And of course, that is the domain 'python.org'.
'www.python.org' is just a host within that domain.

> 	- When was c.l.py founded? When was c.l.py.announce founded?

>From looking at archived news.announce.newgroups postings:

c.l.py -> March 28, 1994

c.l.py.announce -> March 12, 1998

-Justin

 



From tim_one at email.msn.com  Fri Jan 21 19:19:55 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 21 Jan 2000 19:19:55 -0500
Subject: What about REBOL (www.rebol.com)
In-Reply-To: <3888EBBE.D293886C@digicool.com>
Message-ID: <000901bf646e$699de980$caa0143f@tim>

[William Tanksley]
> Heck, not even discovering Joy
> (http://www.latrobe.edu.au/www/philosophy/phimvt/j00syn.html),
> surely the best language ever designed, has not sullied my
> happiness with Python.

[Michel Pelletier]
> Weird, it looks like the bastard child of a paren-less LISP
> and an inverted FORTH.

Yes, Billy *does* leave that impression on people <wink>.  As you say,
though, Joy looks pretty:

> Cool.

for-austerely-elegant-defns-of-coolness-ly y'rs  - tim





From mwh21 at cam.ac.uk  Thu Jan 13 15:09:02 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jan 2000 20:09:02 +0000
Subject: inverse of __repr__ ?
References: <yqtn1qa8aga.fsf@toad.scripps> <001801bf5dc1$ac43a990$f29b12c2@secret.pythonware.com>
Message-ID: <m3u2khivj5.fsf@atrus.jesus.cam.ac.uk>

"Fredrik Lundh" <fredrik at pythonware.com> writes:

> Thien-Thi Nguyen wrote:

> > i've looked through _programming python_ for the inverse of
> > __repr__, but could not find anything.  perhaps i'm missing some
> > clue?
> >
> > or is there no way to extend the reader in a python repl?

No.

> if "reader" means the python source parser, as exposed
> by eval/exec/compile, it's limited to python syntax.  or in
> other words:
> 
>     x = eval(repr(x))
> 
> only holds for simple types.
> 
> dunno what a "repl" is, so I cannot answer the rest
> of your question.

Read-Eval-Print-Loop. Lisp terminology, I think, and one of the things
that makes both Python and Lisp as fantastic as they are.

Cheers,
Michael


From newshoes at mail.com  Tue Jan  4 18:00:37 2000
From: newshoes at mail.com (NewShoes)
Date: 4 Jan 2000 23:00:37 GMT
Subject: http POSTing
Message-ID: <8EB19A9E7newshoesmailcom@205.147.63.94>

i'm trying to create an offline form i can send to a website
through "POST". being new to python and minimal knowledge of
the http protocol, this has stumped me rather quick.

obviously the urllib is out, since it seems i can only
perform "GET" requests. but i can't figure out if the
httplib is lowlevel enough, or if i need to go down to sockets

anyways, if someone has an HTML form like this...


<FORM ACTION="http://www.test.com/form.html" METHOD="POST">
<INPUT TYPE="hidden" NAME="hidd" VAL="en">
<INPUT TYPE="text" NAME="email" VALUE="you at yourmail.com">
<INPUT TYPE="image" SRC=post.gif NAME="POST">
</FORM>


how would one go about sending this off if filled out
from a simple TKinter form??
(quick example code being what i seek here) :]



From loewis at informatik.hu-berlin.de  Thu Jan  6 14:44:07 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 06 Jan 2000 20:44:07 +0100
Subject: Reading tar/gz-files
References: <3874DDAA.E14CC23B@bibsyst.no>
Message-ID: <p6qzoujouig.fsf@informatik.hu-berlin.de>

Thomas Weholt <thomas at bibsyst.no> writes:

> Would it be possible to read out the contents of tar.gz/tgz files using
> Python? Could I, using Linux, use catch the output of a os.exec('tar tvf
> ' + filename) for instance, in a list??

You could use os.popen; this connects you to the output of tar. That
would work fine for tar tvf. See the library reference about usage of
os.popen.

If you want to get to the files as well, I recommend that you unpack
the tar file onto disk. If you absolutely want to live without
temporary files, you can use the 'O' option for receiving the file
contents through the pipe.

Alternatively, you could try to live without an external tar command,
and implement that yourself. The gzip module is already there and
provides you with an uncompressed stream. All you'd need to do is to
interpret the format of tar yourself. See /usr/include/tar.h for a
starting point.

Hope this helps,
Martin


From News at Titanic.co.uk  Sun Jan  2 17:40:26 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Sun, 02 Jan 2000 22:40:26 GMT
Subject: MySQL
References: <pj0v6sg2d06cjuih0ofjhnoqjm5lp6q8lr@4ax.com> <84o6sf$e4j$1@news1.xs4all.nl>
Message-ID: <r2lv6sk6leep7jv66folc55rr97nlup8bg@4ax.com>

Yeah!

It works!

Here is the code, please comment and help me to learn how to write
better code.

>>
#!/usr/bin/python
import MySQLdb, cgi, sys

def bail():
  print "Content-type: text/html\n"  
  print "<H3>Error</H3>"
  sys.exit()

def printaddresses():
  db =
MySQLdb.connect(host="localhost",db='DPAdmin',user="root",passwd="topsecret")
  cursor = db.cursor()
  cursor.execute("SELECT * FROM tblIPAddress ORDER BY Name")
  tuple = cursor.fetchall()
  total = len(tuple)
  if total < 1:
    print "No entries"
  else:
    print "Content-type: text/html\n"
    print "<H3>DP ASSIGNED IP ADDRESSES</H3>"
    print "<P>"
    print "<B>"
    print "<table border cellspacing=0 cellpadding=5>"
    print "<tr>"
    print "<th> Company Name </th>"
    print "<th> IP Address </th>"
    print "</tr>"
    for record in range(total):
      Name, IPAddress =  tuple[record]
      print "<tr>"
      print "<td>" + Name + "</td>"
      print "<td>" + IPAddress + "</td>"
      print "</tr>"
    print "</table>"
  db.close()
  print "</B>"
  print "<FORM METHOD=POST
ACTION='http://www.myaddress.co.uk/cgi-bin/ipaddress.py'>"
  print "<P>"
  print "NAME<INPUT TYPE=text NAME=Name SIZE=50 VALUE=''>"
  print "<P>"
  print "IPADDRESS<INPUT TYPE=text NAME=IPAddress SIZE=30 VALUE=''>"
  print "<P>"
  print "<INPUT TYPE=submit NAME=add VALUE='Add the Address'>"
  print "</FORM>"

class FormData:
  def __init__(self, formdict):
    for fieldname in self.fieldnames:
      setattr(self, fieldname, form[fieldname].value)

#MAIN

sys.stderr = sys.stdout
form = cgi.FieldStorage()
if form.has_key("Name") and not form["Name"].value == "":
  NewName = form["Name"].value
  NewIPAddress = form["IPAddress"].value
  SQL = "INSERT into tblIPAddress (Name,IPAddress) VALUES ('" +
NewName + "','" + NewIPAddress + "')"

db=MySQLdb.connect(host="localhost",db="DPAdmin",user="root",passwd="topsecret")
  cursor=db.cursor()
  cursor.execute(SQL)
  db.close
printaddresses()
<<
On 2 Jan 2000 18:54:07 GMT, boud at rempt.xs4all.nl (Boudewijn Rempt)
wrote:

>Simon Faulkner <News at titanic.co.uk> wrote:
>> Confused - I am!
>
>> I have written a lovely little database stored in MySQL that has a
>> Python front end and uses MySQLdb through _mysql on SuSE with MySQL
>> 3.27 - Fine.  All on my Sony laptop development machine.
>
>> Now I am trying to put it onto my RH6 production server I try to build
>> the MySQLdb-0.1.1 and it reports that it needs _mysql.  Where do I get
>> this from?
>
>> If it is part of MySQL then where is it?  I have the client, server,
>> benchmarks and devel loaded.
>
>> Is there a simple overall explanation of databases in Python anywhere?
>
>As far as I know, everything is simply in the MySQLdb-0.1.1 tarball -
>if you were to unpack it, build it (after checking for local variations,
>I had to do some fiddling due to the places SuSE installs stuff in. The
>file _mysqlmodule.c defines the _mysql module, really.
>
>Is this database thing of yours available for looking at? I've noticed
>a lot of questions about MySQL and Python lately, and having worked on
>one or two apps myself, I though about collecting ideas and writing a
>short paper or tutorial on the subject - you know, pinching ideas here
>and there, digesting and then regurgitating the mess, before presenting
>it to the web...
>
>Interesting points in doing stuff with Python and databases are time
>and dates (a perennial problem - even when you're simply working with
>VB or Delphi and Oracle), numerical formats (ditto, especially when,
>after applying the Microsoft millennium patch the number decimal
>conversion is broken), and of course decent design. 

Simon Faulkner


From gerrit.holl at pobox.com  Fri Jan 21 10:56:52 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 16:56:52 +0100
Subject: Congratulations, Tye!
In-Reply-To: <388774C7.1784B662@callware.com>; from ivanlan@callware.com on Thu, Jan 20, 2000 at 01:49:11PM -0700
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com> <388774C7.1784B662@callware.com>
Message-ID: <20000121165652.A843@stopcontact.palga.uucp>

Ivan Van Laningham wrote on 948372551:
[Tye asked for the Python source and Python license]
> RTFM:  http://www.python.org
[Ivan explains why he said RTFM]

Tye,

Congratulations!

I'm following this newsgroup for almost a year now, and this is the
very first time I saw 'RTFM'! I searched on deja for it, and there
were about 30 matches, of which many where because I had in my .sig
'RTFM: Read That FresMeat' earlier, and others where from someone
who called himself RTFM. It almost never happens someone says RTFM
to someone else in this newsgroup. It's really an achievement!

Congratulations on letting Ivan say RTFM!

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From fredrik at pythonware.com  Wed Jan 19 12:47:23 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 18:47:23 +0100
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com>
Message-ID: <010801bf62a5$3e6c6400$f29b12c2@secret.pythonware.com>

Michel Orengo wrote:
> If you do not have access to the MSDN Library, you can use a COM
> browser: it will give you the properties/methods of the objects.

footnote: MSDN is available on-line, at:
    http://msdn.microsoft.com/library/

you'll find the MAPI docs under "Platform SDK ->
Messaging and Collaboration Services"

</F>




From kc5tja at garnet.armored.net  Thu Jan  6 00:39:47 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 6 Jan 2000 05:39:47 GMT
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net> <N6Vc4.610$hJ1.7143@news.rdc1.tx.home.com>
Message-ID: <slrn878b1i.k6b.kc5tja@garnet.armored.net>

>> Seems like many web host providers, such as MindSpring, offer CGI with
>> Perl, but not other languages, notably Python.
>>
>> Does anyone know who would host a CGI Python site?

Armored Internet (http://www.armored.net) supports Zope and Python CGIs.
Zope is supported under our Unix web hosting packages, and Python CGIs are
supported under both Unix and Windows NT web hosting packages.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From emmanuel.pietriga at inrialpes.fr  Mon Jan 10 02:42:47 2000
From: emmanuel.pietriga at inrialpes.fr (Emmanuel Pietriga)
Date: Mon, 10 Jan 2000 08:42:47 +0100
Subject: JPython serialization problem
References: <387989FF.658A8F68@xrce.xerox.com>
Message-ID: <38798D77.9D3C267A@inrialpes.fr>

I forgot to mention that I'm running JPython1.1beta2  with JDK 1.1.7B



Emmanuel Pietriga wrote:

> My problem is rather "hard" to explain, so I'll take an example:
>
> I have two applications which communicate using a socket:
> One is written in Java and sends serialized objects to the other one,
> written in python, which uses java classes (therefore it runs under
> JPython).
> The python application receives serialized objects from the java
> application.
>
> I use java.io.ObjectOutputStream.writeObject() on the java side, and
> ObjectInputStream.readObject() on the python side.
>
> Of course, the objects I send all implement the Serializable interface.
>
> Here's the problem:
>     if I send an instance of a class which is part of the standard Java
> API  (like Vector, Hashtable, or even awt.Label,.....) everything is
> fine.
>     but if I send objects of a class created by me, no matter how simple
> the class is, I get a NoClassDefFound exception on the python side when
> reading the serialized object from the objectinputstream.
>
> At first I thought the paths were defined incorrectly, but if I create
> an instance of the same class directly on the python side using its
> constructor, it works!  Therefore JPython knows where to find the class
> definition.
>
> Anyone can help?
>
> Thanks.

--
Emmanuel Pietriga
OPERA
emmanuel.pietriga at inrialpes.fr





From Alex.Martelli at think3.com  Fri Jan 28 10:57:55 2000
From: Alex.Martelli at think3.com (Alex Martelli)
Date: Fri, 28 Jan 2000 16:57:55 +0100
Subject: vim and tabs in Python code
Message-ID: <6D8A17398E28D3119F860090274DD7DB4B3E13@pces.cadlab.it>

Nicolas writes:

- The indentation is seen by the interpreter but not
  by me, e.g. a 4-blank tab is seen the same as 4 blanks
  but not interpreted the same. Hard to tell something
  invisible is different from something else invisible.

Right, which is why VIM (and Emacs, and other good editors)
make it easy for you to smooth this out.


- Whatever intelligence you put in vim, will not help
  other kinds of editors, will it?

Of course -- a bad editor remains bad.


> I would appreciate, though, if anyone could give an
> advice about how to configure vim or preprocess
> somebody else's code to make it fit any given convention
> for indentation.

Look into the expandtab, smarttab and softtabstops options
of VIM.

Setting expandtab, any tab you enter becomes the right
number of blanks; setting smarttab, the shiftwidth option
becomes valid also for a tab you press at line start;
softtabstop may make it more natural to erase (backspace)
over multiple blank spaces just inserted by hitting a tab --
or it may not, try it out and see how it works for you.

With :autocmd BufWritePre *.py retab, and similar settings,
you can also ensure that tabs are always eliminated (or,
always inserted!) when a file is written (or read, etc --
here, I've specified it for writing, and for Python source
files only, but it's easy to tweak this!).


Alex



From gmcm at hypernet.com  Fri Jan 21 08:49:06 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 21 Jan 2000 08:49:06 -0500
Subject: Using tabs effectively. Was: Why not use tabs?
In-Reply-To: <3887DF0D.B102B357@Lugoj.Com>
Message-ID: <1263658531-14676592@hypernet.com>

James Logajan wrote:

> .... For example, consider a file that contains the
> following text lines, where ^I indicates a tab:
> 
> A^I^Ix
> AA^I^Ix
> AAA^I^Ix
> AAAA^Ix
> AAAAA^Ix
> AAAAAA^Ix
> AAAAAAA^Ix
> 
> The intent is to have that look like:
> 
> A       x
> AA      x
> AAA     x
> AAAA    x
> AAAAA   x
> AAAAAA  x
> AAAAAAA x
> 
> There is only ONE tab setting that will produce that result. 

Um, there's no tabsetting that will produce that result, but with 
the wonders of email, maybe that's not what you wrote.


- Gordon



From bitbucket at isomedia.com  Thu Jan  6 22:35:56 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Fri, 07 Jan 2000 03:35:56 GMT
Subject: python with ASP - pytest.asp (0/1)
References: <000001bf58c9$8c26ff60$39f4fea9@desktop>
Message-ID: <38755ea1.450061644@news.isomedia.com>

When you copy-and-pasted my code, it appears you lost the tabbing.
I've attached the code in case your newsreader isn't displaying the
tabs happily.
    -Eugene

On Thu, 6 Jan 2000 20:39:51 -0800, "Sameer Chowdhury"
<sameerc1 at yahoo.com> wrote:

>I get the following error with the code listed below:
>
>Python ActiveX Scripting Engine error '80020009' 
>
>invalid syntax 
>
>/pytest2.asp, line 4 
>
>sStr = '<b> Got Title</b>'
>---^
>
>
>So this code does what you want.
>
><%@ LANGUAGE = Python %>
><%
>if len (Request.Form("title")):
>sStr = '<b> Got Title</b>'
>else:
>sStr = '''<FORM ACTION="./pytest.asp" method="POST">
>
><b>Title:&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</b><BR>
><input type="text" value="" size=60 name="title"><P>
><input type=submit value="Submit">
></form>'''
>%>
><html>
><%= sStr %>
></html>
>
>    -Eugene
>
>
>
>
>
>
>__________________________________________
>NetZero - Defenders of the Free World
>Get your FREE Internet Access and Email at
>http://www.netzero.net/download/index.html
>

import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From nielsenjf at my-deja.com  Thu Jan  6 16:07:29 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Thu, 06 Jan 2000 21:07:29 GMT
Subject: (wierd situation) from file import not same as include :-(
Message-ID: <853064$ei7$1@nnrp1.deja.com>

I already tried from file import.
That doesn't help me :-(

Let me explain the situation further:

I have code running in an ASP environment, but I don't want it to know
that it is doing so. So I localize the ASP specific stuff in another
file.

The problem is the module in the other file does not have access to
ASP's objects. And I don't want to send it those objects,
since it is not supposed to know about it.

Any ideas on the best way to deal with this?

john


--
nielsenjf at my-Deja.com


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


From tim_one at email.msn.com  Thu Jan  6 19:52:46 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 6 Jan 2000 19:52:46 -0500
Subject: lazy (?) evaluation, request for advice
In-Reply-To: <006001bf5839$ade66180$c02b2bc1@martelli>
Message-ID: <000401bf58a9$843a6500$932d153f@tim>

[Alex Martelli]
>>> ... I'd much rather let the user write the deal-selection
>>> function in Python, than in either Tcl, or any custom,
>>> ad-hoc "little language".

[Tim, proving that his ability to give offense did not die
 with the changing of the calendar]
>> Against that, you invented ad-hoc semantics as you went along;
>> the only connection to Python appeared to be syntactic.

[Alex]
> Did I?  I used, e.g., spades to mean "the value of a
> variable named `spades' in the current scope[s]".  Isn't
> that normal Python semantics? ...
> I would appreciate understanding how you define "syntactic" in
> this context to justify this extremely peculiar assertion.

The answer to "what's in a name?" is fundamental semantics for any language.
Your language's answer is not Python's answer, hence my objection to calling
the former Python.

> ...
> Where more than one hand is of interest, object notation will
> be used; and where more power is needed, statements will be
> used rather than just an expression.

The last wasn't (isn't) clear.  You described it as a "deal-selection
function", but the example fragments showed only expressions.  If you are in
fact going to treat isolated expressions as if they were functions, it's
again a sense in which your language is not (although may be related to)
Python.  This means you're in the language-design business, as you went to
great lengths in this reply to prove; I can only repeat my advice that
Python is already well-designed and your users will be better served in the
end by letting them use the language as-is; you don't agree, so that's that.

>> Note also that if your users truly are morons, there's
>> nothing to stop *you* from marching over their string and
>> replacing e.g. "spades" with "current_hand.spades" before
>> compilation.

> Is there a way I could use the existing parsing tools to help
> me with this?  I've read through the library reference, but I
> still find that part a tad obscure. 3.14.6.1 and 3.14.6.2 are
> the parts I am thinking about.  If I could examine in the AST
> the variable names used in the expression, matching them with
> the dictionary of the hand features I know how to evaluate, I
> could pre-populate the dictionary to be used as the scope in
> the eval; or, I guess, change the variables as you suggest, to
> get even lazier (for those cases where short-circuit logic
> lets me evaluate a certain feature only in a subset of cases).
>  But the logic for examining, and even more for modifying, the
> AST, appears to be somewhat complex. Are there more online
> examples that I could use?
>
> Or, were you suggesting a similar operation on a purely lexical
> level?  E.g., with tokenize?  Or, a string-manipulation
> approach?

Well, "all of the above", plus John Aycock's SPARKS system and Michael
Hudson's bytecodehacks, *may* be appropriate.  It's simply impossible to be
helpful here until your language is fully defined, which is the first point
at which we can judge how radical a transformation is needed to convert it
into executable Python.

>From what you've said so far, it doesn't *look* like anything fancier than a
regexp replace is needed; e.g.,

import re, string
IMPLIED_OBJECT_NAME = "__whatever"
magic_names = string.split("diamonds clubs hearts spades losers")
_lookfor = re.compile(r"\b(" +
                      string.join(magic_names, "|") +
                      r")\b")
_replace = IMPLIED_OBJECT_NAME + r".\1"

during module initialization, and

   codestring = _lookfor.sub(_replace, codestring)

to do the transformation.  This changes, e.g.,

    spades >= 6 or hearts >= 6

to

    __whatever.spades >= 6 or __whatever.hearts >= 6

The bracketing word-boundary assertions ("\b") in the regexp prevent it from
transforming e.g. biglosers or heartsaches.  Generating a whole parse tree
wouldn't work any better, *provided that* you don't have string literals in
your language, and that the mapping to Python isn't hairier than simple
token substitution.

> I will appreciate further advice about this, and tolerate the
> sneers and name-calling that seem to go with the advice
> (although rarely have I seen discussion quite as sneering as
> this on this group

Alex!  That wasn't sneering -- it was brazen ridicule <wink>.  Seriously,
the exaggerated relentlessness of it all was supposed to nudge you away from
taking at all soooooo seriously.  Didn't work, eh?

> -- reminds me a lot of the ambiance on the Perl group -- this
> one had so far seemed friendlier, though I guess that may have
> been an artefact due to a frequentation started more recently).

Oh, this *is* much friendlier.  Over on c.l.p.m., they would have called
*you* an idiot.  I consistently insulted your *users*.

how-much-friendlier-can-a-computer-guy-get<wink>?!-ly y'rs  - tim





From dworkin at ccs.neu.edu  Wed Jan 19 09:47:36 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 19 Jan 2000 09:47:36 -0500
Subject: Python Rocks!
In-Reply-To: Sposhua's message of "Wed, 19 Jan 2000 11:36:44 +0000"
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <vndn1q340ez.fsf@betelgeuse.ccs.neu.edu> <Pine.WNT.4.21.0001191135140.206-100000@wks_3.e-shopwork.com>
Message-ID: <vnd66wq3ypj.fsf@betelgeuse.ccs.neu.edu>

Sposhua <sposhua at my.pc> writes:

> > > ! - whats wrong with tabs?
> > 
> > They don't look the same in every editor, or with each individual's
> > tab-width settings.
> 
> !? If you use ONLY tabs and not mix it with spaces (who the heck
> wants to jit the space bar 4 times when U can hit tab once?) surely
> all will be well, or have I missed something...?

Yes, you missed the possibility of someone else editing your code.
All of the silliness about tabs vs. spaces only comes up at all when
more than one person uses the same source file.

Also, I _never_ hit the space bar four times to insert a level of
indentation.  I hit the tab key, once.  My editor does an excellent
job of inserting the appropriate number of spaces.

-Justin

 



From seanb at home.com  Mon Jan 24 01:48:31 2000
From: seanb at home.com (seanb at home.com)
Date: Sun, 23 Jan 2000 22:48:31 -0800 (PST)
Subject: tabs do WHAT?
In-Reply-To: <000601bf6625$e80c3340$902d153f@tim>
Message-ID: <200001240648.WAA06269@seanb.sttln1.wa.home.com>

On 23 Jan, Tim Peters wrote:
> [Warren Postma]
>> Let's see how many languages I have run into that make
>> whitespace and/or tabs important:
> 
> Whitespace?  Now *that's* a novel topic c.l.py hasn't seen enough of lately
> <wink>.
> 
>> ...
>> As an embedded language, Python needs to be usable beyond the
>> ivory towers where functional programming or object oriented
>> programming or metaclasses are fodder for discussion.
> 
> Python didn't come from an ivory tower, and indeed even Perl may be more
> heavily used in those fabled lands.
> 
>> Python looked eminently useable by non-programmers until I found
>> out that whitespace was meaningful.  Try to explain scope to
>> someone who doesn't write programs for a living.
> 
> This one is ironic:  Python's treatment of block structure came from formal
> usability studies (conducted with the ABC language), using subjects who
> indeed didn't write programs for a living.  "Introductory colon +
> indentation" is what worked best -- they tried the others too.  See recent
> msgs in DejaNews for more about that.
> 
>> ...
>> Has anyone built a version of Python that totally ignores
>> tabs and whitespace,
> 
> Yes, a patch for that was posted many years ago; it didn't catch on.
> 
>> or a filter that you can use on top of Python?
> 
> See Tools/Scripts/pindent.py in the std distribution.
> 
> although-i'm-not-sure-it's-ever-been-run<wink>-ly y'rs  - tim
> 
> 
> 
Out of curiosity, are there any online copies of these usability studies
that were so formative to pythone?  Such studies might make interesting
reading and some people may have more trust in the results of a study
they could examine online.

Failing online access, is there any public access to write-ups of these
studies?
-- 
Sean Blakey
(206)297-7123
quine = ['print "quine =",quine,"; exec(quine[0])"']; exec(quine[0])




From rgruet at ina.fr  Tue Jan  4 09:36:40 2000
From: rgruet at ina.fr (Richard Gruet)
Date: Tue, 04 Jan 2000 15:36:40 +0100
Subject: String Handling
References: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com>
Message-ID: <38720578.46802AF6@ina.fr>

try this :

import string
s = string.join(string.split('Lucy by Railway Carraige'), '+')

Richard

Simon Faulkner wrote:

> Hi All,
>
> Can anyone give me the Python line/command to change all spaces in a
> string to +s?
>
> ie
>
> "Lucy by Railway Carraige"
>
> becomes
>
> "Lucy+by+Railway+Carraige"
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000104/2fd8e324/attachment.html>

From kuncej at mail.conservation.state.mo.us  Mon Jan 31 11:48:44 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Mon, 31 Jan 2000 10:48:44 -0600
Subject: Date and Time
References: <000201bf6bd2$eecba190$3acbd9c2@peridot.optichrome.com>
Message-ID: <a2jl4.84$aw4.1670@news.more.net>

> > I also have an all-python module that does a lot of date and time
> > things. It might give you some ideas.
>
> I'm surprised that something like this doesn't exist already in the
> standard Python library.

I think it's because doing it right is difficult, and there are several
differing "religions" about what is the right way. You can find
some long discussions in the archives...

I doubt that my datetime.py module fits anyone's notion
of the "right" way to do dates and times. It's just something I
hacked together over time to solve some local needs.

I think that Marc Lemburg has requested that mxDateTime be put
in the python distribution. That's probably the best solution, since
his work is very solid, and since Guido has entrusted other portions
of the distribution to him.

  --Jeff





From smb at bby.com.au  Wed Jan  5 00:08:43 2000
From: smb at bby.com.au (Sarah Burke)
Date: Wed, 05 Jan 2000 15:08:43 +1000
Subject: Sydney scoreboard?
References: <385DB48C.7A7A52D6@bby.com.au> <8On74.6414$Dh3.85256@ozemail.com.au> <YCU74.7437$NV6.31360@news-server.bigpond.net.au> <38601AF6.7D39F091@bby.com.au> <D0da4.3285$lv5.19947@news-server.bigpond.net.au>
Message-ID: <3872D1DB.D6DE0F2A@bby.com.au>

My netscape crashed whilst sending my last reply to this, so apologies
if this message is posted twice.
I would be very interested in a Sydney users group.
Are other people out there interested? Would people be interested in a
more formal or less formal group?

Mark Hammond wrote:

> "Sarah Burke" <smb at bby.com.au> wrote in message
> news:38601AF6.7D39F091 at bby.com.au...
> > Very impressive!
> > Since python is so big here, what sort of community activities are
> there?
> > (I'm only recently
> > arrived from Chicago).
>
> Not many.  We did create a python-au mailing list, but that is very
> quiet -
> http://starship.python.net/mailman/listinfo/python-au
>
> Some sort of social gathering would be good - although it would
> obviously need to be city-by-city.  Melbourne and Sydney both have
> decent numbers of users...
>
> Mark.



From skip at mojam.com  Tue Jan 25 16:38:03 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 25 Jan 2000 15:38:03 -0600 (CST)
Subject: Need Preprocessor Module!
In-Reply-To: <388E124C.B6B9BF9A@pk.highway.ne.jp>
References: <388E124C.B6B9BF9A@pk.highway.ne.jp>
Message-ID: <14478.6075.956183.905937@beluga.mojam.com>

    Akinori> I would like to use preprocessor-like functionality in Python
    Akinori> like we use in C/C++.

The C preprocessor runs at compile-time.  When do you envision a Python
preprocessor running?  If you want compile-time-like behavior and have a
build/install process in place, why not run your Python source through an
existing preprocessor like m4 like so:

    $(INSTALLDIR)/%.py : %.py
	    rm -f $@
	    $(M4) $(M4FLAGS) $< > $@

I use the above in my installation process with no problems.  I typically
use it to include files containing constants, but there's no reason I
couldn't use it to select installation-dependent blocks of code as well.

    Akinori> For example, you can comment multiple lines just by "if 0" and
    Akinori> "endif" with no need to consider indenting.  You don't have to
    Akinori> take care of not embracing another comment section, I mean...

If you want run-time behavior, I think it's a small price to pay to use "if
0:" or similar construct accompanied with the indentation support provided
by a decent text editor like Emacs with python-mode running.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From a.eyre at optichrome.com  Tue Jan 25 07:50:10 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 25 Jan 2000 12:50:10 -0000
Subject: Embedding -- getting error/traceback information not to stderr?
In-Reply-To: <slrn88r54f.ebi.erw@erw-pc.dde.dk>
Message-ID: <000501bf6732$b728bfb0$3acbd9c2@peridot.optichrome.com>

> >3) Set stderr to be stdout

> Well, I wanted the information to be in a variable in memory, which I can
> then send to the user (which isn't looking at stdout :) ).

Ah. I misunderstood what you needed.

I found this sort of thing to be quite useful, especially in cgi scripting:

import sys, traceback, string

def main(argv):
    try:
        all_my_code_here(argv[1])
        return 0
    except:
        s = string.join(apply(traceback.format_exception, sys.exc_info()))
        print s					# Do what you like with it here
        return 1

if __name__ == '__main__':
    sys.exit(main(sys.argv))

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From tismer at tismer.com  Mon Jan 24 13:17:23 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 24 Jan 2000 13:17:23 -0500
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com>
	 <$WXiIAAO+Dj4Ew2W@jessikat.demon.co.uk> <388C7D4A.D873AA@tismer.com> <zMBtEAA$jIj4Ewm7@jessikat.demon.co.uk>
Message-ID: <388C9733.B4F60DC0@tismer.com>


Robin Becker wrote:
> 
> In message <388C7D4A.D873AA at tismer.com>, Christian Tismer
> <tismer at tismer.com> writes

> >Whow. I guess there is a bug in exec. In any case, it cannot be a
> >conceptual but but just a failure of mine. Idle, PyWin, pystone,
> >Metakit, they all work fine.
> >
> >Will look into it after the conference.
> >Thanks for giving the measures.
> >
> ....
> 
> Zope uses several deep changes to standard python eg ExtensionClass,
> Aquisition which may be the cause of the problem.

Shouldn't be so. I do not change the object model
at all. It appears that I missed to give another
reference to an argument.
SLP is designed to work exactly as standard Python,
if the old API calls are used. I have to hunt it later.

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From wilson at nortelnetworks.com  Tue Jan 25 16:44:42 2000
From: wilson at nortelnetworks.com (Stewart Wilson)
Date: Tue, 25 Jan 2000 16:44:42 -0500
Subject: Multidimensional Dictionaries: Possible?
References: <slrn88ppsg.49t.w6uv@voyager.home.com> <oqpuuqx6s1.fsf@titan.progiciels-bpi.ca> <388DE028.ED24A3B1@nipltd.com> <14478.2995.807755.217066@beluga.mojam.com>
Message-ID: <388E194A.E54F93E2@americasm01.nt.com>

Skip Montanaro wrote:
> 
> If what you really want are nested dicts, then you should use something like
> 
>     >>> dict = {}
>     >>> dict[1] = {}
>     >>> dict[1][2] = 'foo'
>     >>> dict.keys()
>     [1]
>     >>> dict[1].keys()
>     [2]
> 

In addition, and to answer your original question on incrementing the
value, if your top level dictionary contains the first key, then you can
do:

dict["key1"]["key2"] = dict["key1"].get("key2",0) + 1

As in this example:

>>> dict = {}
>>> dict["key1"] = {}
>>> dict["key1"]["key2"] = dict["key1"].get("key2",0) + 1
>>> dict["key1"]["key2"]
1
>>> dict["key1"]["key2"] = dict["key1"].get("key2",0) + 1
>>> dict["key1"]["key2"]
2


---
Stewart Wilson
wilson at nortelnetworks.com



From gerrit.holl at pobox.com  Mon Jan 17 14:47:59 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 17 Jan 2000 20:47:59 +0100
Subject: If you can only buy *one* Python book...
In-Reply-To: <t6tg4.305$iL4.4241@news.rdc1.tx.home.com>; from pbible@home.com on Mon, Jan 17, 2000 at 12:11:05AM +0000
References: <N%rg4.4360$3b6.20735@ozemail.com.au> <20000116182502.A1106@better.net> <t6tg4.305$iL4.4241@news.rdc1.tx.home.com>
Message-ID: <20000117204759.B1291@stopcontact.palga.uucp>

Paul E. Bible wrote on 948064265:
> I found 'Learning Python' to be very easy to read and a great book to get
> started programming in Python...

Agreed.

> The 'Python Essential Reference' is a good addition to the online
> documentation as well.

Disagreed. I think the online reference is far the best and most
complete there is. You don't need any other book than 'Learning Python';
you aren't going to *read* the library reference, are you?
Maybe the first few chapters on builtin-types and common modules like os,
string (obsolete in 1.6), sys, and knowing about glob, rexec and
MimeWriter can also be useful, but *reading* it as a whole... no.

> I found 'Programming Python' somewhat difficult to follow myself.

Agreed.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From katz at glue.umd.edu  Sat Jan 22 10:16:20 2000
From: katz at glue.umd.edu (Roey Katz)
Date: Sat, 22 Jan 2000 15:16:20 GMT
Subject: Python dumps core with this.. how come?
References: <38891316.393423@news.erols.com> <005301bf64b5$d2f234d0$f29b12c2@secret.pythonware.com>
Message-ID: <3889c4b6.242514@news.erols.com>

On Sat, 22 Jan 2000 09:50:58 +0100, "Fredrik Lundh"
<fredrik at pythonware.com> wrote:

>looks like a recursive call in getattr (if "res" is
>not already available in the instance, it asks
>__getattr__ to look it up).

Thanks, I got it right after I posted :)
this brings up another concern. Given
the following definition:

   class A:

      def __getattr__( self, name ):
        return -1    
      def __setattr__( self, name, val ):
         self.__dict__[ name ] = val

I see that self.__dict__ must be a special variable
in that setting it somehow avoids a recursive call 
to __setattr__()? 



>> Also, I have another question (very old, and I can't find this in the
>> FAQ):  I want to explicitly take a reference to an integer.
>
>1. reset your brain ;-)
>2. variables are named references, not small boxes
>   that hold data values
>3. *all* variables in python are references.  there
>   is no other thing.
>4. integers cannot be modified in place.

Ok, OK:  this is what I always get when I ask this question,
but then why does this happen:

     >>> a = 3 
     >>> b = a 
     >>> b
     3 
     >>> a = 5
     >>> b
     3

I mean, what happens if I have a complicated path 
to a variable and I want to alias it:

    alias = long_name.hard_to_type[ 1 ].integer_value

    # arbitrary calculations that would be 
    # torturous with any label longer than 'alias'
    if alias > someValue:
       result = (alias * 5 ) / (alias + 4 )
    else:
       result = (alias * 3 ) / (alias + 2 )

>in other words, Python always copies references, but
>you cannot modify objects in place unless their inter-
>face allows you to (numbers, tuples and strings cannot
>be modified in place, while lists and dictionaries can --
>the only difference is that the former doesn't provide
>any methods that let you modify them...)

so where does this definition fit in the context I mentioned?
I understand that if I have a function and pass to it a parameter,
I can modify it in-place (although that I cannot modify numbers,
strings or tuples in-place seems to me inconsistent).  
So for example, these are roughly equivalent:
 
    void setValue(  int& x ) { x = 3  }    // C++
    void setValue(  int *x  ) { *x = 3 }    // C
    def  setValue( x ):    x = 3             # Python

whoops, we cannot modfiy integers in-place, how invonvenient.  Though
we could always wrap them in a class and overload its = operator
(wait! no = operator to overload! double inconvenience!) or have a
specially defined set() function:

    class A:
       def set( self, val ):
            self.value = val

    def setValue( a ):
       a.setValue( 5 ) 

so that works, but do you see how it is sort of polluting? I mean, the
interface is not so simple anymore. 

>Q: when does Python copy a reference and not a value?
>A: always!
ayeee! missed it! terribly sorry :)


ok, here's a third question, and it builds on the above inquiry
regarding a lack of an assignment operator.  What happens when I want
to do convert this for loop:

   for i,j in dest, src:
        dest[ i ] = src[ j ]

into a fast map() expression:

   map( dest.assign_operator, src )

like I mentioned -- there is no assignment operator to overload! 
the trouble here is that there may be more situations in which 
I cannot use  filter() to narrow down the list to only the elements
that I want and then copy it into another list with src.copy().  

Thanks!
Roey


From jpshell at bigfoot.com  Mon Jan 31 12:06:08 2000
From: jpshell at bigfoot.com (Jeffrey P Shell)
Date: Mon, 31 Jan 2000 12:06:08 -0500
Subject: Python1.5.2/Zope 2.1.3 on MacOS X Server notes
Message-ID: <B4BB2B2F.4F4F%jpshell@bigfoot.com>

These are some quick notes from building Python 1.5.2 with threads and
shared libraries on MacOS X Server 1.02.  As a note,

The MacNN Xclave (xclave.macnn.com) reports on configuring Python with
--with-host=powerpc-apple-netbsd .  I didn't try this option with what I did
below, but it couldn't hurt to try on older versions of Rhapsody and newer
developr releases of MacOS X who aren't picked up by the current configure
scripts.

This document, and future updates to it, can be found at
http://www.zope.org/Members/jshell/buildingZopeOnMacOSXServer

---
Building Zope 2.1.3 on MacOS X Server

Last time I told this story, I was having problems getting dynamic
libraries (.so files) to work with Python without configuring with
the --with-next-framework option, which didn't compile to the results
I wanted.  Below are the steps to building Python 1.5.2 with threads
and dynamic libraries on MacOS X Server and to building Zope 2.1.3.

Python::
First - configuring

./configure --with-threads --with-dyld

Next - MAKE

make LDFLAGS="-undefined suppress"
make install

The LDFLAGS="-undefined suppress" option is passed into the linker
and will suppress errors about undefined references.  Without this
setting, dynamic libraries will not get past the linker stage.  Make
install installs Python into the configured area (you should be root
to do this).


Zope::

A change has to be made in one of the files used for
building/configuring Zope that will hopefully be ironed out in the
near future.  For now, do the following

In inst/do.py, change line 135 from
   do('make')
to
   do('make LDFLAGS="-undefined suppress"')

(We'll find a way of fixing this in the future so this shouldn't
have to be edited.  There should be a Python configuration option
when building for setting these LDFLAGS into the default
Makefile.pre.in)


Then, for reasons currently unknown to me, cPickle issues have to be
resolved.  The Zope build process makes two cPickle.so files.  These
files appear to be identical and cause a conflict with each other at
runtime due to multiple symbol definitions, and they cause a
conflict with the cPickle.so that is built with Python.

Lastly:
cd lib/python/
rm cPickle.so

cd lib/python/ZODB/
mv cPickle.so /usr/local/lib/python1.5/lib-dynload/

You should be root to do the last step of moving the Zope cPickle.so
into the default Python area.


This is still too much to do, in my opinion, but it's the best way
I've found yet.  Any suggestions about how to get the LDFLAGS option
set at Python configuration time so that the generated Makefiles that
Zope uses will have them set, let me know.  I tried setting LDFLAGS
as an environment variable before running configure, but it uses
$LDFLAGS at the beggining of configuration to test the C compilor,
and since the '-undefined suppress' option is a linker option, an
error is returned and configuration stops.

Or, any suggestions or advice on how to set Python up to resolve
multiple definitions (due to multiple cPickle.so files) automatically
will be greatly appreciated.  The cPickle in Zope is newer than the
one in Python 1.5.2, I believe.

On my to-do list now (but pretty far down the list, sadly) is to
make this into a MacOS X Server package similar to other Zope
releases (meaning, the Zope package will be a full release with its
own Python).  The timeframe of this is uncertain.

jeffrey at Digicool.com



From schiotz1 at yahoo.com  Fri Jan  7 07:29:08 2000
From: schiotz1 at yahoo.com (Jakob Schiotz)
Date: Fri, 07 Jan 2000 12:29:08 GMT
Subject: Getting SWIG to work with C++/Python
References: <9mad4.42353$Ke.27158@news21.bellglobal.com>
Message-ID: <854m6a$j4o$1@nnrp1.deja.com>

In article <9mad4.42353$Ke.27158 at news21.bellglobal.com>,
  "Jason Maskell" <backov at nospam.csolve.net> wrote:
> Well, I bit the bullet and reinstalled VC5, and got SWIG to compile.
Now, I
> try "swig -c++ -python interf~1.cpp" and it chokes, hard. I try it
with the
> h, same thing.. I've tried variations of that, and I keep getting
loads of
> syntax errors, etc. Did it compile wrong? Is there some hidden syntax
I
> should be using? The other possibility is that since the files I'm
swigging
> are Borland C++ and SWIG itself was compiled with VC++.. But I can't
see
> that being a problem.
>
> Any help?
>
> Cheers,
> Jason
>
>

You should not run SWIG on the .cpp file, but on an interface (.i) file
that you write.  In some cases the .i file may just include the .h file,
but in most cases you need to cut and paste the declarations from the .h
file leaving out the stuff that SWIG does not understand.

There is an _excellent_ manual for SWIG at www.swig.org.  SWIG is a
rather advanced tool, and you really need to look at the manual to use
it.

I use SWIG a lot, I find it relatively easy to use, but I would not have
gotten anywhere without the manual.


Happy programming!

Jakob


--
Jakob Schiotz, CAMP and Department of Physics, Tech. Univ. of Denmark,
DK-2800 Lyngby, Denmark.  http://www.fysik.dtu.dk/~schiotz/
This email address is used for newsgroups and mailing lists
(spam protection).  Official email: schiotz @ fysik . dtu . dk


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


From loewis at informatik.hu-berlin.de  Tue Jan  4 14:57:08 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 04 Jan 2000 20:57:08 +0100
Subject: GCC and Unicode (OT)
References: <slrn87062a.sii.kc5tja@garnet.armored.net>
Message-ID: <p6qzoulehkb.fsf@informatik.hu-berlin.de>

kc5tja at garnet.armored.net (Samuel A. Falvo II) writes:

> Does anyone know of any way to make GCC produce 16-bit characters when
> encoding a Unicode string?  That is, if I have:
> 
> 	wchar_t *myText = L"ABCD";
> 	
> I want myText to point to the following:
> 
> 	0x0041 0x0042 0x0043 0x0044
> 
> instead of:
> 
> 	0x00000041 0x00000042 0x00000043 0x00000044
> 
> I'm using GCC under the Linux operating system.

When generating code for wide-string literals, gcc uses the type
wchar_t. The size of this type is defined by the platform (and its
ABI). gcc on Windows (and AIX?) uses 16-bit wchar_t, gcc on Solaris
and Linux uses 32-bit wchar_t. There is no way to change this.

Please be aware that this has nothing to do with Unicode, on an
abstract level. It is a wide character string, and the encoding of it
is implementation-defined.

Regards,
Martin


From kuncej at mail.conservation.state.mo.us  Tue Jan 18 13:28:20 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Tue, 18 Jan 2000 12:28:20 -0600
Subject: Monitoring file changes in NT
Message-ID: <3h2h4.362$B5.8315@news.more.net>

I'd like my python program to know when
directories and files have changed on
an NT system.

Could anyone share an example of how to use
win32file.FindFirstChangeNotification()
and its relatives?

Thanks.

  --Jeff





From sprawlsr at worldnet.att.net  Wed Jan 19 11:38:14 2000
From: sprawlsr at worldnet.att.net (Robert J. Sprawls)
Date: 19 Jan 2000 16:38:14 GMT
Subject: Looking for the FAQ
Message-ID: <slrn88bqi2.512.sprawlsr@TacHQ.com>

Hello,

	I'm new to this group. I run Linux and it has Python 1.5. I'm
an ANSI C programmer, learning C++ at my University, and I'm
interested in checking python out. Is there a good book that takes a
person from beginning to at least intermediate level? This is probably
something covered in the FAQ, so if you just point me to the FAQ, I'll
take it from there.

Robert



From stianh at ifi.uio.no  Mon Jan 17 02:08:39 2000
From: stianh at ifi.uio.no (Stian Husemoen)
Date: 17 Jan 2000 07:08:39 GMT
Subject: simple question about raw_input
References: <1Zyg4.1372$Pb4.11218@newsfeed.slurp.net>
Message-ID: <slrn885fvm.22l.stianh@birk82.studby.uio.no>

TechSpood wrote:
>I tried this but it doesn't quite work right...can you tell me what I am
>doing wrong?
>
>name=raw_input ("What is your name?")
>choice1=raw_input "hello, %s, would you like to do something?" % name

this might help:
choice1=raw_input("hello, %s, would you like to do something?" % name)


-- 

Stian Husemoen <stianh>


From backov at nospam.csolve.net  Tue Jan 11 19:59:10 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Wed, 12 Jan 2000 00:59:10 GMT
Subject: Please, someone tell me what's going wrong here. ;>
References: <GzLe4.54805$Ke.42098@news21.bellglobal.com> <udQe4.3161$Ef6.805744@news.shore.net>
Message-ID: <ylQe4.55014$Ke.43355@news21.bellglobal.com>

Michael P. Reilly <arcege at shore.net> wrote in message
news:udQe4.3161$Ef6.805744 at news.shore.net...
> Jason Maskell <backov at nospam.csolve.net> wrote:
> I would try to stay away from calling the module initializers, they
> are designed with a purpose in mind.  Go for the documented route:
>
> void initxyzzy()
>   {
>     (void)Py_InitModule("xyzzy", xyzzy_methods);
>   }

Hmm, ok. ImportModule is in the embed demo, which is why it's in my code as
well. Removing it seems to have no effect, so it's gone. ;>

>
> void TForm1::Button1Click(TObject *Sender)
> {
>     int a;
>     Py_SetProgramName(Application->ExeName.c_str());
>     Py_Initialize();
>     a=PyRun_SimpleString("import xyzzy"); /* import into the namespace */
>     a=PyRun_SimpleString("xyzzy.foo()");
>     Py_Finalize();
> }
>
> The module never got imported into the __main__ module, only
> initialized.

Thank you thank you thank you! I was staying away from the import command,
as I mistakenly thought it would try to load a module called xyzzymodule.o
or something. I guess I need to read up the docs some more. Anyway, that's
shattered the last little barrier for me, I can proceed with my embedding. I
owe you a pint! ;>

Cheers,
Jason





From gerrit.holl at pobox.com  Tue Jan 25 10:32:34 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 25 Jan 2000 16:32:34 +0100
Subject: Python compiler?
In-Reply-To: <000401bf661e$da1c4100$902d153f@tim>; from tim_one@email.msn.com on Sun, Jan 23, 2000 at 10:55:28PM -0500
References: <20000123092040.A2010@stopcontact.palga.uucp> <000401bf661e$da1c4100$902d153f@tim>
Message-ID: <20000125163234.C2497@stopcontact.palga.uucp>

Tim Peters wrote on 948664528:
> [J Dubal]
> > Does [Python] have a compiler for Linux and/or windows that can
> > produce native executables?
> 
> [Gerrit Holl]
> > Yes, it has one. It's included in the Python distribution,
> > Tools/freeze.  It's called freeze, and with it, you won't
> > need any Python interpreter.
> 
> Just to prevent any possible misunderstanding, freeze works by sticking a
> copy of the interpreter into the executable.

Ah, so *that's* why it's so big!

> This is probably not what J
> Dubal had in mind, but that's life.  A Python VM is needed to execute Python
> bytecodes just as a Java VM is needed to execute Java bytecodes, and that's
> what "the interpreter" is.

But isn't it possible to write a *compiler* for Python?

> the-natives-are-restless-tonight-ly y'rs  - tim

The natives?
Native Pythoneers?

did-I-already-announce-python.nl.linux.org/gemeenschap/drukste.html?-ly

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From arcege at shore.net  Sun Jan 30 08:55:41 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Sun, 30 Jan 2000 13:55:41 GMT
Subject: mimecntl 1.4
Message-ID: <xpXk4.1061$GX6.283443@news.shore.net>

I have created a new version of my mimecntl module based on patches
from users.  The new version is 1.4.

Mimecntl is a module which encapsulates MIME documents as mutable
objects and subobjects.  Multipart content-type behaves like UserList,
most other times act like files.  The original purpose of this module
was to create a single mechanism for reading and written MIME files
(I found it hard to integrate mimetools and MimeWriter).

The module can be downloaded from
<URL: http://starship.python.net/~arcege/modules/mimecntl.py>

What has been fixed since the last announced release:
* a new method of MIME_document has been added to check file objects;
  this was made as a method to be overridden by subclasses if necessary;
  such objects require a 'seek' method;
* MIME field content now handle case-insensitive;
* a bug copying a Field instance;
* added an old fix that I mistakenly never included to truncate
  then decoding/encoding.

Thanks to Josef Sachs, Bill Bumgarder and Riccardo Murri for their
aid.

  -Arcege



From aahz at netcom.com  Tue Jan 25 15:32:23 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jan 2000 20:32:23 GMT
Subject: re module -- case insensitive matching.
References: <roy-3281DF.15202825012000@netnews.nyu.edu>
Message-ID: <86l18n$7ob$1@nntp9.atl.mindspring.net>

In article <roy-3281DF.15202825012000 at netnews.nyu.edu>,
Roy Smith  <roy at popmail.med.nyu.edu> wrote:

>Is there any difference between:
>  re.compile ('[a-zA-Z]*')
>and
>  re.compile ('[a-z]*', re.I)

Yup.  Use the former; it's faster.  For the latter, the string has to be
converted before the regex touches it.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From fredrik at pythonware.com  Sat Jan 22 03:50:58 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 22 Jan 2000 09:50:58 +0100
Subject: Python dumps core with this.. how come?
References: <38891316.393423@news.erols.com>
Message-ID: <005301bf64b5$d2f234d0$f29b12c2@secret.pythonware.com>

Roey Katz <katz at glue.umd.edu> wrote:
> When I import this module, the interpreter segfaults. 
>
> Thiis module provides a little guard mechanism to keep track of
> changes to registered variables. The error occurs when you ty to
> instantiate class 'basic_monitor'.

looks like a recursive call in getattr (if "res" is
not already available in the instance, it asks
__getattr__ to look it up).

make sure to define "res" in the __init__ function,
and everything will work as you expected.

> Also, I have another question (very old, and I can't find this in the
> FAQ):  I want to explicitly take a reference to an integer.

1. reset your brain ;-)
2. variables are named references, not small boxes
   that hold data values
3. *all* variables in python are references.  there
   is no other thing.
4. integers cannot be modified in place.

in other words, Python always copies references, but
you cannot modify objects in place unless their inter-
face allows you to (numbers, tuples and strings cannot
be modified in place, while lists and dictionaries can --
the only difference is that the former doesn't provide
any methods that let you modify them...)

> is there some sort of rule-list to follow to figure out when
> exactly Python will take a reference (like assigning from an
> object) or a literal value (int's)?

yes, and it's very simple:

Q: when does Python copy a reference and not a value?
A: always!

related FAQ entries:
http://www.python.org/doc/FAQ.html#4.35
http://www.python.org/doc/FAQ.html#4.38
http://www.python.org/doc/FAQ.html#4.50
http://www.python.org/doc/FAQ.html#4.89
and probably a few more.

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! &smiley; -->




From grant at nowhere.  Thu Jan 20 09:01:04 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 20 Jan 2000 14:01:04 GMT
Subject: olemsg32 module used in win32com exchange server example?
References: <slrn8897gs.3cm.grant@grante.comtrol.com> <864ult$fot$1@nnrp1.deja.com> <slrn88ceoc.5ga.grant@grante.comtrol.com> <38863DB3.7FAEC820@callware.com>
Message-ID: <slrn88e58v.67b.grant@grante.comtrol.com>

In article <38863DB3.7FAEC820 at callware.com>, Ivan Van Laningham wrote:

>> >>  2) Does the win32com stuff talk directly to the Exchange
>> >>     server, or does it go through a local copy of Outlook?
>> >
>> >It goes through the MAPI DLLs - the same DLLs that the local copy of
>> >Outlook itself uses.
>> 
>> Cool.  Does that mean that I don't even need to have Outlook
>> installed in order for a Python/COM/MAPI program to work?
>
>I suspect that is true, but there's a gotcha.  Unless I'm wrong, both
>NetScape Mail and Outlook use MAPI dll's, but they expect different
>functions to be found there.  So, when I want to use Outlook to retrieve
>mail, I can configure everything else correctly but I have to change the
>MAPI.DLL to the one the Outlook expects or it barfs.  Then put it back
>when I want to use NetScape. ...

<Boggle>

There some sort of Cosmological Constant that keeps the
MS-Windows universe from collapsing under its own weight and
complexity.

[a whoosh and muted thud]

Never mind.  I see it just did.

-- 
Grant Edwards                   grante             Yow!  WHOA!! Ken and
                                  at               Barbie are having TOO
                               visi.com            MUCH FUN!! It must be the
                                                   NEGATIVE IONS!!


From bsb at winnegan.de  Sun Jan  2 09:27:06 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 02 Jan 2000 15:27:06 +0100
Subject: Pseudo-SQL: searching for Parser/Rewriter
In-Reply-To: "hartmut Goebel"'s message of "Sun, 02 Jan 2000 14:58:30 +0100"
References: <16600378@oberon.noris.de>
Message-ID: <87vh5c35xh.fsf@baal.winnegan.de>

"hartmut Goebel" <hartmut at oberon.noris.de> writes:

> Hello everyone,

> I'm looking for a general purpose tokenizer/parser/rewriter for
> Python. If a parser is available, it best should return a simple
> parse tree.

As soon as the search engine on www.python.org is up again, look out
for Gadfly by Aaron Watters, it's an SQL database engine written in
Python.

[...]

/siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******



From cfollett at gac.edu  Fri Jan 21 13:49:46 2000
From: cfollett at gac.edu (Charles Follett)
Date: Fri, 21 Jan 2000 12:49:46 -0600
Subject: Second plea: help with multicast sockets
Message-ID: <200001211849.MAA09072@giant-curled.mcs.gac.edu>

Whenever I attempt to select a multicast group, like when running
Demos/sockets/mcast.py, I get:

  File "mcast.py", line 89, in openmcastsock
    s.setsockopt(IPPROTO_IP, IP_ADD_MEMBERSHIP, mreq)
socket.error: (22, 'Invalid argument')

Multicast works on my machine, and 'mcast.py -s' sends just fine
(verified with a sniffer.) What is preventing me from listening? Also,
I have two ethernet interfaces on my machine, and I was wondering how
to manually choose what interface is used. Thanks in advanced.

charley



From tseaver at starbase.neosoft.com  Tue Jan 18 15:44:48 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 18 Jan 2000 14:44:48 -0600
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <861jm5$nrj$1@mach.vub.ac.be> <zqGEOEqNnxHC8gbhL5piinIM1bGb@4ax.com> <862egg$4br$1@nntp3.atl.mindspring.net>
Message-ID: <D979B77DECF4C994.A2825C36F10F2D95.8DB19B8AF5090A6D@lp.airnews.net>

In article <862egg$4br$1 at nntp3.atl.mindspring.net>,
Aahz Maruch <aahz at netcom.com> wrote:
>In article <zqGEOEqNnxHC8gbhL5piinIM1bGb at 4ax.com>,
>Dennis Lee Bieber  <wlfraed at ix.netcom.com> wrote:
>>
>>	Mysterious errors usually means a mistake in the indentation
>>applied. And to me, the indentation is easier to pick out then an
>>"endif" on the wrong line...
>
>Unless you've mixed tabs and spaces.  :-(

Tabs in source files are devil's spawn, plain and simple. :)

Demurr'ly,

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From controld at hotmail.com  Sun Jan  2 15:44:26 2000
From: controld at hotmail.com (Mike Apted)
Date: Sun, 02 Jan 2000 12:44:26 PST
Subject: DB API
Message-ID: <20000102204426.5130.qmail@hotmail.com>

Check out:

http://www.webtechniques.com/archives/2000/01/jepson/

Covers different DB's and their pros and cons.

>From: Simon Faulkner <News at Titanic.co.uk>
>To: python-list at python.org
>Subject: Re: DB API
>Date: Sun, 02 Jan 2000 19:44:42 GMT
>
>Thanks for the reply...
>
> >> with most^H^H^H^Hall database servers<<
>Is this a typo or am I missing something?
>
>I have my database working online!!!  GREAT!
>
>Which database do you recommend?  MySQL is free (or v. cheap)
>
>I am keen to try everything.
>
>TIA
>
>Simon
>
>On 2 Jan 2000 12:13:17 GMT, boud at rempt.xs4all.nl (Boudewijn Rempt)
>wrote:
>
> >Simon Faulkner <News at titanic.co.uk> wrote:
> >> Hi All,
> >
> >> Where can I find the instructions for the DB API?
> >
> >> I have managed to figure out the .connect() and .fetchone() methods
> >> but am flagging with the rest.
> >
> >> The Python Database API Specification 2.0 document does not realy show
> >> me the methods or examples that I need to get started and the MySQLdb
> >> documentation is deliberatly vague to discourage use of the _mysql.
> >> methods (so you use the proper API?)
> >
> >If you're using to MySQL, then you'd better not go to far using all
> >the options and stuff from DB-API II - if you use connect, execute and
> >fetchall, you've exhausted the capabilities of the database. No need
> >for commit, rollback or things like that - MySQL doesn't support them.
> >There's one area where MySQL has functionality beyond DB-API II, and
> >that's in the area of autoincrement columns, where you can get quite
> >easily get back the autoincrement values back after insertion.
> >
> >Also, with most^H^H^H^Hall database servers, what you really need to do 
>is
> >ask for a few rows, and fetch them immediately from the server. Keeping
> >large resultsets on the server and nibbling piecemeal from them is a
> >recipe for performance problems.
>
>Simon Faulkner
>--
>http://www.python.org/mailman/listinfo/python-list

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com




From boud at rempt.xs4all.nl  Sun Jan  2 18:02:39 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 2 Jan 2000 23:02:39 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com> <84nfct$bal$1@news1.xs4all.nl> <umav6ssq7uubv4gpb192lp2fs1jcan6hc2@4ax.com>
Message-ID: <84olef$5j1$1@news1.xs4all.nl>

Simon Faulkner <News at titanic.co.uk> wrote:
> Thanks for the reply...

>>> with most^H^H^H^Hall database servers<<
> Is this a typo or am I missing something?

No, not really ;-). When I was typing this, I suddenly realized that
I don't know of any database server where it's a good idea to slurp
thousands of rows over the network to the client. (Still, it is, or was
company policy at the place I work at - according to the lead programmer
'Nothing is as fast as core memory. We need all the data in the core
memory of the client. Heck, they'll start up once a day, at most - it
won't matter if we put 130 mb in VB collections - it'll be fast!' That
programm will never be started again... It's canceled.)

> I have my database working online!!!  GREAT!

Congratters!

> Which database do you recommend?  MySQL is free (or v. cheap)

MySQL is what I use myself. I'm used to Oracle (and have it installed,
indeed - but for most applications it's really overkill). Postgres is
nice - but it doesn't have a DB-API II interface, as far as I know.

> I am keen to try everything.

So am I, but when I saw that SuSE Linux comes with Ingres, Postgres,
MySQL, DB-2 and a few others (Sybase or Informix, not to mention
Flagship), I decided not to bother ;-).

-- 

Boudewijn Rempt  | http://denden.conlang.org


From mskott at image.dk  Sun Jan 23 16:02:58 2000
From: mskott at image.dk (Martin Skøtt)
Date: 23 Jan 2000 22:02:58 +0100
Subject: Simple socket loop
Message-ID: <ln5gqz5p.fsf@jumbo.skott.dk>

Hi
I'm trying to create a simple socket listener interface but i don't
know how to keep it listening. I have the following server loop (from
the python library reference):
from socket import *
HOST = ''
PORT = 50007
s = socket(AF_INET, SOCK_STREAM)
s.bind(HOST, PORT)
s.listen(2)
conn, addr = s.accept()
print 'Connected by', addr
while 1:
    data = conn.recv(1024)
    if not data: break
    conn.send(data)
conn.close()

After having communicated with the client the server stop! How can I
prevent that and keep it listening (and answering)? I guess I have to
put it inside some kidn of loop but how. Please if you could give me a
working example (or a link to one) it would be great because i'm still
new to Python.

-- 
Many regards
Martin Sk?tt
mskott at image.dk

"Every time I learn something new it pushes some old suff out of my brain"
Homer Simpson


From dworkin at ccs.neu.edu  Thu Jan 20 15:34:41 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 20 Jan 2000 15:34:41 -0500
Subject: How to store a function pointer in class?
In-Reply-To: "Sami Hangaslammi"'s message of "Thu, 20 Jan 2000 21:23:40 +0200"
References: <867j5p$qvq$1@mordred.cc.jyu.fi> <vndaem01s1r.fsf@betelgeuse.ccs.neu.edu> <867nbb$1ho$1@mordred.cc.jyu.fi>
Message-ID: <vnd66wo1nz2.fsf@betelgeuse.ccs.neu.edu>

"Sami Hangaslammi" <shang.remove_edu at st.jyu.fi.edu> writes:

> So I wan't to store function references into class variables and
> retrieve them unchanged.

Oh.  I appear to have misunderstood your real goal here.

You could store the function inside a mutable class data member.

For instance:

>>> class Test:  
...   class_attribute = []

>>> def test_func(x):  
...     print "test_func called with",x  

>>> Test.class_attribute.append(test_func)

>>> x = Test.class_attribute[0]

>>> x(1)
test_func called with 1

Is that something like what you were looking for?

-Justin

 



From anders.eriksson at morateknikutveckling.se  Wed Jan 12 07:50:23 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Wed, 12 Jan 2000 13:50:23 +0100
Subject: checking for file existance
Message-ID: <JHh8OAp=9qpii6CdulrM3o2KSTd9@4ax.com>

Hello!

How do I check if a file exist on a webserver? I have tried
urllib.urlopen() but it never give any exception IOError. Is there an
access() that takes an url?

// Anders



From ovazquez at sprint.ca  Mon Jan 17 02:31:10 2000
From: ovazquez at sprint.ca (Orlando Vazquez)
Date: Sun, 16 Jan 2000 23:31:10 -0800
Subject: Very Easy question concerning input
References: <D2zg4.1375$Pb4.11132@newsfeed.slurp.net>
Message-ID: <3882C53E.F8550E02@sprint.ca>

how about this? (i assume in the last `raw_input' you meant to put the
parantheses...)

>>> name = raw_input("What is your name? ")
What is your name? orlando
>>> name
'orlando'
>>> choice1 = raw_input("welcome, %s would you like to do something?" % (name,))
welcome, orlando would you like to do something?

TechSpood wrote:
> 
> can someone tell me how to use a variable from a previous line and get a
> variable from the currentline using raw_input, for example
> 
> I am trying to make a simple program that ask the user his name then says
> "welcome (name) would you like to do something"
> 
> I tried this...but it doesn't work right...
> 
> name=raw_input ("What is your name?")
> choice1=raw_input "welcome, %s, would you like to do something?" % name
> 
> doing this displays a %s instead of the persons name, can you tell me how to
> fix this?
> 
> thanks

--
Orlando Vazquez


From suzuki611 at okisoft.co.jp  Mon Jan 31 18:25:44 2000
From: suzuki611 at okisoft.co.jp (Hisao Suzuki)
Date: 01 Feb 2000 08:25:44 +0900
Subject: lib reference in Unix man format
References: <200001281339.IAA12329@trm.bmo.com>
Message-ID: <u3k8kpkemf.fsf@ares.sys1.ngo.okisoft.co.jp>

> If each built-in function has its own man page, I have to know the function name 
> before I can find its man page. This will be like the manual of Unix system 
> calls. If I don't remember the name of message send operation, i.e., msgsnd, I 
> have to use 'man -s 2 intro' to figure out.

% python
Python 1.5.2 (#1, Apr 23 1999, 11:53:40)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> dir()
['__builtins__', '__doc__', '__name__']
>>> dir(__builtins__)
['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']
>>> 

Hope this helps.

--===-----========------------- Sana esprimo naskas sanan ideon.
SUZUKI Hisao            suzuki611 at okisoft.co.jp, suzuki at acm.org.


From aa8vb at yahoo.com  Wed Jan  5 14:15:20 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 5 Jan 2000 14:15:20 -0500
Subject: Python and GUI options
In-Reply-To: <slrn877386.lok.zeitlin@seth.lpthe.jussieu.fr>
References: <84lreq$hib$1@nnrp1.deja.com> <386fa6b1.58568155@news.telus.net> <20000105085647.C9441@vislab.epa.gov> <slrn877386.lok.zeitlin@seth.lpthe.jussieu.fr>
Message-ID: <20000105141519.A13656@vislab.epa.gov>

Vadim Zeitlin:
 |
 |Sun, FreeBSD, Linux, HP-UX and AIX seem to work fine most of the time
...
 |>wxWindows/wxGTK/wxPython requires someone with experience and persistence
 |>to help it along on a standard SGI UNIX box, in compilation and
 |>configuration.
 |
 | There is one problem inherent to SGI: the native make doesn't support VPATH
 |(and never will according to SGI docs) while wxWindows makefiles require it,
 |so you must install GNU make. Except for this, however, there should be no
 |problems

I wasn't even thinking of the make/gmake part.  Using gmake was the first
thing I did (after asking Robin how I could make it use gmake; it wasn't
apparent).  There are other issues I was referring to.  Some of them I've
swapped mail with Robin Dunn on.

>From a user standpoint, it would be useful if the build system were changed
to a GNU configure sense-and-testwhat-you-have scheme rather than a
configure-this-file, put-overrides-in-this-file, and see-what-breaks
scheme.

Re the other problems.  If build system changes don't do away with it (I
hope they do), more docs on creating the build.local file are a must.
Also, I had to copy this around when I was building wxPython and then its
assorted modules because it didn't look for it except in $PWD.  I had to
make it compile in an RPATH so wxPython could find wxWindows.  This is
probably there for Solaris (-R) but it's not there for SGI (-rpath) since
you're not testing on that OS.  Also, wxGTK and wxPython sense different
compilers (SGI vs. GCC).  I had to coerce them to use the same ones.  I
configured with --with-opengl and it didn't pull in the GL canvas.  Also,
There are other issues, but that's what comes to mind glancing at my
build.local.

-- 
Randall Hopper
aa8vb at yahoo.com
-------------- next part --------------

  # wxPy makefiles expect GNU make
MAKE         = "gmake"
  # Needed on IRIX because Python's mymalloc.c incorrectly defines NULL for C++
OTHERCFLAGS  = "-DNULL=0"
  # Where wxPy puts its .so modules
HELPERLIBDIR = "/home/rhh/software/pythonutil/wxPython-2.1.11"
  # Link with an RPATH (so imports just "work", without LD_LIBRARY_PATH hacks)
OTHERLFLAGS  = "-rpath /home/rhh/software/wxGTK-2.1.11/lib:/home/rhh/software/pythonutil/wxPython-2.1.11"
  # Force use of the same compilers as were used to build wxWindows -- 
  #   otherwise we get undefined symbols because compilers mangle the 
  #   names differently (the joys of C++...)
CCC          = "CC"
CC           = "cc"
  # WXPYSRCDIR isn't really needed unless not under the wxWin build dir.
  #   But it can't hurt.
WXPSRCDIR    = "/scratch_local/rhh/BUILD/wxPython/wxGTK/utils/wxPython/src"


From aa8vb at yahoo.com  Fri Jan 21 10:22:44 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 21 Jan 2000 10:22:44 -0500
Subject: Exiting program
In-Reply-To: <20000121101100.A184338@vislab.epa.gov>
References: <38873769.5048399@122.0.0.250> <20000121101100.A184338@vislab.epa.gov>
Message-ID: <20000121102244.A184499@vislab.epa.gov>

Randall Hopper:
 |volumeregeling:
 | |    sys.exit()                        <-- for some reason this also
 |
 |Keep in mind that this doesn't terminate your program.  It merely raises
 |a SystemExit exception.  ... it can be caught and ignored, which is most
 |likely what is going on:

Should have mentioned this, but in case you have more important fish to fry
right now, just call:

     os._exit(0)

to force an unfriendly, immediate exit (bypassing the exception propagation
mechanism).

-- 
Randall Hopper
aa8vb at yahoo.com



From emile at fenx.com  Sun Jan  9 14:56:18 2000
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 9 Jan 2000 11:56:18 -0800
Subject: Modifying Global Variables from Within a Function
Message-ID: <0ce801bf5adb$9eaf75e0$01ffffc0@worldnet.att.net>

use global ( see below)

Josh Tomkins <josht at iname.com> wrote in message
news:<8EB69CD0Ejosuetompkins at 209.208.0.7>...
> Hello Everyone!
>
> This is, perhaps, a newbie type of question (I am new to Python), but
> here goes:
>
> Can you modify a global variable from inside a function?  Basically
what
> I want to do is this:
>
> a = 0 #global variable
>
> def func1(None):

       global a # or whatever

>      a = 5 #or whatever
>      return
>
> def func2(None):
>      #do stuff here
>      func1(None)
>      #do more stuff with the newly modified a
>      return
>
> this is a contrived example.  In reality, I will need to modify
several
> variables, changing them to zero.  Is this possible?
>
> Thanks,
>
> Josue
> --
> http://www.python.org/mailman/listinfo/python-list
>
>

Emile van Sebille
emile at fenx.com
-------------------







From gerrit.holl at pobox.com  Tue Jan 18 15:22:29 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 18 Jan 2000 21:22:29 +0100
Subject: Suggest: .get() method on lists.
Message-ID: <20000118212229.A542@stopcontact.palga.uucp>

Hello,

What about having a get() method on lists?
I thought it would be useful, for example, for sys.argv. At the moment,
I have this code:

file = sys.stdin
if len(sys.argv) > 1:
    file = sys.argv[1]

But this, would be much easier to read:
file = sys.argv.get(1, sys.stdin)

Of course, it's also possible to use multiple inheritance on
UserList and UserDict creating a DictionairyList for sys.argv,
because it's the only place I can think of such a feature would
be a Good Thing (TM).

it's-funny-to-browse-the-old-archives-and-see-who-already-was-here-ly y'rs - ger

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From sameerc1 at yahoo.com  Thu Jan  6 23:39:51 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Thu, 6 Jan 2000 20:39:51 -0800
Subject: python with ASP
Message-ID: <000001bf58c9$8c26ff60$39f4fea9@desktop>

I get the following error with the code listed below:

Python ActiveX Scripting Engine error '80020009' 

invalid syntax 

/pytest2.asp, line 4 

sStr = '<b> Got Title</b>'
---^


So this code does what you want.

<%@ LANGUAGE = Python %>
<%
if len (Request.Form("title")):
sStr = '<b> Got Title</b>'
else:
sStr = '''<FORM ACTION="./pytest.asp" method="POST">

<b>Title:&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</b><BR>
<input type="text" value="" size=60 name="title"><P>
<input type=submit value="Submit">
</form>'''
%>
<html>
<%= sStr %>
</html>

    -Eugene






__________________________________________
NetZero - Defenders of the Free World
Get your FREE Internet Access and Email at
http://www.netzero.net/download/index.html



From soundhack at hotmail.com  Wed Jan 12 07:06:31 2000
From: soundhack at hotmail.com (Robert)
Date: Wed, 12 Jan 2000 07:06:31 -0500
Subject: PIL 1.0 bugs? (TIFF, Image)
Message-ID: <387c6eb2$0$22474@senator-bedfellow.mit.edu>

Am I doing something wrong? I tried to read and display a simple TIFF file
and came across two errors that I had to hack/go around to get it to work.

>import Image
>im=Image.open("test.tif")
Here there is a ValueError exception in the getscalar method in
TIFFImagePlugin.py

looking at the code, apparently the plugin assumes that the SOURCEFORMAT tag
is a scalar, which is true when the image is grayscale, but isnt for RGB
images (as per the TIFF specfication)

after removing the test for "scalar"ity, I then ran into a problem when I
tried
>im.show()

again looking at the traceback, there was a problem in the _getdecoder and
_getencoder "codec factories", specifically an apply call.
Again, I did an ugly hack to get it to work and now it works as expected.

This is all very basic stuff, (so much that I'm surprised it wasnt
caught and fixed before), so i would appreciate anyone telling me if I'm
missing something simple here.

Robert





From mgushee at havenrock.com  Thu Jan  6 08:03:10 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 06 Jan 2000 08:03:10 -0500
Subject: XML to dictionary?
References: <8518kq$59n$1@nnrp1.deja.com>
Message-ID: <m3puvf5p4h.fsf@havenrock.com>

johnjensen at my-deja.com writes:

> I have a relatively simple XML question, but I can't seem to figure out
> how to do it simply and easily with the xmllib module.

If you are only doing the simplest of XML tasks and don't care about
the longevity of the code, xmllib may work for you. But I gather that
it is considered very inadequate for any kind of serious XML
processing. You should look into the XML-SIG's XML package. And you might
want to join the XML-SIG mailing list, too. Look under SIGs --> XML at 
www.python.org.

> How can I quickly and easily stuff this into a nested dictionary, so
> that:
> 
> print dict['Newsgroup']['comp.lang.python']['Message1']['Subject']
> spits out 'Hi there'

What do you consider quick and easy? :)

AFAIK there is no ready-made function to do this for you. But if you
get the data parsed properly, the rest should be easy. You might
consider using Pyxie, which provides a higher-level interface to some
of the XML (SIG) package's functionality. You could do something like:

from pyxie import File2PYX    # PYX is the notation pyxie uses for
import string		      # parsed XML

f = File2PYX('hithere.xml')
data = f.read()
f.close()
data = string.split(data)

dict = {}
for line in data:
    if line[0] = '(':         # '(' at the beginning of the line in 
			      # PYX indicates an element

... hmm. This might be a problem if you don't know exactly how many
levels deep the elements will be nested. But nothing you can't solve
in an hour or so, I'm sure.

By the way, wouldn't you rather have:

<Newsgroup name="comp.lang.python">

?

Hope this helps a bit.


-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From grant at nowhere.  Wed Jan 19 13:06:05 2000
From: grant at nowhere. (Grant Edwards)
Date: Wed, 19 Jan 2000 18:06:05 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com> <010801bf62a5$3e6c6400$f29b12c2@secret.pythonware.com>
Message-ID: <slrn88bv8d.4sm.grant@grante.comtrol.com>

In article <010801bf62a5$3e6c6400$f29b12c2 at secret.pythonware.com>, Fredrik Lundh wrote:

>footnote: MSDN is available on-line, at:
>    http://msdn.microsoft.com/library/
>
>you'll find the MAPI docs under "Platform SDK ->
>Messaging and Collaboration Services"

Cool.  Thanks again.  BTW, the example code that printed out
the subject headers from inbox messages worked.  It should be
pretty easy to grab the message body (and attachments,
eventually) and ship them out via SMTP.  It seems foolish to
re-invent "fetchmail", but that's what I'm going to do. Should
it become necessary, it should be pretty simple to write a
conduit that runs in the other direction: receiving messages
via SMTP and sending them on with MAPI.  As usual, the error
cases are where 90% of the work will be.

Currently, I don't have to worry about the "send" direction:
our Exchange Server has the SMTP server enabled since that's
how we get e-mail from the outside world.  Otherwise, BOFH
would probably shut that off too since SMTP wasn't invented by
Microsoft.  [I haven't checked lately, maybe Bill has declared
that SMTP was invented by MS.]

I'm going to use mutt for my MUA, and there's nothing BOFH can
do about it since I will be using Exchange as the MTA via MAPI!

[Pardon my gloating -- nobody around here sympathizes with my
dissatisfaction with Outlook.] 

Besides, it will give that clunky P166 that sits under the
table in the corner of my office something to do.

-- 
Grant Edwards                   grante             Yow!  I'd like TRAINED
                                  at               SEALS and a CONVERTIBLE on
                               visi.com            my doorstep by NOON!!


From wlfraed at ix.netcom.com  Wed Jan 19 23:07:51 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Wed, 19 Jan 2000 20:07:51 -0800
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <65118AEEFF5AD3118E8300508B124877073D23@webmail.altiris.com>
Message-ID: <loiGOLTcqnm1hbxIn9kUCUHHu4gR@4ax.com>

On Wed, 19 Jan 2000 14:33:06 -0700, Mike Steed <MSteed at altiris.com>
declaimed the following in comp.lang.python:


> Seminar posters.  I think that's what we're seeing.  They attend seminars
> where they are carefully taught how to write posts that distract and annoy
> the readers of the target news group/mailing list/whatever.  The seminars,
> of course, are sponsored by vast conspiracies that aim to popularize severe
> hackery in its various forms, such as four-letter languages beginning with
> P.
> 
> Okay, I'm kidding, don't get upset.

	May not be that far off... New semester at Universities --
students taking language design courses or similar and trying to
simplify their lives by making all languages look the same... (Isn't
that the problem PL/1 had -- incorporating features from everywhere?)

	Now there is a scary thought... a dialect of FORTRAN with PERL
notation...

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From thomas at xs4all.net  Mon Jan  3 03:48:52 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 3 Jan 2000 09:48:52 +0100
Subject: Python language spec
In-Reply-To: <E124x1i-00053N-00@ethereal>; from nolan_d@bigfoot.com on Sun, Jan 02, 2000 at 08:15:18PM -0600
References: <E124x1i-00053N-00@ethereal>
Message-ID: <20000103094852.B5669@xs4all.nl>

On Sun, Jan 02, 2000 at 08:15:18PM -0600, Nolan Darilek wrote:

> I have a long-standing tradition of taking up personal projects in
> which I bite off much more than I can chew. The end result is that
> several weeks later I emerge with a bit more knowledge, and either a
> semi-working product or much less hair on my head. :) My most recent
> attempt is a Python interpreter written to compile a subset of Python
> into Muq (http://muq.org/~cynbe/muq/) bytecodes for execution natively
> on the server platform.

Not to spoil your fun at playing with yacc & parsing python, but you dont
really need to tackle yacc/parsing yourself to embed an interpreter. Are you
sure you want Muq to take on only a subset of Python expressions, and not a
subset of python builtin modules/functions ? If the latter, you dont need to
do any parsing, you can build a modified __builtins__ namespace and execute
user code in that. 

Markus Stenberg wrote a small patch to embed Python into TinyMUSH 3.0 and
provide basic MUSH functionality to the python code being executed. It's not
up to date, having been written for TinyMUSH 3.0 beta 5, while beta 19 is
the current version, and i doubt you can cut&paste it into Muq in any case,
but the main function is 1 600 line file, which demonstrates at least one
way to limit access to Python libraries.

Of course, if you really want to tackle python code directly, for instance
to avoid common problemns like 'a.apped(b);b.append(a)' and 'while 1:
pass', the patch can't be more than a small hint. If you do that, though,
keep me posted, i've been wanting to work around those problems for a long
time, for embedding purposes, and have only been lacking time to do so ;P

mushi-mushi-ly y'rs,

-- 
Thomas Wouters <thomas at xs4all.net>

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



From annis at biostat.wisc.edu  Fri Jan 14 08:48:15 2000
From: annis at biostat.wisc.edu (William Annis)
Date: Fri, 14 Jan 2000 07:48:15 -0600 (CST)
Subject: Emacs question
In-Reply-To: <009501bf5e69$f908e530$f29b12c2@secret.pythonware.com>
	(fredrik@pythonware.com)
References: <85kr9b$2kh$1@nnrp1.deja.com> <002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com> <jtu2kh52y6.fsf@wazor.biostat.wisc.edu> <009501bf5e69$f908e530$f29b12c2@secret.pythonware.com>
Message-ID: <200001141348.HAA09968@wazor.biostat.wisc.edu>

 From: "Fredrik Lundh" <fredrik at pythonware.com>:
 >William Annis wrote:
 >> "Fredrik Lundh" <fredrik at pythonware.com> writes:
 >> 
 >> > # -*-python-*-
 >> > 
 >> > or
 >> > 
 >> > #  -*- Mode: Python; tab-width: 4 -*-
 >> 
 >>         Those all work, but as I recall those work only on the first
 >> line of the file.  A quick test seems to confirm this.
 >
 >well, Daryl asked for something to put "at the
 >top of the file", so I didn't think I had to point
 >that out ;-)

        Well, that's certainly true.  I suppose I should have
mentioned in passing that I've had trouble with some operating systems
trying to parse this:

#!/usr/bin/env python -- # -*- python -*-

Solaris has no problem with this.  Linux bleats bitterly about it,
which was something of a nightmare when we started using Linux last
year.  I assumed Daryl wanted to write executable scripts and forgot
to mention the line above is not exactly portable, and thus my
preference for the other trick.

        I love it when my mind skips ahead to the punchline.  Perhaps
it's time to work on reducing my caffeine intake again... :)

--
wm



From claird at starbase.neosoft.com  Sun Jan 23 23:47:07 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Jan 2000 22:47:07 -0600
Subject: Tkinter, threads, mainloops, etc
References: <pcobt6c635a.fsf@math.ntnu.no> <pco901g5k6k.fsf@math.ntnu.no>
Message-ID: <3DD253A909A4E545.9A16947EC86B8249.BD16F5CF20E0BE2D@lp.airnews.net>

In article <pco901g5k6k.fsf at math.ntnu.no>,
Harald Hanche-Olsen  <hanche at math.ntnu.no> wrote:
>I am following up to my own posting here, since everybody else is
>probably at the python conference anyway, and/or getting drunk at the
>bar.  Some reckless experimentation shows that, indeed, I can do
>graphs in Tkinter without ever calling any mainloop method - however,
>I have to throw in the occasional call to an update method, or nothing
>is updated until I get the command prompt back.
>
>I'll most likely be doing this live in a seminar in a couple of weeks,
>so if there are any gotchas I'd like to know about them before it's
>too late.
			.
			.
			.
The Tcl crowd often fusses over (generally subtle) errors
in the "notifier".  Search the comp.lang.tcl archives for
that keyword to identify particular releases and platforms
which have been impaired in the past.  Apart from that, yes,
it's not considered too radical for plucky individuals to
fiddle with the event loop.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From samschul at pacbell.net  Sun Jan 16 17:03:41 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Sun, 16 Jan 2000 22:03:41 GMT
Subject: SERIAL I/O ON WIN32
References: <Qvpg4.581$253.12322@news>
Message-ID: <85tf7n$28l$1@nnrp1.deja.com>

In article <Qvpg4.581$253.12322 at news>,
  "bob" <bob at no.com> wrote:
> Hi there...newbie to Python here.
>
> I need to pass information across a COM port. I have the serial-i/o
module
> but no clue on how to use it.
>
> Thanks for any and all help!
>
> Bob
>
>

I do not know which module you have  but try the following link. These
modules are documented.

http://www.vex.net/parnassus/apyllo.py?so=d&find=serial+IO


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


From ovazquez at sprint.ca  Mon Jan 17 02:27:55 2000
From: ovazquez at sprint.ca (Orlando Vazquez)
Date: Sun, 16 Jan 2000 23:27:55 -0800
Subject: Very Easy question concerning input
References: <D2zg4.1375$Pb4.11132@newsfeed.slurp.net>
Message-ID: <3882C47B.39A16495@sprint.ca>

how about this? (i assume in the last `raw_input' you meant to put the
parantheses...)

>>> name = raw_input("What is your name? ")
What is your name? orlando
>>> name
'orlando'
>>> choice1 = raw_input("welcome, %s would you like to do something?" % (name,))
welcome, orlando would you like to do something?

TechSpood wrote:
> 
> can someone tell me how to use a variable from a previous line and get a
> variable from the currentline using raw_input, for example
> 
> I am trying to make a simple program that ask the user his name then says
> "welcome (name) would you like to do something"
> 
> I tried this...but it doesn't work right...
> 
> name=raw_input ("What is your name?")
> choice1=raw_input "welcome, %s, would you like to do something?" % name
> 
> doing this displays a %s instead of the persons name, can you tell me how to
> fix this?
> 
> thanks

--
Orlando Vazquez


From mal at lemburg.com  Thu Jan 20 13:34:03 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 20 Jan 2000 19:34:03 +0100
Subject: Suggest: .get() method on lists.
References: <20000118212229.A542@stopcontact.palga.uucp>
Message-ID: <3887551B.81288AED@lemburg.com>

Gerrit Holl wrote:
> 
> Hello,
> 
> What about having a get() method on lists?
> I thought it would be useful, for example, for sys.argv. At the moment,
> I have this code:
> 
> file = sys.stdin
> if len(sys.argv) > 1:
>     file = sys.argv[1]
> 
> But this, would be much easier to read:
> file = sys.argv.get(1, sys.stdin)
> 
> Of course, it's also possible to use multiple inheritance on
> UserList and UserDict creating a DictionairyList for sys.argv,
> because it's the only place I can think of such a feature would
> be a Good Thing (TM).

Check out get() in mxTools (look on my Python Pages for more
infos on that package). It handles all indexable objects...

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





From justus at my-deja.com  Fri Jan 21 12:15:12 2000
From: justus at my-deja.com (Justus Pendleton)
Date: Fri, 21 Jan 2000 17:15:12 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com>
Message-ID: <86a46g$9c3$1@nnrp1.deja.com>

> I thought they gotten rid of all such languages until I came across
Python.
> Good programming languages should be whitespace tolerant, and most of
them
> are.

Haveyouevertriedtore adEng lishthatdoesn 'tuse whitespaceproperly?

Most human languages aren't really whitespace tolerant (well, at least
the ones I know :-).  When I was learning Japanese, one of the hardest
things about reading it was the fact that they don't use whitespace to
delineate words.

How come now one is out there advocating that we "fix" English by
removing the mandated whitespace?  :')


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


From news at dorb.com  Sat Jan 15 13:38:30 2000
From: news at dorb.com (Darrell)
Date: Sat, 15 Jan 2000 13:38:30 -0500
Subject: Does python support long longs?
References: <x8_f4.26$tX3.648@vixen.cso.uiuc.edu> <85qa3c$hk$1@nntp9.atl.mindspring.net>
Message-ID: <163g4.3375$hN5.139180@typhoon.nyroc.rr.com>

"Aahz Maruch" wrote:
> >2) Does Python support unsigned 32-bit ints?
>
> Nope.
>
Sort of, check out the stuct or  array modules in "ping and traceroute" for
examples.
http://the-tech.mit.edu/~jeremy/python/

> Note that to correct both of these deficiencies, you could in theory
> build your own Pythonic-C extension if this is "mission critical" for
> you.
>

Writing extensions is not hard at all once you've done one or two.
http://starship.python.net/crew/gmcm/scxx.html

--Darrell




From a.eyre at optichrome.com  Mon Jan 31 13:10:51 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 18:10:51 -0000
Subject: Dynamic class construction?
In-Reply-To: <3895B829.39067306@math.okstate.edu>
Message-ID: <001301bf6c16$81e08ac0$3acbd9c2@peridot.optichrome.com>

>> There's already a module which does this...

> I sorta thought there might be a module that does this, but I
> couldn't find it.

You didn't look very hard :)

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

> Um, I still can't find it. It appears to be a builtin module -
> new.__doc__ says

> '     Functions to create new objects used by the interpreter.
> 
> You need to know a great deal about the interpreter to use this!',
> which is helpful,

I thought so :)

> It's not in the "module index", I don't find any interesting
> files contaning the text "classobj", etc.)

What? This one?

http://www.python.org/doc/current/modindex.html

Ah. Be aware this module was only introduced in 1.5.2 IIRC.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From sjz18 at yahoo.com  Wed Jan 12 09:54:48 2000
From: sjz18 at yahoo.com (Stuart Zakon)
Date: Wed, 12 Jan 2000 06:54:48 -0800 (PST)
Subject: class methods?
Message-ID: <20000112145448.29699.qmail@web804.mail.yahoo.com>

Is there no such thing as class scope methods in Python? (Analogous to static
methods in Java, C++).

I have a class scope dictionary that I am using to collect statistics on method
invocations of the class instances and I wanted to have a class method to print
them out. 

However, Python seems to want a class instance to call a method. Does every
method on a class need self as a parameter?

Certainly I could make a separate statistics class and instantiate it once;
however, the nature of the problem is such that it is more natural for the
statistics collection to be private to the original class.

If I am missing something, I apologize, but coming from Java and C++ it just
seemed natural to have this feature. If there can be class attributes, then why
not class methods? My understanding is that this is natural for Smalltalk as
well.

Stuart Zakon
Objects by Design
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com



From robin at jessikat.demon.co.uk  Tue Jan 25 06:06:22 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 25 Jan 2000 11:06:22 +0000
Subject: using threads with python and winnt
References: <1264075593-4560755@hypernet.com> <38835f4c.1380925@news.iol.ie>
 <388C1CE3.3391A5CF@zebrasoft.co.nz> <86ismn$50a$1@athena.netset.com>
 <388D556E.1DC8C084@zebrasoft.co.nz>
Message-ID: <ONPeOGAuOYj4EwBA@jessikat.demon.co.uk>

I've been testing Zope on Win95 (OSR2) for around 9 months and after the
Zope race etc bugs were fixed I find it works really well. I've tried to
thrash it to death with test demands at high rate and it doesn't fail.

My actual public server is in fact NT based, but that's to do with the
overall reliability of NT rather than any threading problems with win9x.

Certainly I know that Win9x fails badly under high memory loads, but all
the threading tests I carried out worked ok as my main memory/cpu hog
was explicitly designed to be a fifo queue (ie only one request at a
time could hog, other hog requests had to wait).
-- 
Robin Becker


From darrell at dorb.com  Thu Jan 27 05:23:39 2000
From: darrell at dorb.com (Darrell)
Date: Thu, 27 Jan 2000 05:23:39 -0500
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
References: <1392272911@nds-3.azg.nl>
Message-ID: <061001bf68b0$94693980$0100a8c0@rochester.rr.com>

From: "P.J.W.S. Vrijlandt" 
> 
> According to the manual,  hasattr is implemented by calling getattr 
> and seeing if an exception occurs, so hasattr will not be faster than 
> try...except...
> 
Looks like hasattr is still faster.

from time import *

def f1():
    t1=time()
    for x in range(1000000):
        if hasattr(dic, 'xx'):
            pass
    print time()-t1

def f2():
    t1=time()
    for x in range(1000000):
        try:
            dic['xxx']
        except:            
            pass
    print time()-t1

f1()

f2()

6.84399998188
40.1719999313

--Darrell





From ahmed.bouzid at unisys.com  Thu Jan 20 10:42:29 2000
From: ahmed.bouzid at unisys.com (abouzid)
Date: 20 Jan 2000 15:42:29 GMT
Subject: question answering service....
Message-ID: <867ad5$afr$36@trsvr.tr.unisys.com>

Hello all:

Just to let you know that there is a new expert answering service
that covers quite a large range of topic areas and is free. Go to 
		http://www.bluefingers.com 

Ahmed



From scherbi at bam.com  Thu Jan 13 11:44:35 2000
From: scherbi at bam.com (Bill Scherer)
Date: Thu, 13 Jan 2000 11:44:35 -0500
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com> <387DF8D7.95DE0101@bam.com>
Message-ID: <387E00F3.1B12359E@bam.com>

Please forgive more for creating noise by not registering your complete
message, and for creating further noise by asking for forgiveness.  Somehow
my brain missed the "by putting something at the top of the file." part.

Bill

Bill Scherer wrote:

> M-x python-mode
>
> should do it.
>
> M-x load-file
> may be needed first to load python-mode.el
>
> (M == escape key for most, I think)
>
> Bill Scherer
>
> dharms98 at my-deja.com wrote:
>
> > I know there is a way to force emacs to go into python mode even if the
> > file doesn't end with .py by putting something at the top of the file.
> >
> > Can someone remind me of how to do that?
> >
> > Thanks,
> >
> > D.
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> > --
> > http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000113/ebfb78cd/attachment.vcf>

From cjw at connection.com  Sun Jan 23 08:03:17 2000
From: cjw at connection.com (Colin J. Williams)
Date: Sun, 23 Jan 2000 08:03:17 -0500
Subject: olemsg32 module used in win32com exchange server example?
References: <4D0A23B3F74DD111ACCD00805F31D8101D8BCC50@RED-MSG-50>
Message-ID: <388AFC15.A6D8F39C@connection.com>

This posting gives the MS attitude on the MAPI dll.

http://support.microsoft.com/support/kb/articles/Q221/8/38.ASP

Colin W.

Bill Tutt wrote:
> 
> > From: Gordon McMillan [mailto:gmcm at hypernet.com]
> > Sent: Friday, January 21, 2000 8:16 AM
> > To: aahz at netcom.com; python-list at python.org
> > Subject: Re: olemsg32 module used in win32com exchange server example?
> >
> >
> > Aahz wrote:
> > > Grant Edwards <grant at nowhere.> wrote:
> > > >In article <38863DB3.7FAEC820 at callware.com>, Ivan Van
> > Laningham wrote:
> > > >>
> > > >>I suspect that is true, but there's a gotcha.  Unless I'm
> > wrong, both
> > > >>NetScape Mail and Outlook use MAPI dll's, but they expect
> > different
> > > >>functions to be found there.  So, when I want to use
> > Outlook to retrieve
> > > >>mail, I can configure everything else correctly but I
> > have to change the
> > > >>MAPI.DLL to the one the Outlook expects or it barfs.
> > Then put it back
> > > >>when I want to use NetScape. ...
> > > >
> > > ><Boggle>
> > >
> > > It's even worse than that.  AFAIK, NT *still* only loads
> > one copy of a
> > > DLL, so that even if you have DLLs in two different
> > directories, it is
> > > impossible simultaneously to run programs that use a
> > same-named DLL if
> > > they need even different versions.
> >
> > Wrong about that. Though 2 different versions of a mapi dll
> > loaded at the same time might yield trouble, it's not because
> > you can't load them. I do it all the time with (different)
> > python15.dlls.
> >
> 
> Esp. wrong about that with Windows 2000 and Windows 98.
> Win2k and Win98 added features so that you could determine
> which specific DLL (even if its a DLL you only create via CoCreateInstance
> (aka COM object))
> is loaded for any given application.
> 
> See the following URLs for more information:
> http://msdn.microsoft.com/library/techart/sidebyside.htm
> http://msdn.microsoft.com/library/techart/dlldanger1.htm
> 
> Bill




From kc5tja at garnet.armored.net  Fri Jan  7 19:01:31 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 8 Jan 2000 00:01:31 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net>
Message-ID: <slrn87cvv9.gcu.kc5tja@garnet.armored.net>

>>(Personally, I think the COM solution is a superior one anyway, but that's
>>just my opinion.)
>
>Sure it is -- but compatibility is even better.

Then why aren't we all using CP/M?

There is a point of diminishing returns when it comes to this whole
"backward compatibility" thing.  Sometimes you just have to pick up your
bags, and leave Hotel Compatibility behind.  Time marches on, after all.

>Mozilla is right now (M12) almost as stable than Netscape 4.7 or IE 5.0

For the record, three people in this office, who have used Mozilla before,
laughed when I mentioned what you said.  In three words, Not Even Close.
First, from personal experience at least, IE 5 is 100% more stable than any
Netscape version I've ever used.  As far as the current versions of Netscape
are concerned, Mozilla doesn't even hold a candle to them.  Which is pretty
suprising since its built from the same codebase (supposedly) as the much
more stable NT version.  People have been saying, "Mozilla is almost as
stable as..." for the last three years. Without fail, every time I try it,
Mozilla segfaults on something trivial on every Linux box I've used it on.
(I'd be willing to bet that the bugs that destabilize Mozilla under Linux
have to do with XPCOM, and their apparent disregard for existing COM
specifications).

Right now, my bets are on Opera, if and when they ever release it.  And yes,
I'd be willing to pay for it.

>(although it still lacks some important features, and both of those
>browsers are terrible for stability).  The codebase is big and
>complicated, but I like the progress they're making.

As do I -- I'm not commenting on their progress.  I was commenting on its
suitability and stability.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From emile at fenx.com  Thu Jan  6 19:49:36 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 6 Jan 2000 16:49:36 -0800
Subject: how to include a file -- not import??
Message-ID: <04a501bf58a9$14ea9b20$01ffffc0@worldnet.att.net>

----- Original Message ----- 
From: Andrew Csillag <andrew at starmedia.net>
<snip>
> exec open('filename').read() 


That's neat!  I'm assuming that in going from py to pyc the
compiler executes this line causing the additional source to
be read in and compiled as though it were part of the containing
source file.

All you need to do is watch out that you rm/del/recompile all 
container pyc's whenever you modify include files...

now-you-see-me-now-you-don't-ly y'rs

Emile van Sebille
emile at fenx.com
-------------------






From jblazi at netsurf.de  Wed Jan 26 12:45:07 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Wed, 26 Jan 2000 18:45:07 +0100
Subject: Rational numbers
Message-ID: <388f30c7_3@goliath.newsfeeds.com>

What do you think, wouldn't it be nice if Pythony had the built-in data type
"rational" as LISP has? (I would like it. The truth is that the packages
are a bit less confortable.

Janos Blazi




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From MSteed at altiris.com  Wed Jan 19 16:49:22 2000
From: MSteed at altiris.com (Mike Steed)
Date: Wed, 19 Jan 2000 14:49:22 -0700
Subject: Python Rocks!
Message-ID: <65118AEEFF5AD3118E8300508B124877073D25@webmail.altiris.com>

> From: Garrett G. Hodgson [mailto:garry at sage.att.com]
> Sent: Wednesday, January 19, 2000 2:36 PM
> To: python-list at python.org
> Subject: Re: Python Rocks!
> 
> 
> Tres Seaver wrote:
> 
> > Tabs in source files are devil's spawn, plain and simple. :)
> 
> no, spaces are.

I\040agree,\040we\040shouldn't\040use\040spaces.

--
M.



From bsb at winnegan.de  Mon Jan 17 16:12:20 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 17 Jan 2000 22:12:20 +0100
Subject: Removing tags from a html-file
In-Reply-To: Thomas Weholt's message of "Mon, 17 Jan 2000 05:52:20 +0100"
References: <3882A004.A5F150F@bibsyst.no>
Message-ID: <87k8l89zd7.fsf@baal.winnegan.de>

Thomas Weholt <thomas at bibsyst.no> writes:

> Hi,
> 
> I want to remove all tags from a html-file, so that only the plain-text
> remains. How, if so, can this be done using htmllib.py?

The following should do the trick

from htmllib import HTMLParser
from formatter import AbstractFormatter, DumbWriter

parser = HTMLParser(AbstractFormatter(DumbWriter()))
parser.feed(YOUR_DATA)

Instatiated w/o arguments, DumbWriter writes to stdout.


 - Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
                  bsb at north.de - http://www.north.de/~bsb/
****** ceterum censeo javascriptum esse restrictam *******



From R.Brodie at rl.ac.uk  Mon Jan 24 04:56:03 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Mon, 24 Jan 2000 09:56:03 -0000
Subject: Whitespace delimiters suck - Seen Occam?
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <86cmom$5pp$1@news-int.gatech.edu>
Message-ID: <86h7jm$o7i@newton.cc.rl.ac.uk>

"Andrew Henshaw" <andrew.henshaw at gtri.gatech.edu> wrote in message
news:86cmom$5pp$1 at news-int.gatech.edu...

> By the way, from my long experience with Occam - I can enthusiastically
> recommend using a folding editor with indentation languages like Python.
> They're made for each other.

Curious. Everyone I knew who programmed in Occam had a profound
dislike for the folding editor - it put me off the idea of using whitespace
as syntax for years. I've got over it now though ;)






From mgushee at havenrock.com  Sun Jan 16 11:29:46 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 16 Jan 2000 11:29:46 -0500
Subject: checking for file existance
References: <Pine.LNX.4.21.0001121324440.10184-100000@fep132.fep.ru>
Message-ID: <m31z7iughx.fsf@havenrock.com>

Oleg Broytmann <phd at phd.russ.ru> writes:

> On Wed, 12 Jan 2000, Anders M Eriksson wrote:
> > How do I check if a file exist on a webserver? I have tried
> > urllib.urlopen() but it never give any exception IOError.
> 
>    There is no such thing as "file-by-WWW". URL gives you a reference to an
> object, and if a WWW server can't resolve the reference, it is not
> mandatory to return the error 404 - WWW server may decide to return another
> object (a page with words and picture saying you have got an error).

If you use httplib instead of urllib, you can access the HTTP status
codes. However, as Oleg points out, you can't always depend on those
codes, so it's questionable whether it's worthwhile.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From nic at csee.uq.edu.au  Sun Jan 30 22:22:25 2000
From: nic at csee.uq.edu.au (Nic Williams)
Date: Mon, 31 Jan 2000 03:22:25 GMT
Subject: Creating a local variable with a parameterized name
References: <38937075.18B75753@rochester.rr.com> <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu> <8704vo$soj$1@nnrp1.deja.com> <vndaelomfpu.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <872v5f$o2q$1@nnrp1.deja.com>

In article <vndaelomfpu.fsf at betelgeuse.ccs.neu.edu>,
  Justin Sheehy <dworkin at ccs.neu.edu> wrote:
> Nic Williams <nic at csee.uq.edu.au> writes:
>
[more cutting]
> > def f(name):
> > 	locals()[name]=123
> > 	print locals()[name]
>
> This is why I didn't recommend that approach, from the Python
documentation:
>
> locals ()
>        Return a dictionary representing the current local symbol
table.
>        Warning: the contents of this dictionary should not be
modified;
>        changes may not affect the values of local variables used by
>        the interpreter.
>
> Yes, I know that it works right now, but it is also quite clearly
> marked as something which could stop working without warning in a
> future version of Python.
>
> If you must recommend something like that, it is only polite to point
> out that it is strictly incorrect code and should not be relied upon.

It is clearly marked having read it carefully, and I will now politely
point out that it is incorrect code :-) I do believe that globals() is
ok though to use and modify, though it does clearly appear that vars()
and locals() should not be modified.

Sorry for missing that after merely running some test code to test my
theory.

nic


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


From News at Titanic.co.uk  Tue Jan  4 09:11:48 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Tue, 04 Jan 2000 14:11:48 GMT
Subject: String Handling
Message-ID: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com>

Hi All,

Can anyone give me the Python line/command to change all spaces in a
string to +s?

ie

"Lucy by Railway Carraige"

becomes

"Lucy+by+Railway+Carraige"

Muchos gratcias etc

Simon
Simon Faulkner


From tim_one at email.msn.com  Fri Jan 14 16:36:32 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 14 Jan 2000 16:36:32 -0500
Subject: class.__getattr__ behavior:  feature or bug? 
In-Reply-To: <200001142112.PAA27786@ws89.krumm0.com>
Message-ID: <000801bf5ed7$6cd5e540$a0a2143f@tim>

[Huayin Wang]
> >>> class S:
> ...    def __getattr__(self, name):
> ...       if name == 'c':
> ...          self.a = self.a + 1
> ...          return self.a
> ...       else: return None
> ...
> >>> s=S()
> >>> s
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: call of non-function (type None)
> >>>

Feature.  Try sticking "print name" in __getattr__ to see what's happening.
Putting "s" on a line by itself implicitly asks Python to invoke repr(s) (to
build an output string), so Python tries to look up the special method
s.__repr__.  Your __getattr__ returns None, and Python complains that None
isn't a sensible __repr__ method.

To fix that, don't do this <wink>, define S.__repr__, or capture the name
"__repr__" in your __getattr__ and return something sensible for it.

Note that __getattr__/__setattr__ are "advanced" features -- meaning mostly
that they're low-level hooks bristling with subtleties.

your-next-post-will-be-wondering-whether-an-infinite-
    __getattr__-loop-is-a-bug<0.5-wink>-ly y'rs  - tim





From fredrik at pythonware.com  Wed Jan 19 15:52:03 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 21:52:03 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000119233305.25615A-100000@sunse>
Message-ID: <006901bf62bf$5faf1ad0$f29b12c2@secret.pythonware.com>

Evguenii Smogailov <sei at sunse.jinr.ru> wrote:
> New operators:
> 1) 'eq' (like '==');
> 2) 'ne' (like '!=','<>');
> 3) 'last' (like 'continue');
> 4) '..' for generating sequences (like '1..20' or 'a..z'),
> like in Perl.
> 
> Generally for improving readability,
> (and Perl compatibility).

oh, come on.  since when is "Perl compatibility" a
guiding principle for Python's design?

what's wrong with you folks?

</F>




From rhaper at houston.rr.com  Sun Jan 23 13:22:25 2000
From: rhaper at houston.rr.com (Rod Haper)
Date: Sun, 23 Jan 2000 12:22:25 -0600
Subject: Looking for Solaris 2.6/Python 1.5.2 make test fix
References: <86ajtj$8f8$1@srv38.cas.org> <3888DDFC.38F7BBC@sage.nelesautomation.com> <20000121235034.D10685@xs4all.nl>
Message-ID: <388B46E1.53F1044C@houston.rr.com>

It turned out that I had mismatched the libc.so version reference in
/usr/lib/libc.so as the result of a quicky update for an application
that need a newer version.  I've been using the CVS tree for updates
since 1.5.2 was released and was surprised when my last update didn't
run.  I had forgotten about the libc.so mod but tracked it down after
you pointed out the obvious.  Thanks.


Thomas Wouters wrote:
> 
> On Fri, Jan 21, 2000 at 04:30:20PM -0600, Rod Haper wrote:
> 
> > I'm also getting a segmentation fault on Red Hat Linux 6.1 at the same
> > point in 'make test' but no core file is generated:
> 
> > rm -f ./Lib/test/*.py[co]
> > PYTHONPATH= ./python ./Lib/test/regrtest.py
> > make: *** [test] Segmentation fault
> > PYTHONPATH= ./python ./Lib/test/regrtest.py
> > make: *** [test] Segmentation fault
> 
> Something probably went wrong with building your Python binary. Can you
> start the python executable ? do you get the prompt ? Can you execute code ?
> Python should run fine on rh6.1, unless you have a buggy compiler or a buggy
> computer, or you built a buggy module statically. How were you building it ?
> Which version of python was it, how did you run 'configure' and what did you
> change to Modules/Setup ? (You *do* need to change a few things in there.)



From loewis at informatik.hu-berlin.de  Fri Jan 28 14:45:03 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Jan 2000 20:45:03 +0100
Subject: Rational numbers
References: <388f30c7_3@goliath.newsfeeds.com>
Message-ID: <p6qoga6j800.fsf@informatik.hu-berlin.de>

"Janos Blazi" <jblazi at netsurf.de> writes:

> What do you think, wouldn't it be nice if Pythony had the built-in data type
> "rational" as LISP has? (I would like it. The truth is that the packages
> are a bit less confortable.

What's wrong with

>>> from yarn import Rat
>>> Rat(4,5) * Rat(7,12)
Rat(7, 15)

I can't imagine that you can get this much more comfortable, without
breaking existing programs. I guess you'd like 4/5 return a rational
number, even though it currently gives you 0. I would not like that.

Regards,
Martin



From skip at mojam.com  Fri Jan 28 13:09:51 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 28 Jan 2000 12:09:51 -0600 (CST)
Subject: tabs do WHAT?
In-Reply-To: <86sfmd$ktl$1@mach.vub.ac.be>
References: <000301bf6706$99fca7c0$632d153f@tim>
	<86joac$l2s$1@mach.vub.ac.be>
	<86k30t$i5r$1@nntp2.atl.mindspring.net>
	<landauer-4BDC4F.15462726012000@news.apple.com>
	<86pcii$600$1@mach.vub.ac.be>
	<38918A29.8AB61742@electricorb.com>
	<86sfmd$ktl$1@mach.vub.ac.be>
Message-ID: <14481.56175.546895.236262@beluga.mojam.com>

    Thomas> If braces are used, you can choose between a myriad of tools to
    Thomas> get consistent indentation (there are lots of C and C++
    Thomas> beautifiers out there). 

You have the same problem here as getting people to indent code in a sane
manner.  Getting people to use indent or similar tools on a consistent basis
is like herding cats.

Regarding screwing up the indentation, many people have pointed out that
editors such as vim and emacs support indentation fairly well.  In practice,
if your editor supports controlled indentation, it just isn't a problem.  On
those extremely rare occasions in the past six years or so of Python
programming, when I have had problems, emacs' ediff-revision command has let
me compare (in a nicely highlighted fashion) the current buffer with the
last version I checked back into CVS.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098
"Languages that change by catering to the tastes of non-users tend
not to do so well." - Doug Landauer



From s.schwarzer at ndh.net  Mon Jan 24 20:09:24 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Tue, 25 Jan 2000 02:09:24 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <388670D2.2FA842F6@starmedia.net> <Pine.SOL.3.91.1000120001532.29235B-100000@sunse> <tIeGOHk6cMDxz41yCwn96Iy45Dgx@4ax.com>
Message-ID: <388CF7C4.D6247285@ndh.net>

Hello Dennis,

Dennis Lee Bieber schrieb:
>         The only languages I've ever seen that use a text string for
> those operators are FORTRAN and VMS DCL.

The only programming languages I (personally) know to pay attention
to amount of whitespace are Fortran, Assembler and Python ;-) .

(BTW, this confused me to learn Perl instead of Python in the first
place, because that property of Python reminded me too much of
Fortran.)

Stefan


From wjk at wjk.mv.com  Wed Jan 26 17:34:13 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Wed, 26 Jan 2000 17:34:13 -0500
Subject: Newbie question on paging..
References: <388DF473.1ED172A4@wjk.mv.com> <slrn88s0il.e0a.grant@grante.comtrol.com>
Message-ID: <388F7664.9C79C1DA@wjk.mv.com>

Thankyou for your response and all who responded.
Learning something everyday!




From R.Brodie at rl.ac.uk  Tue Jan 25 05:08:50 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Tue, 25 Jan 2000 10:08:50 -0000
Subject: Multidimensional Dictionaries: Possible?
References: <slrn88ppsg.49t.w6uv@voyager.home.com>
Message-ID: <86jsnk$s2q@newton.cc.rl.ac.uk>

"Jerry Gardner" <w6uv at hotmail.com> wrote in message news:slrn88ppsg.49t.w6uv at voyager.home.com...
>
> Are multidimensional dictionaries possible in Python? I'm sure they're
> possible, but I can't figure out the syntax.

Indeed so. When you get the hang of it is actually easier to follow
than Perl's reference syntax.

> Here's how I do it in Perl:
<snipped - because it brings back painful previous Perl programs>

>>> dict={}
>>> dict['key1'] = {}
>>> dict['key1']['key2'] = 'thing'
>>> dict['key1']['key2']
'thing'
>>> dict
{'key1': {'key2': 'thing'}}

Hence in one:
>>> dict2 = {'key1': {'key2': 'thing'}}





From edcjones at erols.com  Tue Jan 25 23:51:39 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Tue, 25 Jan 2000 23:51:39 -0500
Subject: Interesting article on programming languages
Message-ID: <388E7D5B.7840983F@erols.com>

I saw the following on slashdot:

GameSpy posted a six-page essay of Tim Sweeney's entitled "A
Critical Look at Programming Languages." Tim Sweeney is, of
course, the lead coder for the Unreal game engine (one of the
most licensed 3D game engines for the PC.)

The article is at http://www.gamespy.com/articles/devweek_b.shtm





From gapalma at quiknet.com  Thu Jan 20 22:15:36 2000
From: gapalma at quiknet.com (George Palma)
Date: Thu, 20 Jan 2000 19:15:36 -0800
Subject: Configure EMACS for Python
Message-ID: <3887CF58.BEF947CA@quiknet.com>

I have recently begun to use Linux (moving fromm OS/2) and want to continue to use

Python as a scripting language. Can anyone give me advice on how I can configure EMACS

to format Pyhton code? I reasonably sure that someone has figured this out and I do

not want to reinvent the wheel.

--
George Palma
Loomis, CA
Running on Linux, OS/2, and DOS


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000120/1a26844d/attachment.html>

From guido at CNRI.Reston.VA.US  Mon Jan  3 12:37:59 2000
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Mon, 03 Jan 2000 12:37:59 -0500
Subject: Python Conference -- Early Bird Registration ends Jan 5!
Message-ID: <200001031737.MAA24342@eric.cnri.reston.va.us>

This is the last warning.  The conference is getting booked full,
don't wait till the last moment!

If you haven't registered and paid by January 5, you will paying full
price...  So, be smart and register NOW.  Also don't forget to book
your hotel room by January 3 to qualify for the conference rate!

Some highlights from the conference program:

- 8 tutorials on topics ranging from JPython to Fnorb;
- a keynote by Open Source evangelist Eric Raymond;
- another by Randy Pausch, father of the Alice Virtual Reality project;
- a separate track for Zope developers and users;
- live demonstrations of important Python applications;
- refereed papers, and short talks on current topics;
- a developers' day where the feature set of Python 2.0 is worked out.

Our motto, due to Bruce Eckel, is: "Life's better without braces."

Come and join us at the Key Bridge Marriott in Rosslyn (across the
bridge from Georgetown), January 24-27 in 2000.  Make the Python
conference the first conference you attend in the new millennium!

The early bird registration deadline is January 5.  More info:

    http://www.python.org/workshops/2000-01/

The program is now complete with the titles of all presentations.
There is still space in the demo session and in the short talks
session.

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



From pinard at iro.umontreal.ca  Mon Jan 17 17:07:04 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Jan 2000 17:07:04 -0500
Subject: Incoming PO utilities 0.4 announcement
Message-ID: <oqwvp82vzr.fsf@titan.progiciels-bpi.ca>

Hi, people.

My few PO utilities (formerly `xpot' and PO mode) are slowly taking form
as a distribution.  I'll send a separate announcement, using the means
I already have for other packages.  However, compared to them, I do not
think PO utilities are worth their own announce/pretest mailing lists.

It makes some sense announcing it to team leaders, and maybe once more to
the Python list (because a few Python-related problems have been repaired).
In any case, I hope the distribution point should not change much for some
while, so there is no need to repeat the announcement just for a change of
version numbers.  I'm not sure where the equilibrium lies between avoiding
pollution and sharing news with those interested.  Tell me what you think.

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





From aahz at netcom.com  Fri Jan 28 10:56:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 28 Jan 2000 15:56:09 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <slrn88n40a.2qh6.cjc26@bridget.mindriot.net> <948708715.22839@news.adtranzsig.de> <slrn891l2f.5tk.cjc26@bridget.mindriot.net>
Message-ID: <86se6p$qu9$1@nntp5.atl.mindspring.net>

In article <slrn891l2f.5tk.cjc26 at bridget.mindriot.net>,
Cliff Crawford <cjc26 at nospam.cornell.edu> wrote:
>
>	(kicked(the ball))(John)
>
>In that last one, "kicked" represents a curried (or
>Schoenfinkelized :) function; it is a function which takes a noun
>phrase X as its argument and which returns another function which
>takes a noun phrase Y as its argument and returns true if Y
>kicked X.  I.e.:
>
>	kicked = lambda(X) (lambda(Y) (if Y kicked X: 1
>	                               else: 0))

My head hurts.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From warlock at eskimo.com  Wed Jan  5 20:39:27 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Wed, 05 Jan 2000 17:39:27 -0800 (PST)
Subject: Folding (was Re: Python not a Very High-Level Language?)
In-Reply-To: <20000105172721.A1624@Ridcully.home>
Message-ID: <XFMail.20000105173927.warlock@eskimo.com>

On 05-Jan-2000 Malcolm Tredinnick wrote:
> On Tue, Jan 04, 2000 at 09:18:35AM -0800, Jim Richardson wrote:
>> On 2 Jan 2000 13:31:09 GMT, 
>>  lin li, in the persona of <linli at computer.org>,
>>  brought forth the following words...:
>> 
>> >Message by boud at rempt.xs4all.nl (Boudewijn Rempt) on: 1/1/00 10:40:48 PM
>> >>I've tried a lot of folding editors over the years, and I've never
>> >>found one that was as pleasant to use as, say, Word's outline editor.
>> >>Perhaps that's just because I was expecting that, instead of, for
>> >>instance, Origami. What I'd like would be an editor that didn't need
>> >>to insert all kind of cruft into the source files, but could work
>> >>on a set of preferences and rules to present the source in a nice way.
>> >
>> >Just an idea, how about folding on the indentation levels? We just have
>> >to forget about all these other languages then.
>> >
>> >Lin
>> Works for me:) anyone up to chewing on the vim sources for this :)
> 
> Before we (I am interested in this, too) go leaping off into modifying vim
> sources, be aware that folding is on the agenda for Vim 6. So if you are
> serious about modifying sources, etc, it would be worth getting in contact
> with the developers to find out what they have in mind, so at least we can be
> consistent. It would be a royal pain to have one sort of folding for Python
> and another type of mindset for everything else that is foldable.
> 
> Cheers,
> Malcolm
> 
> --
> Despite the high cost of living, have you noticed how popular it remains?

I am not about to try hacking the vim code, I don't know C well enough to do
any good there, All I can do is add an echo of "me too" to the desire for a
good folding mode in vim for python (and anything else, but principally python)

Jim Richardson
        Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
        Linux, because life's too short for a buggy OS.




From andrew.henshaw at gtri.gatech.edu  Sat Jan 22 11:42:41 2000
From: andrew.henshaw at gtri.gatech.edu (Andrew Henshaw)
Date: Sat, 22 Jan 2000 11:42:41 -0500
Subject: Whitespace delimiters suck - Seen Occam?
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu>
Message-ID: <86cmom$5pp$1@news-int.gatech.edu>

Roy Smith  wrote:
...
>I'm in love with python, but this idea of white space as significant 
>syntax is the biggest wart on the language by far.

My all-time favorite language is Occam ("as simple as necessary"), which
was popular for parallel processing on the Transputer.  It is a beautiful 
language that uses indentation for scoping and block constructs.  Unlike
Python, however, the indentation was more rigidly defined (everything
about Occam is rigidly defined).  Each indent is two spaces.  No tabs, no
odd numbers of spaces, no ambiguity.  I'm a newbie to Python, and
I was very pleased to find that it used indentation for blocking, but I have
been a bit concerned about the loose aspects.  It has already bitten me
(mildly) on some examples that I've copied and pasted from newsgroups.  But if
it's the worst problem I experience with Python, then I'm in great shape!

By the way, from my long experience with Occam - I can enthusiastically
recommend using a folding editor with indentation languages like Python.  
They're made for each other.

-- 
Andrew Henshaw


From jblaine at shell2.shore.net  Tue Jan  4 13:16:51 2000
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Tue, 04 Jan 2000 18:16:51 GMT
Subject: PIL 1.0 and palettes with ImageDraw
Message-ID: <slrn874e8j.f3v.jblaine@shell2.shore.net>

Hi folks,

I've gotten up to speed so far to do everything I want (make pie charts)
with PIL.  Yay!

However, my charts are very drab from using the default palette for mode
"P" (I am writing GIFs...please hold off on the PNG spiel).

I have read through the PIL 1.0 handbook and cannot figure out how to
change my palette properly.

If I do:

  im = Image.new("P", (ImageWidth, ImageHeight))
  im.palette = ImagePalette.ImagePalette("RGB")

I get:

    File "/jblaine/mylocal/lib/python15/Image.py", line 375, in load
      if self.im and self.palette and self.palette.rawmode:
  AttributeError: rawmode

and aside from that, I can't figure out how to just get a "simple"
"rainbow" palette (wow, I'm pretty imaging-lingo savvy, eh?).

Maybe I can kill 2 birds with one stone here and also mention that I
want "Black" to be transparent in my output images."

Anyone have any pointers or code I can look at?  PIL comes with no
demos that perform ImageDraw.* things that I can find.

Thanks for your time


From bradmarshall at fastpointcom.com  Thu Jan 27 15:38:30 2000
From: bradmarshall at fastpointcom.com (brad marshall)
Date: Thu, 27 Jan 2000 12:38:30 -0800
Subject: Cookies
References: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com> <14461.65478.595789.251070@ghost.bbn.com>
Message-ID: <3890ACC6.24815613@fastpointcom.com>

OK,

I don't think my message went through the first time.  SOrry if this is a
repeat.

Tim, I can't find your Cookie module anywhere to download.  COuld you
point me in the right direction?

Brad



From robin at jessikat.demon.co.uk  Tue Jan 11 07:03:41 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 11 Jan 2000 12:03:41 +0000
Subject: Python plugin
References: <kizIvIAD2me4EwPp@jessikat.demon.co.uk>
 <000e01bf5bea$512f6a40$d2a0143f@tim>
Message-ID: <gV05SBAdwxe4Ew8N@jessikat.demon.co.uk>

In article <000e01bf5bea$512f6a40$d2a0143f at tim>, Tim Peters
<tim_one at email.msn.com> writes
>[Samuel A. Falvo & Robin Becker bump heads over the True
> Pukeworthiness or Lack Thereof of COM; then, in good
> Python fashion, make up before blood is spilt]
>
>Admirably civilized, fellows!  Tip o' the hat.
>
>[Robin]
>> OK I guess we all need a bit of adrenalin now and then heh heh
>> heh aaaaaarrrggghhhhh the Timbot is presumably busy.
>
>Don't count on the timbot, Robin!  I've noticed with increasing concern an
>increasing (albeit non-monotonic) tendency for it to act as bland as toast
>this year.  If this keeps up, I'm going to have to slash its electricity.
>
>it's-easy-to-be-nice-when-you're-not-starving-ly y'rs  - tim
>
>
>
No one expects the random inquisition. The Timbot must have scanning
devices better than the NSA.
-Dear NT administrator, my palette has no Norwegian Blue-ly yrs- 
Robin Becker


From gerrit.holl at pobox.com  Mon Jan 17 14:44:16 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 17 Jan 2000 20:44:16 +0100
Subject: Version history of python
In-Reply-To: <20000117165510.A3786@stopcontact.palga.uucp>; from gerrit.holl@pobox.com on Mon, Jan 17, 2000 at 04:55:10PM +0100
References: <85v0ih$409$1@nnrp1.deja.com> <20000117165029.A3761@stopcontact.palga.uucp> <20000117165510.A3786@stopcontact.palga.uucp>
Message-ID: <20000117204416.A1291@stopcontact.palga.uucp>

Gerrit Holl wrote on 948124510:
> Hi again,
> 
> gerrit wrote on 948124229:
> > File attached.
> 
> *really* attached this time.

My apologies: I didn't know the file was *that* big.

regards,
Gerrit.



From someone at deneb.cygnus.argh.org  Sat Jan 22 12:24:08 2000
From: someone at deneb.cygnus.argh.org (Florian Weimer)
Date: 22 Jan 2000 18:24:08 +0100
Subject: [Tkinter/Win32] Message box on app start -> mouse is dead
References: <87g0vv6wlc.fsf@deneb.cygnus.argh.org> <3884E2E8.F3F6A83E@endea.demon.nl>
Message-ID: <87zotyc953.fsf@deneb.cygnus.argh.org>

Niels Diepeveen <niels at endea.demon.nl> writes:

> Displaying a message box outside the main loop seems to mess up tk's
> message dispatching somehow. One way to work around this is to use a
> separate Tk instance and throw it away after displaying the message:

> tk = Tk()
> Button(tk, text="Exit", command=sys.exit).pack()
> xxx = Tk()
> xxx.withdraw()
> tkMessageBox.showinfo("Test app", "Click on Ok, then on Exit.\n"
>                       + "Do you still see the problem?", master=xxx)
> xxx.destroy()
> tk.mainloop()

Another possibilty (which I actually use): Calling the initialization
code from the main loop (using Tk.after_idle).


From thomas at xs4all.net  Mon Jan 31 10:04:02 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 31 Jan 2000 16:04:02 +0100
Subject: Case Sensitivity and Learnability
In-Reply-To: <slrn898vr0.ffp.neelk@brick.cswv.com>; from neelk@brick.cswv.com on Sun, Jan 30, 2000 at 06:13:45PM +0000
References: <000501bf6a9f$8f501e00$78a0143f@tim> <slrn898vr0.ffp.neelk@brick.cswv.com>
Message-ID: <20000131160402.D19725@xs4all.nl>

On Sun, Jan 30, 2000 at 06:13:45PM +0000, Neel Krishnaswami wrote:

> Well, I'd prefer Python to be case-insensitive, just as a matter of my
> own preference. Eyeball grep is for me basically case-insensitive, and
> I have spent hours trying to find bugs caused by case errors, because
> my eyes just slide over the difference.

Funny, my eyeball grep is decidedly case-sensitive. N and n look very
differently, and I scan them very differently, too. I sooner have trouble
with 'I', 'l', '1' and in some fonts 'i', or '0' and 'O' and in some fonts
'Q'. On my old Atari ST I used to have a perfect screen font (which came
with the Warp 9 screen accelerator) but unfortunately I have not been able
to find an adequate substitute on any of the UNIX or Windows systems i've
been working on, and I've been looking for over 5 years now ;)

I personally think python should not be case insensitive (not that I fear
that, though.) -- I like to think 'when names _look_ different, they _are_
different.'. It's just when they look alike that I have to doublecheck for
'1's and 'l's and such ;)

Courier-bold-1y y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From dharms98 at my-deja.com  Thu Jan 13 10:34:04 2000
From: dharms98 at my-deja.com (dharms98 at my-deja.com)
Date: Thu, 13 Jan 2000 15:34:04 GMT
Subject: Emacs question
Message-ID: <85kr9b$2kh$1@nnrp1.deja.com>


I know there is a way to force emacs to go into python mode even if the
file doesn't end with .py by putting something at the top of the file.

Can someone remind me of how to do that?

Thanks,

D.


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


From claird at starbase.neosoft.com  Mon Jan 17 10:26:30 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 17 Jan 2000 09:26:30 -0600
Subject: LANL software contest
References: <of11z7jd8lo.fsf@kirsch.dialup.fhg.de> <3880CEC0.C1A660F7@scriptics.com> <of1g0vytkyy.fsf@kirsch.dialup.fhg.de> <38832BE8.81EFC77D@mayo.edu>
Message-ID: <E944E07C655AF327.9A90D78300B69607.3C74E3DED0FC804C@lp.airnews.net>

In article <38832BE8.81EFC77D at mayo.edu>,
Bob Techentin  <techentin.robert at mayo.edu> wrote:
>Harald Kirsch wrote:
>
>[snip]
>
>> But as mentioned in another post, I just don't like it that they don't
>> say: ``we just like Python more than Tcl'' but ``Tcl is easier to
>> learn and read than ..., but not as easy as Python.''
>> 
>> I doubt they have hard facts to prove that, but if the contest will be
>> a success, everybody will later cite their statement as if it were a
>> hard fact. And a new myth will be born like `helmets reduce the danger
>> of bicycle riding' (read rec.bicycle.rides if you think this is a
>> proven fact).
>
>I'm not so sure.  When I read Guido's proposal to DARPA for developing a
>teaching environment for Python
>(http://www.python.org/doc/essays/cp4e.html), I saw the first ever (for
>me) reference to studying the "learnability" of a language.
>
>    We already have some evidence of where changes might be necessary. 
>    Prof. Randy Pausch at Carnegie Mellon University (see below) has
>    conducted some usability studies of Python within their limited 
>    problem domain. Their users seemed most confused by the case 
>    sensitivity of Python's variable names and by the truncation of 
>    integer division. 
>
>I would argue (without any actual physical evidence) that you could
>teach Tcl even more easily to beginners.  Tcl's dead-simple syntax
>should be easy to teach to non-programmers, but has anybody actually
>studied how hard/easy it is to learn?
			.
			.
			.
Briefly, no.

Imagine:  programmers go around arguing about various
qualities pertinent in software engineering (reliability,
efficiency, ...), and very, very few of the claims extant
in our domain ("OO is more scalable"; "HLL coding is
faster"; ...) have been tested (let alone re-tested) with
any pretense of rigor.  There definitely is evolution in
behavior (look at magazine titles--structured programming,
objects everywhere, XML empire, ...), but it's best
modeled as fashion change.

Learnability is an even more extreme case of this.  Guido
worked in the '80s with one of the few groups ever to do
the hard (but illuminating) empirical work of learnability
experiments.

I know of nothing comparable done with Tcl.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From mhammond at skippinet.com.au  Wed Jan  5 22:05:24 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 06 Jan 2000 03:05:24 GMT
Subject: Python and ASP
References: <_ji34.8719$523.343739@news.chello.at> <bBk34.1288$Cl3.6689@news-server.bigpond.net.au> <slrn84t6n0.q3j.aa126@fan1.fan.nb.ca> <BZA34.2308$Cl3.11615@news-server.bigpond.net.au> <387351CA.1EA77B30@gssec.bt.co.uk>
Message-ID: <UDTc4.11594$lv5.51278@news-server.bigpond.net.au>

Microsoft have defined an API called "Active Scripting", that allows
applications to host scripting languages, without actually knowing the
languages they are hosting. Microsoft provide a number of Active
Scripting applications - ASP, IE4 and 5, WSH and one or 2 others. They
also provide 2 active scripting languages - VBScript and JScript.
Naturally, the MS languages are most popular with MS applications.

Python supports active scripting as soon as you install
win32all-???.exe. Then, any Active Scripting application also supports
Python.

However, each application defines an object model - ie, ASP, WSH and
IE all have different object models. Thus, VBScript code written for
ASP will not work under WSH or IE, etc. In the same way, Python code
written for ASP will not run under IE, WSH or even standard
Python.exe.

Hope this makes sense - look for the samples when you install win32all

Mark.



"Alan Gauld" <alan.gauld at gssec.bt.co.uk> wrote in message
news:387351CA.1EA77B30 at gssec.bt.co.uk...
> Mark Hammond wrote:
> > No - you can do Python in ASP.
>
> Care to elaborate?
>
> > Not really.  Python programs running under ASP make extensive use
of the ASP
> > object model, and hence won't work outside an ASP environment.
>
> Again, how is that done? Any Web pages you can reference?
> I don't see any ASP references on python.org
>
> > Just like
> > VBScript or JScript code designed for ASP wont work outside of
ASP.
>
> On a (tangentially) related theme has anyone written whatever
> plumbing is needed to make python work under
> Windows Scripting Host? MS do VBscript and JScript and
> mention a Perl version available for download.
> Is there a Python one?
> I know the pythonwin extensions render it semi redundant
> but it would be nice to keep up with the Jones :-)
>
> > including those sent to mailing lists).  4 spams per week, and I
have been
> > using a unprotected email address on Usenet for years and years.
What
> > exactly are you afraid of?
>
> Maybe the 7 or 8 spans I get per _day_... and thats after
> filtering which catches another 3 or 4!
>
> But OTOH, I still don't protect my mail coz I find them such
> a pain. I just don't reply to protected addresses!
>
> Alan G.
> --
> =================================================
> This post represents the views of the author
> and does not necessarily accurately represent
> the views of BT.




From dworkin at ccs.neu.edu  Tue Jan 18 14:58:28 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 18 Jan 2000 14:58:28 -0500
Subject: Python Rocks!
In-Reply-To: Nick Belshaw's message of "Tue, 18 Jan 2000 13:12:52 +0000"
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk>
Message-ID: <vndn1q340ez.fsf@betelgeuse.ccs.neu.edu>

Nick Belshaw <nickb at earth.ox.ac.uk> writes:

> > Get used to it, and don't use tabs.

> ! - whats wrong with tabs?

They don't look the same in every editor, or with each individual's
tab-width settings.

This isn't a python-specific thing.  I recommend against using tabs
for indentation in almost any source code, regardless of the language.

-Justin

 



From alex at mindlube.com  Mon Jan  3 02:46:33 2000
From: alex at mindlube.com (Alex Rice)
Date: Mon, 03 Jan 2000 00:46:33 -0700
Subject: cannot install wxPython on linux
Message-ID: <387053D9.4EB9C9B8@mindlube.com>

Sure could use some help here. 

I have tried building my own libs & succeeded through glib and gtk+ 1.2.6, but
errors out on wxGTK 2.0.1.

I have now installed rpms, but wxPython has undefined symbols :-(

Does anyone have a statically linked distro of wxPython?

$ rpm -qa | grep "gtk\|wx"
gtk+-1.2.3-2
gtk+-devel-1.2.3-2
wxGTK-2.0.1-1
wxPython-2.1.11-1

$ ldconfig -v | grep "gtk\|wx"
        libwxPyHelpers.so => libwxPyHelpers.so
        libwx_gtk2.so.0 => libwx_gtk2.so.0.1
        libgtk-1.2.so.0 => libgtk-1.2.so.0.2.1
        libobgtk.so.1 => libobgtk.so.1.2.1
        libgtkxmhtml.so.1 => libgtkxmhtml.so.1.0.1
        libgnorbagtk.so.0 => libgnorbagtk.so.0.0.0
        libwxPyHelpers.so => libwxPyHelpers.so
        libwx_gtk2.so.0 => libwx_gtk2.so.0.1
        libgtk-1.2.so.0 => libgtk-1.2.so.0.2.1
        libobgtk.so.1 => libobgtk.so.1.2.1
        libgtkxmhtml.so.1 => libgtkxmhtml.so.1.0.1
        libgnorbagtk.so.0 => libgnorbagtk.so.0.0.0

Python 1.5.2 (#1, Sep 14 1999, 22:35:43)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from wxPython.wx import *
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/site-packages/wxPython/__init__.py", line 20, in ?
    import wxc
ImportError: /usr/lib/libwxPyHelpers.so: undefined symbol:
CopyObject__C8wxObjectR8wxObject
>>> 

Running Redhat 6.0/i386.

-- 
Alex Rice <alex at mindlube.com>


From samschul at pacbell.net  Sat Jan 15 20:34:00 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Sun, 16 Jan 2000 01:34:00 GMT
Subject: How to use Pythonwin's ability to access windows dll's???
References: <1264144125-438654@hypernet.com>
Message-ID: <85r768$isb$1@nnrp1.deja.com>

In article <1264144125-438654 at hypernet.com>,
  gmcm at hypernet.com wrote:
> Sam Schulenburg writes:
>
> > I can access windows dll's using Sam Rushing's calldll.pyd and
windll.py. My
> > question is are thes features built into Pythonwin?
>
> Nope. Everything in the Win32 extensions is wrapped (much
> of it via SWIG).  Sam's stuff is regarded as dangerous,
> because it's too easy to crash Python.
>
> Wimps.
>
> real-men-juggle-chain-saws-ly y'rs
>
> - Gordon
>
>

Does that mean that real programmers use Sam's routines, and when
Python crashes they hack at it until it works :)
Seriously what is the purpose of the LoadLibrary function if not to
access Dll's>

Sam Schulenburg


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


From cfelling at iae.nl  Sat Jan 29 08:08:23 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 29 Jan 2000 14:08:23 +0100
Subject: Case Sensitivity and Learnability
References: <B4191059635E81B1.8AFDD386EFACF4A3.6FF161EA0151B5EB@lp.airnews.net> <86u4uf$26lq$1@thoth.cts.com>
Message-ID: <86uoo7$1a4$1@vvs.superst.iae.nl>

Will Rose <cwr at crash.cts.com> wrote:
> Tres Seaver <tseaver at starbase.neosoft.com> wrote:
> : Thinking about the notion of case sensitivity as a barrier to CP4E, as
...
> Your arguments are reasonable, but they are arguments against systems
> which are both case-insensitive _and_ case-preserving.  The trouble with
> CI/CP is that the compiler sees a different set of tokens from those seen
> by the programmer, and it's sometimes difficult for the programmer to

I don't think he is refering to this part of the problem. Swapping case at
random *on its own* makes the source hard to read, let alone this confusing
of what you see and what the compiler sees.

> realise this.  If unskilled programmers find case too difficult or
> distracting, the cure is to force a single case, either upper or lower. 
> This could be done by an (optional?) compiler switch.

this would make it impossible to import library modules that do depend on
case distinction. Or would it be a per module switch, settable in the source?
Now that will improve things:)

I, for one, vehemently dislike case insensitive languages, as they force me
to use ugly naming conventions like:

  class tv_class: pass 
  tv = tv_class()

yack, instead of the much clearer:

  class TV: pass
  tv = TV()

Admitted you have to know this convention of using case to distinct classes
from variables:), but then in natural language we use this same trick to
distinguish persons from objects like in Bill and bill. So how hard can it
be for a newbie:) OTOH, I'm easily convinced by proper usability studies.

-- 
groetjes, carel


From robin at jessikat.demon.co.uk  Sat Jan  1 12:07:22 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 1 Jan 2000 17:07:22 +0000
Subject: pickle != cPickle
Message-ID: <Q2upQCAKRjb4EwQS@jessikat.demon.co.uk>

I notice that cPickle and pickle sometimes give different results. This worries me as I thought
these two were now strictly comparable. What sort of things can cause this? It appears they
still give the same answer on unpickle.

eg
>>> dp=pickle.dumps(d['LNUX'])
>>> dc=cPickle.dumps(d['LNUX'])
>>> len(dp),len(dc)
(1502, 1525)
>>> dc
"(i__main__\012_YahooData\012p1\012(dp2\012S'Low'\012p3\012(lp4\012I233\012aI215\012......
>>> dp
"(i__main__\012_YahooData\012p0\012(dp1\012S'Low'\012p2\012(lp3\012I233\012aI215\012......
>>> np=pickle.loads(dp)
>>> nc=pickle.loads(dc)
>>> dir(np)==dir(nc)
1
>>> for s in dir(np):
...     print s, getattr(np,s)==getattr(nc,s)
... 
Close 1
High 1
Low 1
Open 1
Volume 1
date 1
finish 1
start 1
ticker 1
>>> 
-- 
Robin Becker


From scherbi at bam.com  Thu Jan 13 11:10:00 2000
From: scherbi at bam.com (Bill Scherer)
Date: Thu, 13 Jan 2000 11:10:00 -0500
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com>
Message-ID: <387DF8D7.95DE0101@bam.com>

M-x python-mode

should do it.

M-x load-file
may be needed first to load python-mode.el

(M == escape key for most, I think)

Bill Scherer

dharms98 at my-deja.com wrote:

> I know there is a way to force emacs to go into python mode even if the
> file doesn't end with .py by putting something at the top of the file.
>
> Can someone remind me of how to do that?
>
> Thanks,
>
> D.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000113/fe92c9fb/attachment.vcf>

From peter.stoehr at weihenstephan.org  Mon Jan 17 17:59:35 2000
From: peter.stoehr at weihenstephan.org (Dr. Peter Stoehr)
Date: Mon, 17 Jan 2000 23:59:35 +0100
Subject: Tkinter hangs again under Windows
References: <85v092$3n4$1@nnrp1.deja.com>
Message-ID: <38839ED7.7FAB2941@weihenstephan.org>

Hi Hans-Joachim,

how about posting the script or the relevant part of it?

    Peter

jbhr at my-deja.com wrote:
> 
> By searching the FAQ and Deja News, the last time I saw this
> mentioned was May 1999.
> 
> Python 1.5.2, newest Windows Installer with Tcl/Tk 8.0.5, Windows 95b
> 
> Well, now I have the problem: tk.mainloop() seems to just ignore
> any event, but only under certain circumstances.
> 
> My program opens a file and displays some information of it in a
> Listbox. If the program is started with a filename argument, this
> file is read, otherwise a filename is prompted via
> tkFileDialog.askopenfilename(). The info is then shown.
> Now, if the program is started with a filename, everything (well,
> almost :-)) is working fine - I can select items in the listbox,
> use the menu or the button. If the program is started without a
> filename, the file select box comes up, lets me select a file, and
> the info is then displayed as well. The difference starts when I try
> to select something - nothing happens. A Ctrl-C always breaks in
> Tkinter.py, line 492: mainloop.
> This problem doesn't exist under Linux. And it's the second Windows PC
> with the very same problem.
> 
> There's another (small) problem: I want the last entry in the listbox
> to be shown and selected by default. The selection works, but the line
> is only seen for a brief moment, then the list scrolls down one line for
> no apparent reason ... Just to give complete information: I'm using
> Fredrik Lundh's very handy AutoScrollbar.
> 
> This happened just when I wanted to show my boss how easy it is to
> write a nice little program with python ... ;-)
> 
> Hans-Joachim Widmaier
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.

-- 
---------------------------------------------------------------------------
   Dr. Peter Stoehr --- Teisenbergweg 6 --- 85435 Erding --- 08122/47232
      http://www.weihenstephan.org/home/ak/support/team/jpg/peter.jpg
---------------------------------------------------------------------------
		I'm the terror that flaps through the night


From mwh21 at cam.ac.uk  Sat Jan 22 11:09:19 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jan 2000 16:09:19 +0000
Subject: Python dumps core with this.. how come?
References: <38891316.393423@news.erols.com> <005301bf64b5$d2f234d0$f29b12c2@secret.pythonware.com> <3889c4b6.242514@news.erols.com>
Message-ID: <m3ln5i3x74.fsf@atrus.jesus.cam.ac.uk>

Pre-point: Python is Python, not some other programming language you
may have used in the past. When you are writing Python code, try to
write Python, not brutally translate C (or whatever) to Python. This
can be tricky, depending on how ingrained your habits are, but it's
the only path to happiness.

katz at glue.umd.edu (Roey Katz) writes:

> On Sat, 22 Jan 2000 09:50:58 +0100, "Fredrik Lundh"
> <fredrik at pythonware.com> wrote:
> 
> >looks like a recursive call in getattr (if "res" is
> >not already available in the instance, it asks
> >__getattr__ to look it up).
> 
> Thanks, I got it right after I posted :)
> this brings up another concern. Given
> the following definition:
> 
>    class A:
> 
>       def __getattr__( self, name ):
>         return -1    
>       def __setattr__( self, name, val ):
>          self.__dict__[ name ] = val
> 
> I see that self.__dict__ must be a special variable
> in that setting it somehow avoids a recursive call 
> to __setattr__()? 

Yup. Look in Objects/classobject.c around about line 580 if you want
the gory details.

> 
> 
> >> Also, I have another question (very old, and I can't find this in the
> >> FAQ):  I want to explicitly take a reference to an integer.
> >
> >1. reset your brain ;-)

This is the best advice in /F's post.

> >2. variables are named references, not small boxes
> >   that hold data values
> >3. *all* variables in python are references.  there
> >   is no other thing.
> >4. integers cannot be modified in place.
> 
> Ok, OK:  this is what I always get when I ask this question,
> but then why does this happen:
> 
>      >>> a = 3 
>      >>> b = a 
>      >>> b
>      3 
>      >>> a = 5
>      >>> b
>      3

Lets rewrite your example a bit:

d = {}
d['a'] = 3
d['b'] = d['a']
d['b'] ==> 3
d['a'] = 5
d['b'] ==> 3

That's more-or-less what's going on under the hood. The line 'a=5'
changes "what 'a' points at" and does *not* change "what" 'a' points
at. Is this helping?

> I mean, what happens if I have a complicated path 
> to a variable and I want to alias it:
> 
>     alias = long_name.hard_to_type[ 1 ].integer_value
> 
>     # arbitrary calculations that would be 
>     # torturous with any label longer than 'alias'
>     if alias > someValue:
>        result = (alias * 5 ) / (alias + 4 )
>     else:
>        result = (alias * 3 ) / (alias + 2 )

This is all fine so long as you don't need to change the value
referred to. You can alias up to the last '.', eg:

s = package.module.sub_module.object[2].sub_field
s.attr = s.attr + 1

> >in other words, Python always copies references, but
> >you cannot modify objects in place unless their inter-
> >face allows you to (numbers, tuples and strings cannot
> >be modified in place, while lists and dictionaries can --
> >the only difference is that the former doesn't provide
> >any methods that let you modify them...)
> 
> so where does this definition fit in the context I mentioned?
> I understand that if I have a function and pass to it a parameter,
> I can modify it in-place (although that I cannot modify numbers,
> strings or tuples in-place seems to me inconsistent).  
> So for example, these are roughly equivalent:
>  
>     void setValue(  int& x ) { x = 3  }    // C++
>     void setValue(  int *x  ) { *x = 3 }    // C
>     def  setValue( x ):    x = 3             # Python
> 
> whoops, we cannot modfiy integers in-place, how invonvenient.  Though
> we could always wrap them in a class and overload its = operator
> (wait! no = operator to overload! double inconvenience!) or have a
> specially defined set() function:
> 
>     class A:
>        def set( self, val ):
>             self.value = val
> 
>     def setValue( a ):
>        a.setValue( 5 ) 
> 
> so that works, but do you see how it is sort of polluting? I mean, the
> interface is not so simple anymore. 

None of the above are as simple as

def getValue(): return 3

I dislike this style you use in C, too. As Python has reference
semantics, you don't have the C gotcha of trying to return a four
megabyte structure on the stack, and a sensible exception system means
you don't need to have *every* function return a HRESULT <wink>.

To return more than one result, return a tuple. Or pass in some
instance object that more expresses what you are trying to acheive.

> >Q: when does Python copy a reference and not a value?
> >A: always!
> ayeee! missed it! terribly sorry :)
> 
> 
> ok, here's a third question, and it builds on the above inquiry
> regarding a lack of an assignment operator.  What happens when I want
> to do convert this for loop:
> 
>    for i,j in dest, src:
>         dest[ i ] = src[ j ]

I don't think you mean this. This is (loosely) equivalent to

dest[ dest[0] ] = src[ dest[1] ]
dest[ dest[0] ] = src[ dest[1] ]

I think you mean

for i in range(len(src)):
    dest[i] = src[i]

> into a fast map() expression:
> 
>    map( dest.assign_operator, src )

map(operator.__setitem__,[dest]*len(src),range(len(src)),src)

should do what you want;

map(dest.append,src)

or

map(dest.__setitem__,range(len(src)),src)

*may* do what you what, depending on circumtances.

Bizarrely, 

filter(dest.append,src)

is probably more memory efficient (the map form conses up a list of
return values, but as append always returns None - which is false -
filter `rejects' every element) - but if you write code like this, you
deserve to lose.

> like I mentioned -- there is no assignment operator to overload! 
> the trouble here is that there may be more situations in which 
> I cannot use  filter() to narrow down the list to only the elements
> that I want and then copy it into another list with src.copy().  

In general, don't be too obsessive about turning loops into maps and
filters. You're probably not saving as much as you think. for is far
clearer, and if every cycle is *that* precious, why Python?

(maphash #'(lambda (k v) (setf (gethash dest k) v)) src)-ly y'rs
Michael


From dfan at harmonixmusic.com  Wed Jan 19 10:28:31 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 19 Jan 2000 10:28:31 -0500
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <l03102803b4aa8c9e7bfe@[193.78.237.141]>
Message-ID: <wkzou2nkrk.fsf@turangalila.harmonixmusic.com>

Just van Rossum <just at letterror.com> writes:

| >> ! - whats wrong with tabs?
| >
| >They don't look the same in every editor, or with each individual's
| >tab-width settings.
| 
| But it's easy to turn that argument around:
| 
| Tabs can be nice *because* they "don't look the same in every
| editor, or with each individual's tab-width settings": they allow
| the *viewer* of the source code to decide what layout looks best for
| him/her. Tabs are also far superior if you prefer a proportional
| width font (yes, people do that, too!).

Almost without exception, when I am handed a file that uses tabs, I
can tell what tab width was used to generate the file, and all other
tab widths look ugly.  The most common tipoff is right-hand-side
comments that line up only with one particular tab width.

So my conclusion, after seeing hundreds of files that use tabs, is
that in general they don't successfully allow the viewer to decide
what layout looks best; the file's going to look bad unless you pick
the One True Tab Width.

-- 
Dan Schmidt | http://www.dfan.org


From hniksic at iskon.hr  Sat Jan  8 22:19:37 2000
From: hniksic at iskon.hr (Hrvoje Niksic)
Date: 09 Jan 2000 04:19:37 +0100
Subject: fastest way to merge lists.
References: <roy-D565B0.21255908012000@netnews.nyu.edu>
Message-ID: <9t97lhkvsmu.fsf@mraz.iskon.hr>

Roy Smith <roy at popmail.med.nyu.edu> writes:

> I'm not yet sure if care if the final list is sorted or not.  For
> now, I think it doesn't matter.  What's the fastest way to do that?
> It seems to me:
> 
> d = {}
> for s in list1:
>    d[s] = 1
> for s in list2:
>    d[s] = 1
> for s in list3:
>    d[s] = 1
> for s in list4:
>    d[s] = 1
> for s in list5:
>    d[s] = 1
> list = d.keys()
> del d   # if I care about memory
> 
> would be pretty good.  Anything better?

How about:

# cat
biglist = list1 + list2 + list3 + list4 + list5
# sort
biglist.sort()
# uniq
index = 0
while index < len(biglist) - 1:
  if biglist[index] == biglist[index + 1]:
    del biglist[index]
  else:
    index = index + 1

I haven't compared this solution with yours, but note that it's
directly equivalent to Unix-ese `cat | sort | uniq'.


From mlh at idi.ntnu.no  Wed Jan 19 11:17:45 2000
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Wed, 19 Jan 2000 17:17:45 +0100
Subject: JPython edit Component?
Message-ID: <864o7f$lih$1@kopp.stud.ntnu.no>

I just wondered if anyone had made a Java Component for editing Python -
i.e. a TextArea or something with indentation,
and preferrably with some rudimentary Emacs key-bindings... Possibly even a
JTextArea (or something) with
syntax coloring... And even an interactive shell would be nice... It would
be cool to have some IDLE-like functionality
(at least the basics) available in Java programs using JPython as an
extension language etc.

If someone has done something like this, or is interested in cooperating
with me in doing it, then sound off! :)

--

  Magnus
  Lie
  Hetland                                mailto:<anything>@hetland.org





From codeslave at worldnet.att.net  Sat Jan  8 10:46:53 2000
From: codeslave at worldnet.att.net (Eric Hausig)
Date: Sat, 08 Jan 2000 10:46:53 -0500
Subject: Error using string.strip
References: <38775227.2CC0D685@worldnet.att.net>
Message-ID: <38775BED.7D4595B3@worldnet.att.net>

Thanks to all for the response!  import string did it!

I'm going through the online tutorial at www.python.org.  Unfortunately,
they didn't make mention of import prior to the String section of the
tutorial.  Hehe....this is a really neat language so far.

Blushing a little,

Eric

Eric Hausig wrote:

> Hello there,
>
> I'm very new to Python.  (In fact, I'm walking through the tuturial
> right now!)
>
> I'm getting the following message:
>
> >>> string.strip('str') + 'ing'
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> NameError: string
> >>>
>
> I'm running Redhat 6.1 and have installed all of the dev rpms off the
> CD.   Is there some library I have to manually include?  If someone
> could give me a little insight, I would be thankful.  (PS - I'm
> currently a VB/SQL programmer trying to make the switch to Linux.  =)
>
> Sincerely,
>
> Eric Hausig
> aka Codeslave2k



From joec at brain.acme.org  Fri Jan 21 08:27:26 2000
From: joec at brain.acme.org (Joe Cotellese)
Date: Fri, 21 Jan 2000 13:27:26 GMT
Subject: Tkinter coverage in books? [Was: Re: Python and Tkinter Programming by John E. Grayson
References: <3880792E.79A874B6@weihenstephan.org> <869fo6$jcs$1@srv38.cas.org> <002d01bf6408$80654660$f29b12c2@secret.pythonware.com>
Message-ID: <m3vh4n1rak.fsf@brain.acme.org>

"Fredrik Lundh" <fredrik at pythonware.com> writes:

> lvirden at cas.org wrote:
> 
> http://www.pythonware.com/people/fredrik/tkintrobook.htm
> 
> </F>

Shameless plug? :-) 

Actually, I've been waiting for an update to these
tutorials for a while. Can't wait to get my hands on the book.

Joe Cotellese


From tescoil at rtpro.net  Sat Jan  1 13:14:24 2000
From: tescoil at rtpro.net (Tesla Coil)
Date: Sat, 01 Jan 2000 12:14:24 -0600
Subject: text-based builder
References: <386E21AF.BEFE9E1D@dnet.net.id>
Message-ID: <386E4400.F59A30A0@rtpro.net>

On 31 Dec 1999, Ahmad Munif wrote:
> i'm using wxPython to build GUI Applications.  but
> i don't know what tool that i can use to build text
> based application.

I found these to do what with maybe later...

Curses Programming with Python How-To:
http://www.python.org/doc/howto/curses/curses.html

ncursesmodule:
http://andrich.net/python/selfmade.html
(SuSE 6.3 has this in RPM, absent documentation
the tar provides).

Tinter:
http://office.iximd.com/~dwalker/tinter.py



From a.eyre at optichrome.com  Thu Jan 13 08:46:48 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Thu, 13 Jan 2000 13:46:48 -0000
Subject: CGI problem
In-Reply-To: <387C68DD.71D3F226@commsecure.com.au>
Message-ID: <000601bf5dcc$a3149590$3acbd9c2@peridot.optichrome.com>

> > > "(8) Exec format error exec of /usr/local/apache/htdocs/...
> > > Premature end of script headers: /usr/local/apache/htdocs/..

If you have access to the logs for apache, it should have logged
the traceback for the error. That might shen some light on the
problem.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From steve.mnard at sympatico.ca  Thu Jan 27 09:32:52 2000
From: steve.mnard at sympatico.ca (Steve Menard)
Date: Thu, 27 Jan 2000 09:32:52 -0500
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com> <slrn88pg1h.scl.wtanksle@hawking.armored.net> <m3r9f3veja.fsf@chinon.cnrs-orleans.fr>
Message-ID: <UIYj4.75$q64.2465@weber.videotron.net>

Konrad Hinsen <hinsen at cnrs-orleans.fr> wrote in message
news:m3r9f3veja.fsf at chinon.cnrs-orleans.fr...
> wtanksle at hawking.armored.net (William Tanksley) writes:
>
> > I have a question, though.  What's that about the Palm?  When will the
> > Palm version be available?  What will it require?  What will it provide?
> > Will it only work on the Palm III, or will it be okay on anything above
as
> > well (Palm V, Visor, TRG, etc)?
>

    Any timefame for the palm port? Any details? I have looked at doing a
port myself, and am still in the early stages. I'd like to know what
features of python will be supported and of course how long I'll have to
wait :)

                Steve




From aa8vb at yahoo.com  Wed Jan  5 08:40:52 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 5 Jan 2000 08:40:52 -0500
Subject: Python and GUI options
In-Reply-To: <wkr9g1gyr1.fsf@sccsi.com>
References: <84lreq$hib$1@nnrp1.deja.com> <386e8a42@omega> <84lreq$hib$1@nnrp1.deja.com> <84lv5k$rqv$1@news1.xs4all.nl> <wkr9g1gyr1.fsf@sccsi.com>
Message-ID: <20000105084052.B9441@vislab.epa.gov>

Jody Winston:
 |On the SGI wxWindow and GTK still are not as stable as either Linux or
 |Windows 98

Second Jody's comment.  Also on UNIX, wxPython/wxGTK/wxWindows has more
geometry negotiation problems (widgets trees not arranging correctly;
overlapping, etc.) than Tk/Tkinter does.

For now I'll stick with Tkinter, but keep an eye on wx* and a few of the
others.

-- 
Randall Hopper
aa8vb at yahoo.com



From hildeb at www.stahl.bau.tu-bs.de  Fri Jan 21 09:45:50 2000
From: hildeb at www.stahl.bau.tu-bs.de (Ralf Hildebrandt)
Date: 21 Jan 2000 14:45:50 GMT
Subject: How to commit suicide with a SIGSEGV?
Message-ID: <slrn88gs8u.77g.hildeb@stahlw06.stahl.bau.tu-bs.de>

I need the python interpreter to "commit suicide" with a special signal. 
How do I get the pid of the interpreter to be able to use s.th. like

os.kill(mypid, signal.SIGSEGV)


From pinard at iro.umontreal.ca  Thu Jan 20 18:11:42 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 20 Jan 2000 18:11:42 -0500
Subject: Small `du' reformatter :-)
Message-ID: <oqogagfidt.fsf@titan.progiciels-bpi.ca>

Hi, people.

I knew I wrote a small Python tool to reformat `du' output, but I lost it
for a while, and unexpectedly stumbled on it today, looking for something
unrelated.  Here it is.  So if I loose it again, I'll ask you! :-)

-------------- next part --------------
#!/usr/bin/python
# Produce directory hierarchy with sizes, well sorted and indented.
# Copyright ? 1996, 1999 Progiciels Bourbeau-Pinard inc.
# Fran?ois Pinard <pinard at iro.umontreal.ca>, 1996.

# Idea from Pierre Rioux <riouxp at bch.umontreal.ca>, 1996-07-12.

import os, re, string, sys

def main(*arguments):
    # Get data properly sorted.
    items = map(presort,
                os.popen('du %s' % string.join(arguments)).readlines())
    items.sort()
    items = map(postsort, items)
    # Erase unneeded fragments of vertical lines.
    clean(items)
    items.reverse()
    clean(items)
    items.reverse()
    # Erase leftmost white columns.
    skip = 0
    split, size = items[0]
    while not split[skip]:
        skip = skip + 1
    # Produce display.
    for split, size in items:
        sys.stdout.write('%7s ' % size)
        for item in split[skip:-1]:
            if item:
                sys.stdout.write('|  ')
            else:
                sys.stdout.write('   ')
        sys.stdout.write('\\_ %s' % split[-1])

def presort(line):
    size, path = string.split(line, '\t')
    return path, size

def postsort((path, size)):
    return string.split(path, '/'), size

def clean(items):
    # The basename is always written.  Intermediate directories
    # normally print as `|', yet None is used to inhibit the line.
    draw = []
    for split, size in items:
        while len(draw) > len(split) - 1:
            draw.pop()
        while len(draw) < len(split) - 1:
            draw.append(0)
        draw.append(1)
        for counter in range(len(split) - 1):
            if not draw[counter]:
                split[counter] = None

if __name__ == '__main__':
    apply(main, sys.argv[1:])
-------------- next part --------------

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

From aahz at netcom.com  Wed Jan 26 11:05:05 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jan 2000 16:05:05 GMT
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com>
Message-ID: <86n5vh$q2$1@nntp5.atl.mindspring.net>

In article <388E7D5B.7840983F at erols.com>,
Edward C. Jones <edcjones at erols.com> wrote:
>
>GameSpy posted a six-page essay of Tim Sweeney's entitled "A
>Critical Look at Programming Languages." Tim Sweeney is, of
>course, the lead coder for the Unreal game engine (one of the
>most licensed 3D game engines for the PC.)
>
>The article is at http://www.gamespy.com/articles/devweek_b.shtm

Thanks!  I think I saw a few wrong details, but it was a really good
overview.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From roy at popmail.med.nyu.edu  Tue Jan  4 21:01:54 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Tue, 04 Jan 2000 21:01:54 -0500
Subject: how to copy a dictionary
Message-ID: <roy-5C6597.21015404012000@netnews.nyu.edu>

If I do:

d = {'hedgehog': 'spiney norman'}
temp = d
temp['scotsman'] = 'earnest potgorney'

I end up changing the original dictionary, d.  It's obvious that what's 
going on is when I do temp = d I get a pointer to the same object 
instead of a new object.  My question is, how do I force a new object to 
be created, so when  modify it, I don't also modify the original?


From a.eyre at optichrome.com  Wed Jan  5 13:15:24 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Wed, 5 Jan 2000 18:15:24 -0000
Subject: Executing a script.
In-Reply-To: <84io2q$ki2$1@nnrp1.deja.com>
Message-ID: <002501bf57a8$d6099e20$3acbd9c2@peridot.optichrome.com>

> I am using NT 4.0 (sp 5) and MSVC 6.0 (sp3) if the makes a difference.

http://www.python.org/doc/FAQ.html#8.7

http://www.python.org/doc/FAQ.html#8.10

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From hannah at schlund.de  Wed Jan 12 09:21:54 2000
From: hannah at schlund.de (Hannah Schroeter)
Date: 12 Jan 2000 14:21:54 GMT
Subject: Imaging-1.0 Compilation & OpenBSD
References: <Pine.BSO.4.10.10001111652020.6301-100000@doug.speedex.net>
Message-ID: <85i2m2$i8u$1@c3po.schlund.de>

Hello!

In article <Pine.BSO.4.10.10001111652020.6301-100000 at doug.speedex.net>,
Douglas Shawhan  <doug at doug.speedex.net> wrote:
>[...]
>gcc -fpic  -IlibImaging -I/usr/local/include -I/usr/local/include
>-DWITH_TKINTER -I/usr/local/include -g -O2 -I/usr/local/include/python1.5
>-I/usr/local/include/python1.5 -DHAVE_CONFIG_H -c ./_imaging.c
>[...]
>./_imaging.c:2303: tk.h: No such file or directory
>*** Error code 1

>--------------------end my pain--------------------------
>tk.h does indeed exist:

>/usr/local/include/tk8.0/tk.h

>Any clues?

Provide -I/usr/local/include/tk8.0 to the compiler, too.

>Thanks.

>d

Regards, Hannah.


From bitbucket at isomedia.com  Thu Jan  6 11:31:56 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Thu, 06 Jan 2000 16:31:56 GMT
Subject: Simple cartesian product
References: <000801bf571c$0cf448a0$1a2d153f@tim> <iacembwr7qg.fsf@vier.idi.ntnu.no> <Pine.LNX.4.10.10001061110050.6929-100000@gaia.gfy.ku.dk> <20000106132958.E7015@xs4all.nl>
Message-ID: <3874bff0.409436067@news.isomedia.com>

If you don't want to be mistaken for an American, you should 1) know
what ISO stands for and 2) never ask whether ISO is a "real standard
in the ANSI sense."  To do the latter is similar to asking whether
Guiness is beer in the "Budweiser" sense.

We Americans shouldn't have too much trouble with YYYY-MM-DD.  In my
neck of the woods, we're used to seeing it always MM/DD (Year assumed)
or MM-DD-YY.  The year rarely makes it to the front, so YYYY-MM-DD is
completely foreign and thus presents a fresh start to actually think
about it, instead of simply whining "it's too much work to change and
do it right."*  The more observant can take a clue anytime in the
second half of each month.

Besides the good reason to use the proper international date format
because that's what most** of the world does, I prefer to use it
because it sorts properly. :)

     -Eugene
* Retort: it costs twice as much to buy two sets of all the tools.
<grr.>
** ('most' has any number of definitions.)


On Thu, 6 Jan 2000 13:29:59 +0100, Thomas Wouters <thomas at xs4all.net>
 wrote:

>On Thu, Jan 06, 2000 at 11:13:29AM +0100, Thomas Knudsen wrote:
>
>> > Now - you're not going to get that "new millennium"-discussion
>> > started, are you? We are still in the same millennium, century, and
>> > decade as last year. The third millennium starts with 2001.01.01.
>
>> You should, however, use proper ISO8601 syntax and write 2001-01-01
>> This is actually quite important: We really *need* a consistent,
>> global format for numeric date representation. And since we actually
>> have it - *use it!*
>
>So is this YYYY-MM-DD or YYYY-DD-MM ? This creates a lot of confusion
>whenever I need to exchange dates with americans, or with people who think
>_I'm_ american. (the nerve.) Or is ISO8601 a real standard in the
>ANSI sense, and thus claim two different things in different sections of the
>standard, and later claim it's *both* YYYY-MM-DD and YYYY-DD-MM ? :)
>
>History---will-teach-us-nothing-ly y'rs.
>-- 
>Thomas Wouters <thomas at xs4all.net>
>
>Hi! I'm a .signature virus! copy me into your .signature file to help me spread!
>

import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From joec at brain.acme.org  Fri Jan 21 09:54:19 2000
From: joec at brain.acme.org (Joe Cotellese)
Date: Fri, 21 Jan 2000 14:54:19 GMT
Subject: database connection
References: <s8grjo5jeh697@corp.supernews.com>
Message-ID: <m3k8l31n9q.fsf@brain.acme.org>

"David Metzler" <davidm at softdesigns.com> writes:

> I would like to create some py programs that connect to a database like
> mysql or postgres on either win 98 or Linux.
> I could not find the file to include. The python.org directions talk about
> compiling but I do not have MS C.
> Is there a mysqldb available to install in Win 98 without compiling.
> How about a rpm or binary for RH 6.0
> 
> If not for mysql perhaps another database?

I'm using a Python MYSQL module. I don't remember where I got it but I
do recall that two exist. The module I use is MySQLdb. I seem to
remember having problems getting this compiled. I don't remember
exactly what the problems were.

Hope that helps
Joe


From b.k.Hoover at unet.com  Sun Jan  2 11:25:49 2000
From: b.k.Hoover at unet.com (b.k.Hoover at unet.com)
Date: 2 Jan 2000 16:25:49 GMT
Subject: New stuff for our newsgroup
Message-ID: <VUI463A2.4FNF3T66@unet.com>

I found this great homepage with lots of new stuff:

http://everybody.isfunky.com

Have fun!!!


From bsb at winnegan.de  Thu Jan  6 04:21:29 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 06 Jan 2000 10:21:29 +0100
Subject: module path:funny bug, funny solution
In-Reply-To: Pierre Imbaud's message of "Thu, 06 Jan 2000 00:49:08 +0100"
References: <3873D874.62A35E7B@saiph.com>
Message-ID: <877lhnftd2.fsf@baal.winnegan.de>

Pierre Imbaud <pierre at saiph.com> writes:

> I run into a funny bug, working on a python script, importing modules
> laying in the same directory. The script performed some chdir. I debug
> under xemacs, executing the script with py-execute-buffer. At some
> point, I had to modify an imported module: I put a reload(module) in
> the code: surprise! raises a "ImportError: No module named xx"!
> what happens is quite simple: sys.path[0] is '', that means: first
> search for modules in the current directory. First time, modules are
> found ok. when trying to reload, the current directory has changed!
> Now, I think its funny. I didnt when it happened.
> 
> The solution I found? between import os and import custom modules:
> 
> try:
>     os.chdir(startDirectory)            # succeeds at subsequent runs
> except:
>     startDirectory = os.getcwd()        # run at first run.
> 
> well, it works, but may I let such a puzzle in my code? The
> unadvertized reader will think Im plain mad...

There is no need to chdir just for reloading modules, simply change
sys.path[0] upon startup. If you are concerned about others looking at 
your code, you might use the technique from the attached files

# Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******
-------------- next part --------------
A non-text attachment was scrubbed...
Name: debugging.py
Type: application/octet-stream
Size: 265 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000106/5e80ed7e/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: debugme.py
Type: application/octet-stream
Size: 162 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000106/5e80ed7e/attachment-0001.obj>

From gmcm at hypernet.com  Tue Jan 18 17:50:38 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 18 Jan 2000 17:50:38 -0500
Subject: **kwargs
In-Reply-To: <862j09$o1k$1@nnrp1.deja.com>
Message-ID: <1263886656-955174@hypernet.com>

sp00fD writes:

> Does **kwargs represent a dictionary?  

Yes.

> If so, can I have a method like
> such
> 
> def foo(self, foo=None, bar=None, **kwargs)
> 
> and then do something like
> 
> this_dict = {}
> ..insert stuff into this_dict..
> 
> my.foo(foo="something", bar="something else", this_dict)

Not quite. To call foo with the named args wrapped up as a 
dictionary, you need to use apply:
  apply(foo, ("something", "something else"), this_dict)
or
  this_dict['foo'] = "something"
  this_dict['bar'] = "something else"
  apply(foo, (), this_dict)
 
Note that foo cannot tell which of the above ways was used, 
(ie, "foo" and "bar" are not in kwargs when foo examines it).

BTW, while foo-abuse is common, I've never seen it carried to 
such extremes.

- Gordon



From gmcm at hypernet.com  Sun Jan  9 19:17:59 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 9 Jan 2000 19:17:59 -0500
Subject: control structures (was "Re: Sins")
In-Reply-To: <3878B8EE.B4939F3A@connection.com>
Message-ID: <1264659011-22815940@hypernet.com>

Colin J. Williams wrote:
> 
> skaller wrote:
> > 
> > Gordon McMillan wrote:

[John Skaller expresses disgust with searching a list with 
"for", because he finds needs to use a flag]
> > >
> > > That's the reason that Python has the "for
> > > .. else" construct that William Tanksley
> > > hates so much.
> > >
> > >  for i in X:
> > >    if X[i]==e:
> > >      # do the "found" thing
> > >      break
> > >  else:
> > >    print "Not Found"
> > 
> > That doesn't work. The 'do the "found" thing'
> > is in the wrong place: it must be outside the loop,
> > since it is usually the 'normal continuation'
> > of the routine.

[Colin] 
> Aside from that.  

Aside from what? In the Skaller vocabulary "broken" and 
"doesn't work" mean "are not according to Skaller 
preferences". He likes to think he's shocking people, but he 
has only drained those terms of meaning.

>It seems to me that the weakness of the
>    for ...
>    else 
> structure is that the else clause is executed whether the 
> 'for' succeeds or not. The Language Reference Manual has:
> 
>  "The expression list is evaluated once; it should yield a    
> sequence. The suite is then executed once for each item in the
> sequence, in the order of ascending indices. Each item in turn is
> assigned to the target list using the standard rules for
> assignments, and then the suite is executed. When the items are
> exhausted (which is immediately when the sequence is empty), the
> suite in the else clause, if present, is executed, and the loop
> terminates."

I don't know what you mean by "succeeds". The "else" clause 
is skipped if the loop is terminated.
 
> Illustrative code is appended.
> 
> More useful would be to execute the else clause only if the
> sequence is empty.  This might be expressed as:

As one who regularly uses the searching algorithm above, I 
find it highly useful. I find "if seq:" to be useful for testing for an 
empty sequence. I'm not going to defend the spelling of the 
construct, or it's description in the Language Ref, but it *is* 
very useful.

- Gordon



From nhv at cape.com  Fri Jan 14 20:14:22 2000
From: nhv at cape.com (Norman Vine)
Date: Fri, 14 Jan 2000 20:14:22 -0500
Subject: Tcl/Tk Path problems in Python 1.5.2 installation
References: <387F80E5.A0AF82A5@sourcesignal.com>
Message-ID: <85ohls$28s1$1@newsie2.cent.net>

David Nichols  wrote :

> I just installed python 1.5.2 (running py152.exe and answering 'yes - I
> want to install Tcl/Tk') and have been unable to get the path set
> up correctly for graphical applications.  For instance, if I go to the
> idle directory (//E/Programs/Py152/Tools/idle) and try to launch it I
> get the following result.
>
> ================================
> BASH.EXE-2.02$ python idle.

Hi David,

Are you trying to run python from a Cygwin bash shell ??

the (//E/Programs/Py152/Tools/idle) makes me think so !

If so, I think you will have to build your own Python with Cygwin.

Norman Vine





From JamesL at Lugoj.Com  Tue Jan 25 00:41:25 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Mon, 24 Jan 2000 21:41:25 -0800
Subject: OFFTOPIC: *-ly y'rs,
References: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com> <3dvh4jz0m9.fsf@amarok.cnri.reston.va.us>
Message-ID: <388D3785.8BB2FBE4@Lugoj.Com>

"Andrew M. Kuchling" wrote:
> 
> Agent Drek <drek at MonsterByMistake.Com> writes:
> > Where did they come from and are they patented :)
> > How can one join the signoff-ly club?
> 
> The first occurrence, at least in the python-list archives, was quite
> a while ago:
> 
>   in-any-case-the-best-christmas-present-i-got-today!-ly y'rs - tim
>         -- Tim Peters, 29 Dec 91
> 
> I imagine Tim probably used this on other mailing lists before
> python-list.

Unless that sign off is incorporated into Python 2 to indicate termination
of a block rather than indent/dedent, thousands of programmers will avoid
using Python.


From billtut at microsoft.com  Wed Jan  5 16:04:14 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Wed, 5 Jan 2000 13:04:14 -0800 
Subject: Python and ADSI (more ADSI really)
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BCBB9@RED-MSG-50>


> -----Original Message-----
> From: Paul Robinson [mailto:paul.robinson at quantisci.co.uk]
> 

> [WINNT://mycomputer abysmially slow vs. WINNT://mydomain/mycomputer]

Another trick is to try this suggestion:

If you want to bind to a specific computer (instead of using a domain name),
then add ',computer' clause as part of the ADsPath to speed it up. 

e.g Set comp = GetObject("WinNT://mycomputer,computer")
    Set usr = comp.GetObject("user", "jsmith")

Bill





From bjorn at roguewave.com  Mon Jan  3 13:03:50 2000
From: bjorn at roguewave.com (bjorn)
Date: Mon, 03 Jan 2000 11:03:50 -0700
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <386E7BD4.514764EA@ndh.net> <84nv0u$bu9$1@nntp9.atl.mindspring.net> <3870DB37.761034EA@roguewave.com> <75B5A3A6EB0EDE04.FCE8CED8D9B414D5.95006D8FA0ECEB31@lp.airnews.net>
Message-ID: <3870E485.998FF2FD@roguewave.com>


Tres Seaver wrote:

> In article <3870DB37.761034EA at roguewave.com>,
> bjorn  <bjorn at roguewave.com> wrote:
> >Aahz Maruch wrote:
> >
> >> In article <386E7BD4.514764EA at ndh.net>,
> >> Stefan Schwarzer  <s.schwarzer at ndh.net> wrote:
> >> >
> >> >The other topics which the author on the O'Reilly site, Greg Wilson,
> >> >mentions might also be worth considering. For example, I never needed
> >> >tuples (explicitly) yet, i. e. in a sense that lists would not work.
> >> >(If tuples would be taken out of the language or their use deprecated)
> >> >tuple un/packing would become list un/packing and so on.)
> >>
> >> Anyone who claims that tuples are unnecessary has never used a tuple as
> >> a dict key.  That requirement necessitates differentiating between a
> >> mutable and immutable list -- might as well stick with tuples.
> >> --
> >
> >So what you're saying is that dicts are broken since they can't take
> >lists as keys <wink>.
>
> What reasonable use can you propose for an associative container which allows
> mutable keys?  You might as well revert to a list of (key,value) pairs, since
> you're going to do a linear search on each lookup anyway if the keys are
> allowed to change.

If you look at any hashtable implementation for C, you'll see that the keys can
be char* (or any other mutable type for that matter).  There are really only two
requirements for the keys in an associative container (i) they must be hashable,
and (ii) they must remain unchanged for the duration of the container (or more
specifically, the hash value must remain unchanged).  It's might be nice (for a
suitable definition of nice) to have the language hold your hand through this,
but it is far from neccessary.  Sometimes it really does suffice to tell the
programmer that if he changes a key he'll get undefined behavior.

-- bjorn




From robin at jessikat.demon.co.uk  Thu Jan 20 15:00:27 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 20 Jan 2000 20:00:27 +0000
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu> <FonA21.B26@world.std.com>
Message-ID: <gzg1xRAbl2h4EwWf@jessikat.demon.co.uk>

In article <FonA21.B26 at world.std.com>, Will Ware <wware at world.std.com>
writes

I'm certainly not going to be the one to raise the issue of
boustrophedon (back and forth) indenting.

>Paul M (paul.magwene at yale.edu) wrote:
>: Dear Pythoneers...
>
>A great evil has been perpetuated in the design of the Python
>language. It grieves me to be the one to bring this to the world's
>attention but it must not continue to go unacknowledged. Persons of
>right indentation have once again been marginalized and excluded
>because _every_editor_ever_written_ indents from the LEFT SIDE OF THE
>PAGE. The brutal, systematic, uniform, codified, societally
>legitimized vicious assault on right indentation has gone unnoticed,
>despite my many pleas to Greenpeace and Amnesty International. I am
>forced to bring this issue to the very feet of the Python community
>itself, and beg, nay, demand, that the humanity of right-indenters be
>recognized and respected. Let the bloodshed end here.
>
>Right-indenters are not a vindictive lot. We do not insist that the
...
-- 
Robin Becker


From skip at mojam.com  Thu Jan  6 14:21:38 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 6 Jan 2000 13:21:38 -0600 (CST)
Subject: control structures (was "Re: Sins")
In-Reply-To: <005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
	<14452.45436.641890.197509@beluga.mojam.com>
	<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
Message-ID: <14452.60226.853060.116123@beluga.mojam.com>

    Darrell> Don't know if this if fair. But exceptions are slow when raised.

Agreed in the case of your two-branch case.  If you have a state machine
with a bunch of states, try/except/except/... appears to perform about as
well as if/elif/elif/.../else:

    import string, time

    def func1(size):
	s = string.letters + string.digits
	for i in xrange(size):
	    for l in s:
		try:
		    raise intern(l)
		except "a":
		    pass
		except "b":
		    pass
		except "c":
		    pass
		except "d":
		    pass
		except "e":
		    pass
		except "f":
		    pass
		except "g":
		    pass
		except "h":
		    pass
		except "i":
		    pass
		except "j":
		    pass
		except "k":
		    pass
		except "l":
		    pass
		except "m":
		    pass
		except "n":
		    pass
		except "o":
		    pass
		except "p":
		    pass
		except "q":
		    pass
		except "r":
		    pass
		except "s":
		    pass
		except "t":
		    pass
		except "u":
		    pass
		except "v":
		    pass
		except "w":
		    pass
		except "x":
		    pass
		except "y":
		    pass
		except "z":
		    pass
		except "A":
		    pass
		except "B":
		    pass
		except "C":
		    pass
		except "D":
		    pass
		except "E":
		    pass
		except "F":
		    pass
		except "G":
		    pass
		except "H":
		    pass
		except "I":
		    pass
		except "J":
		    pass
		except "K":
		    pass
		except "L":
		    pass
		except "M":
		    pass
		except "N":
		    pass
		except "O":
		    pass
		except "P":
		    pass
		except "Q":
		    pass
		except "R":
		    pass
		except "S":
		    pass
		except "T":
		    pass
		except "U":
		    pass
		except "V":
		    pass
		except "W":
		    pass
		except "X":
		    pass
		except "Y":
		    pass
		except "Z":
		    pass
		except "0":
		    pass
		except "1":
		    pass
		except "2":
		    pass
		except "3":
		    pass
		except "4":
		    pass
		except "5":
		    pass
		except "6":
		    pass
		except "7":
		    pass
		except "8":
		    pass
		except "9":
		    pass

    def func2(size):
	s = string.letters + string.digits
	for i in xrange(size):
	    for l in s:
		if l == "a":
		    pass
		elif l == "b":
		    pass
		elif l == "c":
		    pass
		elif l == "d":
		    pass
		elif l == "e":
		    pass
		elif l == "f":
		    pass
		elif l == "g":
		    pass
		elif l == "h":
		    pass
		elif l == "i":
		    pass
		elif l == "j":
		    pass
		elif l == "k":
		    pass
		elif l == "l":
		    pass
		elif l == "m":
		    pass
		elif l == "n":
		    pass
		elif l == "o":
		    pass
		elif l == "p":
		    pass
		elif l == "q":
		    pass
		elif l == "r":
		    pass
		elif l == "s":
		    pass
		elif l == "t":
		    pass
		elif l == "u":
		    pass
		elif l == "v":
		    pass
		elif l == "w":
		    pass
		elif l == "x":
		    pass
		elif l == "y":
		    pass
		elif l == "z":
		    pass
		elif l == "A":
		    pass
		elif l == "B":
		    pass
		elif l == "C":
		    pass
		elif l == "D":
		    pass
		elif l == "E":
		    pass
		elif l == "F":
		    pass
		elif l == "G":
		    pass
		elif l == "H":
		    pass
		elif l == "I":
		    pass
		elif l == "J":
		    pass
		elif l == "K":
		    pass
		elif l == "L":
		    pass
		elif l == "M":
		    pass
		elif l == "N":
		    pass
		elif l == "O":
		    pass
		elif l == "P":
		    pass
		elif l == "Q":
		    pass
		elif l == "R":
		    pass
		elif l == "S":
		    pass
		elif l == "T":
		    pass
		elif l == "U":
		    pass
		elif l == "V":
		    pass
		elif l == "W":
		    pass
		elif l == "X":
		    pass
		elif l == "Y":
		    pass
		elif l == "Z":
		    pass
		elif l == "0":
		    pass
		elif l == "1":
		    pass
		elif l == "2":
		    pass
		elif l == "3":
		    pass
		elif l == "4":
		    pass
		elif l == "5":
		    pass
		elif l == "6":
		    pass
		elif l == "7":
		    pass
		elif l == "8":
		    pass
		elif l == "9":
		    pass

    size=1000
    t1=time.time()
    func1(size)
    print 'Time:',time.time()-t1

    t1=time.time()
    func2(size)
    print 'Time:',time.time()-t1

Running the above yields:

Time: 5.52181506157
Time: 4.23444199562

;-)

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 gvwilson at nevex.com  Fri Jan 21 05:18:48 2000
From: gvwilson at nevex.com (Greg Wilson)
Date: Fri, 21 Jan 00 10:18:48 GMT
Subject: Open Source Design Competition - Software Carpentry
Message-ID: <mt2.0-27365-948449928@news.informatik.uni-bonn.de>

Los Alamos National Laboratory                     CodeSourcery, LLC

                          Software Carpentry
                  http://www.software-carpentry.com

                    Open Source Design Competition

                         $100,000 in Prizes!

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

The Software Carpentry project is pleased to announce its first Open
Source design competition, with prizes totaling $100,000.  Students
and professionals from any country, working individually or in teams,
are invited to submit design outlines for:

* a platform inspection tool to replace autoconf;

* a dependency management tool to replace make;

* an issue tracking system to replace gnats and Bugzilla; and

* a unit and regression testing harness with the functionality of
  XUnit, Expect, and DejaGnu.

The best four entries in each category will be awarded $2500, and
invited to submit full designs by June 1, 2000.  The best design in
each category will then receive an additional $7500, while runners-up
will each receive $2500.  Once winning designs have been announced,
$200,000 will be available through open bidding for implementation,
testing, and documentation.

Participants may submit separate entries in one or more categories by
March 31, 2000.  Entries must be in English, and no more than 5000
words long.  For more information, see the Software Carpentry web site
at http://www.software-carpentry.com.  All of the project's work will
be Open Source; all tools will be written in, or scriptable with,
Python, and will be required to run on both Linux and Microsoft
Windows NT.

The competition will be judged by a panel that includes the following
noted software developers, authors, and computational scientists:

    Stephen Adler       Brookhaven National Laboratory
    Frank Alexander     Los Alamos National Laboratory
    Donnie Barnes       Red Hat
    Chris DiBona        VA Linux
    Paul Dubois         Lawrence Livermore National Laboratory
    Andrew Hunt         Pragmatic Programmers, LLC
    Stephen R. Lee      Los Alamos National Laboratory
    Josh MacDonald      University of California, Berkeley
    Brian Marick        Reliable Software Technologies
    Doug Mewhort        Queen's University
    Bruce Perens        co-founder of the Open Source Initiative
    Dave Thomas         Pragmatic Programmers, LLC
    Jon Udell           author of Practical Internet Groupware
    Guido van Rossum    inventor of Python
    Tom Van Vleck       TransIlluminant
    Phil Wadler         Bell Labs
    Scot Wingo          AuctionRover

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

The Software Carpentry project is sponsored by the Advanced Computing
Laboratory at the U.S. Department of Energy's Los Alamos National
Laboratory (http://www.acl.lanl.gov), and administered by
CodeSourcery, LLC (http://www.codesourcery.com).  The project's aim is
to encourage adoption of better software development practices by
making software tools easier to use, and by documenting design,
testing, and related activities.  For more information on the project,
or to let us know that you intend to submit a proposal, see
http://www.software-carpentry.com, or mail info at software-carpentry.com.

-- 
----------- 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 a.eyre at optichrome.com  Mon Jan 31 11:05:37 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 16:05:37 -0000
Subject: IOError: (0,"Error") when closing "os.popen"ed file?
In-Reply-To: <14485.44976.913646.228004@temoleh.chem.uu.nl>
Message-ID: <000f01bf6c05$03519de0$3acbd9c2@peridot.optichrome.com>

> I could try to upgrade this machine (it appears to be a RH6.0 and not
> 6.1 as I thought earlier). But if that doesn't fix it, how would I go
> about tracing the problem? Do I have to edit "printf's" into the
> python source? I'm running quite a large package (controlling a data
> acquisition robot, 45.1k lines of Python) to produce this problem, and
> the occurrence seems to be timing-dependent.  Horrible debugging
> scenarios doom in front of me....

I'd start off by taking Python out of the equation, and writing a mini
app which uses the standard posix popen function. See if you can
duplicate the problem. Read Python-1.5.2/modules/posixmodule.c for the
gory details.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From rgruet at ina.fr  Tue Jan  4 08:02:31 2000
From: rgruet at ina.fr (Richard Gruet)
Date: Tue, 04 Jan 2000 14:02:31 +0100
Subject: IDLE breakpoints and Go debugging command
Message-ID: <3871EF67.961E8E4E@ina.fr>

Hi there,

I can't get a breakpoint working with IDLE (0.5) and incidentally can't
GO in the debugger
.
I know that I have to open the debugger to avoid getting a beep, so I
open the debugger,
set a breakpoint on a source line via the right-button menu: no beep
that time, but no
feed-back is visible to confirm that the operation succeeded.
And when I launch the execution (run script), the debugger stops on 1st
line but subsequent 'go' only step one line (like a step over) instead
of running until the breakpoint is reached.

Is this behaviour known ? is it a bug ? I searched the archives but
found no answer.
I'm working on NT 4.0 SP3, Python 1.5.2, IDLE 0.5 (last CVS snapshot).

Thanks for any help,

Richard



From JamesL at Lugoj.com  Fri Jan 21 11:41:43 2000
From: JamesL at Lugoj.com (James Logajan)
Date: Fri, 21 Jan 2000 08:41:43 -0800
Subject: Using tabs effectively. Was: Why not use tabs?
In-Reply-To: <1263658531-14676592@hypernet.com>
References: <3887DF0D.B102B357@Lugoj.Com>
Message-ID: <3.0.6.32.20000121084143.00916800@pop.ablecom.net>

Are you displaying e-mail using a Courier font (or other fixed-width font)
or with a variable width font? If the latter, the x characters may not line
up.

Both were cut and paste from vi. The first was with "set list" (I did erase
the "$" at the end of each line however). The tab setting was set to 4. All
other values between 1 and 8 will not make the x characters line up.

Otherwise I'm baffled as to why you can't reproduce the indented (intended)
result. Unless there really *is* a conspiracy going on. Hmmm....

At 08:49 AM 01/21/2000 -0500, Gordon McMillan wrote:
>James Logajan wrote:
>
>> .... For example, consider a file that contains the
>> following text lines, where ^I indicates a tab:
>> 
>> A^I^Ix
>> AA^I^Ix
>> AAA^I^Ix
>> AAAA^Ix
>> AAAAA^Ix
>> AAAAAA^Ix
>> AAAAAAA^Ix
>> 
>> The intent is to have that look like:
>> 
>> A       x
>> AA      x
>> AAA     x
>> AAAA    x
>> AAAAA   x
>> AAAAAA  x
>> AAAAAAA x
>> 
>> There is only ONE tab setting that will produce that result. 
>
>Um, there's no tabsetting that will produce that result, but with 
>the wonders of email, maybe that's not what you wrote.
>
>
>- Gordon
>
>





From gregc at cgl.ucsf.edu  Tue Jan 18 22:42:21 2000
From: gregc at cgl.ucsf.edu (Greg Couch %CGL)
Date: 19 Jan 00 03:42:21 GMT
Subject: Compiling Python under Borland C++ Builder 4
References: <9nKf4.59481$Ke.56920@news21.bellglobal.com>
Message-ID: <gregc.948253341@cgl.ucsf.edu>

Nice to know that there are other people using C++ Builder 4 with Python.

We've been successful just using BCB for .pyd's and .dll's.  That way we
can use the standard Windows Python distribution.

There are a couple of tricks:

1.	When making a BCB python15.lib from python15.dll, implib doesn't
	put leading underscores on symbols that already have an underscore.
	The workaround is to use impdef, fix up each line to rename symbols,
	and use implib on that.

2.	BCB puts leading underscores on all symbols, so Python fails to
	import the pyd/dll correctly (Python expects initMODULE and BCB
	generates _initMODULE).  The solution is to add a MODULE.def with
	the following:

		EXPORTS
			initMODULE = _initMODULE
	
	and link it in when building the pyd/dll.
	
Hope this helps,

	Greg Couch
	UCSF Computer Graphics Laboratory
	gregc at cgl.ucsf.edu


From b2blink at hotmail.com  Mon Jan 17 13:33:26 2000
From: b2blink at hotmail.com (Ulf Engstrøm)
Date: Mon, 17 Jan 2000 13:33:26 CET
Subject: MySQL
Message-ID: <20000117123326.79616.qmail@hotmail.com>

Has someone made an easy-to-install package for using with MySQL on Windows? 
All I find out there is 'then you do this, then you do that' and whatever I 
try to do it fails somewhere, seems like none of the contributing ppl have 
tried it on a Win-platform. Is there such a thing somewhere or has anyone 
built it and can tell me exxactly how to do, 'cause all my efforts've been 
in vain so far.
Thanx n brgds
Ulf Engstr?m
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com




From fdrake at acm.org  Fri Jan  7 09:50:27 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 7 Jan 2000 09:50:27 -0500 (EST)
Subject: Now What? :o)
In-Reply-To: <387541F5.1606D7C6@rubic.com>
References: <n2bd4.64$%Ph.18874529@news.randori.com>
	<387541F5.1606D7C6@rubic.com>
Message-ID: <14453.64819.631752.383886@weyr.cnri.reston.va.us>

Jeff Bauer writes:
 > 1. Use os.sep rather than slashes or backslashes
 > for your path separators.

  Even better, use the "path algebra" functions in os.path; these
result in more readable code than:

	myfile = mydirectory + os.sep + myname


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From m.faassen at vet.uu.nl  Mon Jan  3 16:30:02 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 3 Jan 2000 21:30:02 GMT
Subject: [OT] OpenSource Python Books?
References: <84g70v$tcq$1@nnrp1.deja.com> <386BC2A4.628D9C9A@callware.com> <3d7lhwrq05.fsf@amarok.cnri.reston.va.us> <14443.59621.975171.406252@cmpu.net> <2881B3B392F6F1D6.245633A232601BA7.19E9E2260D812F76@lp.airnews.net>
Message-ID: <84r4cq$gbe$1@newshost.accu.uu.nl>

Cameron Laird <claird at aftershock.neosoft.com> wrote:
> In article <14443.59621.975171.406252 at cmpu.net>,
> Kendall Clark  <kclark at ntlug.org> wrote:
> 			.
>>Also, only marginally off-topic, what is the standard view of what
>>one's 2nd language should be if one's first language is Python?
> 			.
> Dutch.

This is true -- it is a well known cosmic conspiracy that all Python
programmers speak Dutch as well. Of course I have a major head start,
being a native Dutch speaker. :)

so-you-should-all-listen-to-me ly yours,

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


From prestonlanders at my-deja.com  Thu Jan  6 18:11:55 2000
From: prestonlanders at my-deja.com (Preston Landers)
Date: Thu, 06 Jan 2000 23:11:55 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com>
Message-ID: <8537fi$jta$1@nnrp1.deja.com>

In article <14452.45436.641890.197509 at beluga.mojam.com>,
  skip at mojam.com (Skip Montanaro) wrote:

> I find nothing wrong with it.  Some people object to using try/except
for
> anything but handling errors however.

IMHO these people need to be taken out, shown a good time, maybe a
little wine and dancing.  you know, let off a little steam.

I've had at least one situation where it made perfect sense to use
try/except to handle non-error conditions.  In fact, I had defined a
bunch of exceptions relating to specific conditions, and it also made
sence to define a success condition and catch that rather than handle it
as a special case (different control structure.)

--
|| Preston Landers <prestonlanders at my-deja.com> ||


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


From aahz at netcom.com  Fri Jan 21 10:38:55 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 15:38:55 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de>
Message-ID: <869uif$63h$1@nntp9.atl.mindspring.net>

In article <m3n1q0fzg7.fsf at greebo.nodomain.de>,
Bernhard Herzog  <herzog at online.de> wrote:
>
>To reuse one of Edward's examples:
>
>  def filename(*fragments, **, root=os.getcwd()):
>      """Resolves relative paths."""
>      # Assume os.path.join joins arbitrarily many fragments.
>      return apply(os.path.join, (root,) + fragments)
>
>IMO, the root=os.getcwd() should be moved outside the argument list.
>It's assigned at definition time and can't be changed in any way when
>the function is called, so it's not an argument. Something like this
>would be better:
>
>  def filename(*fragments), root=os.getcwd():
>      """Resolves relative paths."""
>      # Assume os.path.join joins arbitrarily many fragments.
>      return apply(os.path.join, (root,) + fragments)

I think both are horrid.  Why not simply

  root = os.getcwd()
  def filename(*fragments):
      """Resolves relative paths."""
      # Assume os.path.join joins arbitrarily many fragments.
      return apply(os.path.join, (root,) + fragments)

Of course, this is broken code, anyway.  The current working directory
can change any time, and this code should therefore be

  def filename(*fragments):
      """Resolves relative paths."""
      # Assume os.path.join joins arbitrarily many fragments.
      root = os.getcwd()
      return apply(os.path.join, (root,) + fragments)
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From m.faassen at vet.uu.nl  Wed Jan 19 16:33:04 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Jan 2000 21:33:04 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com>
Message-ID: <865aig$nss$1@newshost.accu.uu.nl>

tye4 <tye4 at yahoo.com> wrote:
[snip]
> Don't get me wrong. Indentation tremendously improves readability. However,
> it should not be part of the language. That way, we don't have to put an
> immense amount of effort just trying get everything into the right block.

Argh! You discovered Python's terrible hidden weak spot! I don't know why
the Python Secret Underground hasn't 'silenced' you yet, usually we..um..
_they_ are so efficient.

The evil alien whitespace eating nanoviruses -- when the evil aliens 
release them -- will destroy Earth's Pythonic infrastructure completely. We'll
all have to revert to Perl and C. It'll be incredibly disastrous. However,
knowledge of the evil alien whitespace eating nanites is *classified*. It's
stored in our highly secret high tech 'timbot' system. I don't know how you
ever found out, unless timbot has been 'leaking' to Usenet again. We've
fixed that particular bug a thousand times before, but somehow timbot has
this odd tendency to find new ways to leak. Shouldn't have written it in Perl,
I suppose..

Guido van Rossum, under the guise of the Computer Programming For Everybody
Project, is doing DARPA funded research into protecting the Earth from this 
bizarre and evil alien threat. He and a small team of PSU-ers are putting
an immense effort into developing advanced heuristic algorithms that can 
hopefully manage to keep everything in the right block. (the timbot is
an early offspin of this project).

And you give it all away!

> There is not single programming language that forces such a bizarre
> indentation rule.

You are right -- Python, since version 1.5.1 actually moved towards the
use of _invisible_ block delimiters. You can't see them, but they are 
there, the solemn, faithful guards against the day the aliens release their
nanovirus. Whenever the 'python' executable imports an earlier Python
module without such invisible guards it will silently add them. It uses
a subtle compression technique, designed by the timbot, that ensures the byte
count of the source file remains exactly the same. JPython uses an 
optimizing 'just-in-time' hotspot compiler based on secret transmeta
technology to do the same. Viper uses ocaml as its implementation language,
which due to advanced type-inference does not have this problem at all.

And you give it all away!

> If thousands of programmers have a problem with this, why isn't it fixed.

But it is! The project is merely _secret_, for a very good reason.

And you give it all away!

And now it may be too late -- the aliens are already onto us. They'll 
design _new_ nanoviruses that eat block delimiters instead. Suddenly your
C code is unsafe! Behold the drama:

if (foo == bar) {
  printf("bar!\n");
  if (foo == baz) {
    printf("Also baz!\n");
  }
} else {
  printf("nonbar!\n");
}

an evil alien delimiter eating virus passes by and eats some delimiters,
and ooh! the semantics have changed:

if (foo == bar) 
   printf("bar!\n");
   if (foo == baz) {
     printf("Also baz!\n");
} else {
  printf("nonbar!\n")
}

Can the Earth be protected from this menace? Will the League of Dynamic C
Coders come up with a way to add the missing delimiters? They can't depend on
whitespace to do it -- the aliens have a weapon against this already.

In fact, Python with its invisible block delimiters might be the language
language to stand! You wouldn't want that to happen, do you?

Let's all continue on the PSU's path, as Guido indented.

no-I-really-can't-be-serious-about-this-ly yours,

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


From skip at mojam.com  Sun Jan  9 09:32:34 2000
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 9 Jan 2000 08:32:34 -0600 (CST)
Subject: control structures (was "Re: Sins")
In-Reply-To: <387890F6.2D1136F8@maxtal.com.au>
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
	<14452.45436.641890.197509@beluga.mojam.com>
	<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
	<14452.60226.853060.116123@beluga.mojam.com>
	<011501bf58a4$7b0db540$0100a8c0@rochester.rr.com>
	<3877513B.73B762F4@maxtal.com.au>
	<004901bf5a08$11a35be0$0100a8c0@rochester.rr.com>
	<387890F6.2D1136F8@maxtal.com.au>
Message-ID: <14456.39938.51446.999159@beluga.mojam.com>

    >> This is a common thing, does this express the idea ?
    [ ... out-of-line example snipped ... ]

    skaller> Yes, that provides the required functionality, but has a
    skaller> serious problem: cases have to be defined 'out of line', and
    skaller> therefore do not have access to the current scope.

    skaller> So I'm looking for a syntax for providing the above
    skaller> functionality _inline_.

While not strictly inline, why not define your actions as methods of a
common class, then share the state they need access to as instance
attributes?

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 rdudfield at my-deja.com  Tue Jan 18 20:10:05 2000
From: rdudfield at my-deja.com (rdudfield at my-deja.com)
Date: Wed, 19 Jan 2000 01:10:05 GMT
Subject: Best way to work in debug code?
References: <85umq4$tlr$1@nnrp1.deja.com> <slrn887f3r.viv.quinn@drachma.ugcs.caltech.edu>
Message-ID: <8632t2$42f$1@nnrp1.deja.com>

Thanks alot for your reply, this is just what I was looking for :)

In article <slrn887f3r.viv.quinn at drachma.ugcs.caltech.edu>,
  quinn at drachma.ugcs.caltech.edu (Quinn Dunkan) wrote:
> On Mon, 17 Jan 2000 09:19:08 GMT, rdudfield at my-deja.com
> <rdudfield at my-deja.com> wrote:
> >Hello,
> >
> >
> >Just wondering how you people write debug code in your functions
> >/methods?
> >
> >I personally write a "if (DEBUG_LEVEL >= 5):" around the code.
However
> >this reduces readability.
>
> I have a dprint(msg) function that writes msg to stderr if debugging
is set,
> or to a file if debug_log is set, and occaisionally use assertions.
I've
> never found it useful to get more complicated than that.  The only
time I've
> found it useful is for long-running processes that try to do
complicated
> stuff.
>
> If you put code in if __DEBUG__: (or assert expr) then python -O will
> "optimize" it away.  If you don't want to see debugging stuff in code,
perhaps
> a folding editor is the way to go (vim will do it eventually).
>
> And then, if your code needs that much debugging, perhaps simplifying,
> componentizing, and writing test harnesses is a better solution.
>

That's one of my problems.  Componentizing that is.  If there is a stuff
up in one of my parts of my program it can often flow on to other parts.

I allready have a function to test most of the more than trivial
classes, and methods.

I don't use many assertions though, I might start putting them in.


Thanks again.

Rene.


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


From tseaver at aftershock.neosoft.com  Tue Jan 11 17:14:51 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 11 Jan 2000 16:14:51 -0600
Subject: Zope and jpython
References: <857447$bc5$1@nnrp1.deja.com>
Message-ID: <3AAA8B4AEBE3AA0F.1C6AF9EE7CDCB423.FBC76EE73443400B@lp.airnews.net>

In article <857447$bc5$1 at nnrp1.deja.com>, Salva  <salvador at my-deja.com> wrote:
>Is anybody porting Zope to jpython ?
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.

No -- the answer lies in the ExtensionClass and Acquisition machinery that
underly Zope's magic -- they are fairly heavily entwined with CPython's
mechanics.

The list archives (http://classic.zope.org:12080/archives/Catalog/) will
show the discussion.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From wmcclain at salamander.com  Wed Jan  5 14:48:30 2000
From: wmcclain at salamander.com (Bill McClain)
Date: Wed, 5 Jan 2000 13:48:30 -0600
Subject: NEWS + threads
References: <0t387365ddi452en79%milek@korweta.task.gda.pl>
Message-ID: <1e3xrxz.5xo0stryd4xeN@[192.168.0.2]>

Robert Milkowski <milek at task.gda.pl> wrote:

> HI.
>       Does somebody make a small news reader in python?
> Or only threads like tin (based on subject and references)?
> Or meybe small cook-book how to make threads... step by step...

I've been working on this recently and have done reference-header
threading. Contact me via email for details. It's easy once you know
how...

My goal is to replicate the functionality of MacSOUP, an offline reader
I have used for years. I have a working program, but it is entirely TTY
based; no GUI or curses interface at all. Maybe later.

I'm running the new code on Linux.

-Bill 


From PartyG2816 at aol.com  Mon Jan  3 14:50:28 2000
From: PartyG2816 at aol.com (PartyG2816 at aol.com)
Date: Mon, 3 Jan 2000 14:50:28 EST
Subject: Considering using python for something, and need a bit of info
	:)
Message-ID: <0.48473f40.25a25784@aol.com>

And I have another small question..  After Python is embedded within my app, 
would the scripts written therein be able to modify an existing thing within 
the app? IE, you who know a bit about muds, let's say i have an object, a 
piece of wood for example. My builder wants to change the objects description 
from A long crooked stick. to A smoking, charred stick of unusual 
crookedness. after fire is applied to it. (just first thing that popped into 
my head).  Could the objects properties be modified within that script? The 
object would be an instance of a C++ class in existance within the program 
with a script attached to it.  

The embedding section of the python.org page says you would call 
Py_initialize() to initialize the interpreter, and Py_String/Py_something 
else (cant remember either of the exact names off the top of my head... :P) 
and pass the string/script file into it.

So I pass in my script, how could I let that script modify the actual object 
within the parent program though? :) Is that type of thing possible, or what? 
I don't see why there wouldnt be a way *ponders*, just trying to eliminate 
all of my concerns/questions before i spend my time looking too deeply into 
the subject of me using it in this case :)

JD



From aahz at netcom.com  Sun Jan  9 16:09:48 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jan 2000 21:09:48 GMT
Subject: Modifying Global Variables from Within a Function
References: <8EB69CD0Ejosuetompkins@209.208.0.7>
Message-ID: <85ates$vl$1@nntp6.atl.mindspring.net>

In article <8EB69CD0Ejosuetompkins at 209.208.0.7>,
Josh Tomkins <josht at iname.com> wrote:
>
>Can you modify a global variable from inside a function?  

Yup, and Emile gave you the answer for how to do that.  However, I am
going to challenge you to avoid that way of programming, particularly
given that you need to modify several variables.  Here are three good
ways to deal with the situation without using the "global" keyword:

* Use a dictionary or a list; that way you pass around a single variable
that contains modifiable elements.

* Pass around a class instance with the variables inside it.

* Move your functions into a class (making them methods) and then the
variables will be global to the class instance.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From battery841 at usa.net  Fri Jan 14 12:37:13 2000
From: battery841 at usa.net (battery841 at usa.net)
Date: Fri, 14 Jan 2000 17:37:13 GMT
Subject: DSN?
Message-ID: <85nms7$6al$1@nnrp1.deja.com>

Hey,
	I am a new Python user, so bear with my stupidity/ignorance.
	I am writing a simple log analyzer for www.thestuff.net.  It's a
Python script that runs through a _mysql layer to get to the mySQL
database.  I've been told that I need to (or at least should) use DSN
method of contacting mySQL.  How do I do this?
Thanks
Kevin


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


From moshez at math.huji.ac.il  Mon Jan 31 22:00:39 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 1 Feb 2000 05:00:39 +0200 (IST)
Subject: Namespace pollution
In-Reply-To: <949332864.553076@news.adtranzsig.de>
Message-ID: <Pine.SOL.3.96.1000201045819.23158I-100000@sundial>

On Mon, 31 Jan 2000, Dirk-Ulrich Heise wrote:

> ...
> I've got one problem with all this stuff:
> I end up with dozens of helper functions that are
> called exactly one time. They pollute the namespaces
> of the module or class they're defined in. I hate polluted
> namespaces.
> 
> How do others cope with this problem (that does not only
> apply to Python)?

Well, like the timbot said "namespaces are a honking good idea -- let's do
more of those". Use packages rather then modules, and then all of a
module's helper function stay "private". Use OO -- classes were meant
exactly to reduce namespace clutter by keeping the functions "inside" the
objects.

--
Moshe Zadka <mzadka at geocities.com>. 
INTERNET: Learn what you know.
Share what you don't.




From zakons at objectsbydesign.com  Tue Jan 11 05:57:05 2000
From: zakons at objectsbydesign.com (Stuart Zakon)
Date: Tue, 11 Jan 2000 05:57:05 -0500
Subject: Python + UML =
Message-ID: <387B0C81.D1490C97@objectsbydesign.com>

Python and the UML are two of the hottest, new  technologies for
object-oriented development.

In our latest ViewPoint, we take a look at how these two can work
together:
http://www.objectsbydesign.com/projects/python_uml.html

And an example of using UML with Python:
http://www.objectsbydesign.com/projects/python_example.html

Stuart Zakon
Objects by Design






From arcege at shore.net  Sat Jan  8 13:05:12 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 08 Jan 2000 18:05:12 GMT
Subject: calling C functions from Python
References: <852ck5$7da$1@noao.edu>
Message-ID: <s%Kd4.2786$Ef6.704898@news.shore.net>

Matt Cheselka <cheselka at tucana.noao.edu> wrote:
: Hello, 

: I'm fairly new to Python and I'm wondering if there is a way to call C
: functions from within python?  For instance (very simple) I have an object file
: that contains a function to print "hello world" to the screen.  In the python
: code I create a button to which I bind a call to this print function.  Does
: something like this exist?

Matt,

You will need to create a C extension to access C functions.  You can do
this with SWIG to let it automatically create the extension, or you can
write one yourself.

Reference docs are at:
SWIG homepage
  <URL: http://www.swig.org/
Python Documentation: Extending and Embedding
  <URL: http://www.python.org./doc/current/ext/ext.html>
Python Documentation: Python/C API
  <URL: http://www.python.org/doc/current/api/api.html>
How to write a Python Extension
  <URL: http://starship.python.net/crew/arcege/extwriting/pyext.html>

Good luck,
  -Arcege



From News at Titanic.co.uk  Sun Jan  2 14:44:42 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Sun, 02 Jan 2000 19:44:42 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com> <84nfct$bal$1@news1.xs4all.nl>
Message-ID: <umav6ssq7uubv4gpb192lp2fs1jcan6hc2@4ax.com>

Thanks for the reply...

>> with most^H^H^H^Hall database servers<<
Is this a typo or am I missing something?

I have my database working online!!!  GREAT!

Which database do you recommend?  MySQL is free (or v. cheap)

I am keen to try everything.

TIA

Simon

On 2 Jan 2000 12:13:17 GMT, boud at rempt.xs4all.nl (Boudewijn Rempt)
wrote:

>Simon Faulkner <News at titanic.co.uk> wrote:
>> Hi All,
>
>> Where can I find the instructions for the DB API?
>
>> I have managed to figure out the .connect() and .fetchone() methods
>> but am flagging with the rest.
>
>> The Python Database API Specification 2.0 document does not realy show
>> me the methods or examples that I need to get started and the MySQLdb
>> documentation is deliberatly vague to discourage use of the _mysql.
>> methods (so you use the proper API?)
>
>If you're using to MySQL, then you'd better not go to far using all
>the options and stuff from DB-API II - if you use connect, execute and
>fetchall, you've exhausted the capabilities of the database. No need
>for commit, rollback or things like that - MySQL doesn't support them.
>There's one area where MySQL has functionality beyond DB-API II, and
>that's in the area of autoincrement columns, where you can get quite
>easily get back the autoincrement values back after insertion.
>
>Also, with most^H^H^H^Hall database servers, what you really need to do is
>ask for a few rows, and fetch them immediately from the server. Keeping
>large resultsets on the server and nibbling piecemeal from them is a
>recipe for performance problems.

Simon Faulkner


From nemeth at iqsoft.hu  Sat Jan  8 05:42:25 2000
From: nemeth at iqsoft.hu (Nemeth Miklos)
Date: Sat, 08 Jan 2000 11:42:25 +0100
Subject: Python plugin
Message-ID: <38771491.BA08010@iqsoft.hu>

> Nemeth Miklos  wrote:
> > Is there a (webbrowser) plugin for executing Python scripts?
> > I'm having a lot of problem with Java, and I am seeking alternatives.
>
> What about JPython?
>
> http://www.jpython.org
>
>
Thank you for all suggestions.

(1) I am completely fed up with Java after two years of permanent
sucking (I suppose this may be the correct word).
I am threatened by all languages or tools having the letter "J" anywhere
in its name.
Castanet was also suggested as an alternative - I tried it 1.5 yeras
ago: never again anything from Marimba.

(2) No more COM or any Microsoft-only garbage. The only reason I am
looking for plugin because (apart from simple HTML) this seemed to be
supported by the browser giants. Everything else (Java applets - a
nightmare!, COM, DHTML) are browser specific.

My conclusion after this discussion thread:
(1) Grail was not mentioned as an alternative. I am a bit sad that this
project was a failure.
(2) Why on earth do I want to use plugins!? If I want to create a
sophisticated (ie. non-HTML) client, I do not need any browser, I may do
it easily with PyGTK (I tried it it seems very fast and simple to use),
PyQt, etc. communicating via HTTP, XML-RPC, CORBA (all supported by
Python) with my server. If HTML is acceptable for my application I
should use Zope with any browser.

NM






From tjreedy at udel.edu  Sat Jan 22 23:00:37 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Sat, 22 Jan 2000 23:00:37 -0500
Subject: Python Rocks!
References: <200001211334.IAA28640@python.org> <86ajlq$661@news.or.intel.com>
Message-ID: <86dues$7ln$1@news.udel.edu>

"tye4" <tye4 at yahoo.com> wrote in message
news:86ajlq$661 at news.or.intel.com...
> Yes, I think I'll let you guys wallow in your ignorance.

Two emails to the above address both bounced, at least from my mail server;
I prefer people to post with either real addresses or honestly fake
addresses
(such as fake at nowhere.com).

This has been a clever troll -- start with a title that appears to praise
and then start the burn.
The result has been a near record number of responses,  but now that whoever
has made
his/her intentions clear, let's end this and let him/her leave with his/her
'victory'.

Terry J. Reedy





From tjg at avalongroup.net  Thu Jan 20 19:08:35 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Thu, 20 Jan 2000 16:08:35 -0800
Subject: Tkinter/Pmw and focus events
Message-ID: <3887A383.5D921ADB@exceptionalminds.com>

Hi again,

Back to my other pet subject at the moment, Tkinter. I've been using Pmw
and having really good success with it, especially since I develop on
Linux and my clients run on Win98! Pmw allows one to assign the
attribute 'modifiedcommand' to many of its widgets. That function is
called any time the widget is modified. Is there a similar attribute for
got-focus and lost-focus events?

I'd like to implement something like is in Delphi/Paradox where when the
cursor is outside the field the numbers are formatted all pretty
($1,234.56), and when the cursor is inside the field it is automatically
reformatted to a 'normal' number (1234.56).


-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From tjg at avalongroup.net  Wed Jan 19 22:31:55 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Wed, 19 Jan 2000 19:31:55 -0800
Subject: Mapping Column Names to results using DBI
References: <8825686B.007D977F.00@avnet.com> <3886458C.3256ED8A@exceptionalminds.com> <20012000.5@sanctum.jae.ddns.org>
Message-ID: <388681AB.E702BE34@exceptionalminds.com>

"Juergen A. Erhard" wrote:

>     Timothy> I guess the above means that the DBI spec *doesn't* call
>     Timothy> for this behaviour. I actually find that quite
>     Timothy> surprising, [...]
> 
> <aol>Me too</aol>.
 
> That's why I prefer PyGreSQL to the DB-Sig API.  PyGreSQL does it just
> the way one would expect a language with dictionaries to behave:
> 
>    db.query("select * from table").dictresult()
> 
> returns a list of dictionaries keyed on column names.

I first started working with PyGreSQL (and really liked it, and I still
cobble together small pieces of utility code using it), but couldn't get
it to work right under Windows, plus some of my other project
requirements needed ODBC, so I went with an ODBC interface.
 
> The DB-Sig API seems, at least to me, to be doing things in a very
> traditional way... I guess that's because most of the DB people are
> used to this ugliness <0.3 wink>.

I don't know how the sig works, but I just filled out a PSA membership
form so I could allow the DB-Sig the benefit of my great storehouse of
cluelessness!

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From michelorengo at netscape.com  Tue Jan 18 11:42:53 2000
From: michelorengo at netscape.com (Michel Orengo)
Date: Tue, 18 Jan 2000 16:42:53 GMT
Subject: using threads with python and winnt
References: <1264075593-4560755@hypernet.com> <38835f4c.1380925@news.iol.ie>
Message-ID: <388498FF.D59B6846@netscape.com>

Sean Mc Grath wrote:

> >Kreiner Karl or Charly or somebody wrote:
> >
> >>             does anyone have expiriences with using threads and winnt ?
> [...]
> <gmcm at hypernet.com> wrote:
> >Threads work dandy on NT
> [...]
>
> Does anybody know how good/bad/indifferent the threading stuff works
> on Windows 95/98? I am considering using threads for an app
> under development but I'm a bit worried (read "terrified") about
> the issues that might pop up in '95 and '98.
>
> regards,
> Sean McGrath
>

I've developped a very simple application with two threads: one for the GUI
(wxPython) and one for a loop. The GUI starts the loop in another thread and
stops it by setting up an event.
So far, this works on win NT and 95/98.
But again, it is *very* simple. I must say it is very scary to work on win98.

Regards,

Michel Orengo



From daniels at mindspring.com  Thu Jan 27 21:16:47 2000
From: daniels at mindspring.com (Alan Daniels)
Date: 28 Jan 2000 02:16:47 GMT
Subject: vi and python?
References: <85uc1m$5i8$1@pegasus.tiscalinet.it> <86l7iq024mg@news2.newsguy.com>
Message-ID: <slrn891vbt.u8.daniels@localhost.localdomain>

On Tue, 25 Jan 2000 16:22:14 -0600, the infinitely wise Ben Beuchler
(insyte**NOSPAM**@emt-p.org) spoke forth to us, saying...

>Does anyone have any particular tips for python development within
>vim? Particular shortcuts to test code, autoindentation settings,
>etc...  I would particularly like to be able to use some of the idle
>features with, perhaps, gvim...

For what its worth, I'm a vim fan too, and I felt the default syntax
highlighting for Python under VIM was a bit spartan, so I wrote my
own. It colors things like the "%s"-type symbols in string, numbers,
"self" since its practically a keyword, etc. Its available at:

http://www.mindspring.com/~daniels/vim/python.vim

Hope you find it useful (if you're interested in that sort of thing,
that is).

An-endlessly-tweakable-editor-is-a-joy-forever'ly yours,
Alan.

-- 
=======================
Alan Daniels
daniels at mindspring.com
daniels at cc.gatech.edu


From jmarshall at alum.mit.edu  Mon Jan 24 12:55:28 2000
From: jmarshall at alum.mit.edu (Joe Marshall)
Date: Mon, 24 Jan 2000 12:55:28 -0500
Subject: What about REBOL (www.rebol.com)
References: <001401bf63f3$3a6eb5c0$ec2d153f@tim>
Message-ID: <mm0j4.48027$Rj5.26574@dfw-read.news.verio.net>

I don't want to add too much noise to an off topic discussion, but
I'd like to clarify something Tim Peters said.

Tim Peters <tim_one at email.msn.com> wrote in message
news:001401bf63f3$3a6eb5c0$ec2d153f at tim...
>
> One large & basic difference is that Python is primarily object-oriented,
> while REBOL is primarily function-oriented.  I count REBOL as a member of
> the lexically scoped branch of the Lisp tree (despite that it *looks*
> nothing like Lisp, a REBOL "block" is a lexical closure, and the semantics
> are virtually identical to Scheme's)

REBOL abandoned lexical closures in version 2.0, so the semantics have
become quite different from Scheme's.





From aahz at netcom.com  Sat Jan 15 12:17:32 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jan 2000 17:17:32 GMT
Subject: Does python support long longs?
References: <x8_f4.26$tX3.648@vixen.cso.uiuc.edu>
Message-ID: <85qa3c$hk$1@nntp9.atl.mindspring.net>

In article <x8_f4.26$tX3.648 at vixen.cso.uiuc.edu>,
steve s chen  <sschen at ux8.cso.uiuc.edu> wrote:
>
>1) Does Python support long long datatypes?  

Not really.  But it does have its *own* long type, which can literally
be as large as the memory of your machine will support (i.e. millions of
digits).

>2) Does Python support unsigned 32-bit ints?

Nope.

Note that to correct both of these deficiencies, you could in theory
build your own Pythonic-C extension if this is "mission critical" for
you.

>3) A less significant issue -- is there a utility that analyzes 
>	 a segment of python code verifying the dynamic name-bindings?
>	 NameError exceptions weren't being reported until run-time;
>	 does that mean we must QA each control path to ensure 'correct'
>	 execution?

Yup.  The way to build large Python applications is to componentize and
loosely-couple the hell out of everything.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From morse at harborcom.net  Tue Jan 11 13:53:51 2000
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Tue, 11 Jan 2000 18:53:51 GMT
Subject: ANN: pyFLTK_20000111_133528
Message-ID: <387b7b85.109214131@news.oh.verio.com>

pyFLTK is a wrapper for the Fast-Light Toolkit - a cross platform GUI library.
The wrapper also includes a program to convert the FLTK GUI designer output  to
Python code.

The library is available at:  http://fltk.netpedia.net/


Recent Changes:
----------------------
1/11/00 KPD
	completed install-python option

1/8/00 KPD
	fixed test/boxtype.py, test/PyAppWithGUI
	added to INSTALL file
    
1/7/00 KPD
	added flconvert support for declarations and classes
	added ifdef Python sentries to browser changes
	added demo runner test/demos.py
	fixed browser data bug

1/6/00 KPD
	added data support to browser, still needs some work to handle
    arbritary script object types

1/4/00 KPD
	updated README
	reorganized notes
	added test/PyAppWithGUI
	changed fluid parser to allow scoped indentifiers

12/29/99 KPD
	fixed Python Menu callback bug - menus weren't using the new
		CallBackStruct
	fixed MS-Windows build
	fixed function return values in src/Fl_Wrapper
	in swig/py_timeout.i: MS VCC couldn't handle pointer initializaition, 
		had to change to assignment
	removed default paramter value from contrib/Check_Browser.cc
	removed test/arc.py because it can't work with the current 
		system - no way to subclass a widget in Python yet

11/19/99 KPD
	fixed Python Fl_Widget callbacks to make them work with data





From nickluckcuck at netscapeonline.co.uk  Tue Jan 18 10:52:33 2000
From: nickluckcuck at netscapeonline.co.uk (nickluckcuck)
Date: Tue, 18 Jan 2000 15:52:33 +0000
Subject: python cgi, hosting
Message-ID: <38848C41.F358E7D5@netscapeonline.co.uk>

Does anyone know of one of the free web hosting who offers more than
perl cgi execution??

Thanks, Nick.



From pauldubois at home.com  Fri Jan 21 05:23:25 2000
From: pauldubois at home.com (Paul F. Dubois)
Date: Fri, 21 Jan 00 10:23:25 GMT
Subject: Numerical Python now at SourceForge
Message-ID: <mt2.0-27365-948450205@news.informatik.uni-bonn.de>

We are in the process of moving Numerical to SourceForge, a service provided
by VA Linux for Open Source software projects.

Here are some of the things you can do:

1. Visit the Numerical Python web page, http://numpy.sourceforge.net. A link
on that page will direct you to the Numerical Python Project pages, where
you will find downloads, read about and report bugs, and browse the CVS
source tree.

2. Using CVS, you can check out a the latest source by using the following
instructions.

cvs -d:pserver:anonymous at cvs.numpy.sourceforge.net:/cvsroot/numpy login
     When prompted for a password for anonymous, simply press the Enter key.
cvs -z3 -d:pserver:anonymous at cvs.numpy.sourceforge.net:/cvsroot/numpy co
Numerical

Later updates from within the module's directory do not need the -d
parameter.

Official releases are as shown on the project page. By using the latest
source via anonymous CVS, you are getting both the latest bug fixes and the
latest accidently-introduced bugs. Cavaet downloader.

3. Currently 3 people are enabled as developers: David Ascher, Travis
Oliphant, and myself. Once we are a little more setttled in we will be open
to adding developers.

4. Once you have the source tree you can make a compressed tar file for
moving to other machines by executing, in the Numerical directory,
    python setup.py dist
To avoid confusion, we suggest that you rename the file produced so that it
does not match the official release number. Or, you can change that number
in setup.py. This feature requires your Python have Distutils installed
(http://www.python.org/sigs/distutils-sig).

Note: if you browse the CVS tree you will see a top-level directory "numpy"
but there are no files underneath. This was the result of an early error and
when we can figure out how to destroy it, we will. The module you want is
"Numerical". The SourceForge homepage (http://sourceforge.net) will lead you
to some CVS documentation if you are unfamilar with it.

Note: all other LLNLDistribution modules are still in
LLNLDistribution11.tgz, available via xfiles.llnl.gov. Over the next month
or two we will be moving these to new homes. The xfiles machine itself has
become somewhat unstable and the server often dies on weekends. Security
changes will require us to move everything anyway, so this is a good time
for some housekeeping. In the meantime, please be patient with us.

This change does not mean that LLNL has lessened its involvement with Python
or its commitment to supporting Numerical. The intent is just to move the
project forward by opening up the development process. Fortuitously,
SourceForge has sprung full-blown from the brow of Zeus -- or maybe it just
seems that way to me.

<P><A HREF="http://numpy.sourceforge.net">Numerical Python</A> -
Numerical facility for Python, now hosted at
SourceForge.  (16-Jan-2000)

-- 
----------- 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 breiter at usf.Uni-Osnabrueck.DE  Tue Jan 11 08:44:26 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 11 Jan 2000 13:44:26 GMT
Subject: XML-RPC Server
References: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de>
Message-ID: <85fc3q$fg4$4@newsserver.rrzn.uni-hannover.de>

In article <85faal$fg4$3 at newsserver.rrzn.uni-hannover.de>,
	bernhard at intevation.de (Bernhard Reiter) writes:

> Does anybody have a simple XML-RPC server example flying around?

Of course I found a rudimentary example in the
XML-RPC for Python package(xmlrpclib). 

	Bernhard


From dworkin at ccs.neu.edu  Tue Jan 18 11:00:35 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 18 Jan 2000 11:00:35 -0500
Subject: Breaking out of walk()
In-Reply-To: "Roberts, Robert J"'s message of "Sat, 15 Jan 2000 17:01:08 -0800"
References: <38811854.F2BC197E@rl.gov>
Message-ID: <vndr9ff4bfg.fsf@betelgeuse.ccs.neu.edu>

"Roberts, Robert J" <ROBERT_J_ROBERTS at rl.gov> writes:

> Is there a way to gracefully break out of os.path.walk()?

Put the os.path.walk() in a try-except block, and throw an exception
to break out.

-Justin

 



From darcy at vex.net  Fri Jan  7 09:02:37 2000
From: darcy at vex.net (D'Arcy J.M. Cain)
Date: 7 Jan 2000 14:02:37 GMT
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <854rlt$q3q$1@news.tht.net>

Chuck Esterbrook <echuck at usit.net> wrote:
> Seems like many web host providers, such as MindSpring, offer CGI with
> Perl, but not other languages, notably Python.

> Does anyone know who would host a CGI Python site?

Vex.Net does.  See http://www.vex.net/ in particular the tech page.

-- 
D'Arcy J.M. Cain <darcy at vex.net>           |  Democracy is three wolves
http://www.vex.net/                        |  and a sheep voting on         
+1 416 425 1212     (DoD#0082)    (eNTP)   |  what's for dinner.


From dgoodger at bigfoot.com  Wed Jan 19 13:18:13 2000
From: dgoodger at bigfoot.com (David Goodger)
Date: Wed, 19 Jan 2000 13:18:13 -0500
Subject: Looking for open source hosting site
Message-ID: <B4AB6A14.1FB8%dgoodger@bigfoot.com>

I was surfing several weeks ago and came across a site that offered free
hosting of open source projects. There was an application form, in which you
detail your project, and (if I recall correctly) it would assign you a
subdomain (i.e., if the site was www.site.org, your project would be
www.project.site.org) and give you access.

Unfortunately, I didn't bookmark the site, and can't find it now. Does
anybody know the site I'm talking about? Please e-mail, and I'll summarize.

Thanks in advance!

-- 
David Goodger
dgoodger at bigfoot.com



From nolan_d at bigfoot.com  Mon Jan  3 01:34:31 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Mon, 03 Jan 2000 00:34:31 -0600
Subject: More grammar issues
Message-ID: <E12514Z-0005qV-00@ethereal>

All right. I've begun translating a sizable subset of the standard
Python grammar into Yacc. I'm building a parse tree with the output,
but I'm again confused. Most rules have clearly-defined outputs. These
near the end confuse me, though:

test: and_test ('or' and_test)* | lambdef
and_test: not_test ('and' not_test)*
...

To simplify matters, I've chosen not to implement certain definitions
in my first prototype for simplicity, so for the above example I'm
eliminating lambdef: once I get a working parser and compiler then
I'll begin implementing the language features which I've left
unwritten at the moment.

I'm confused about what types of expressions might fit these
definitions, and how one would go about Yacc-ifying them. It was
easier for me to handle most other definitions since I could visualize
what they might look like, but these have me stumped. :) Thanks in
advance for any help.



From b2blink at hotmail.com  Tue Jan 25 10:31:39 2000
From: b2blink at hotmail.com (Ulf Engstrøm)
Date: Tue, 25 Jan 2000 10:31:39 CET
Subject: tabs do WHAT?
Message-ID: <20000125093139.91268.qmail@hotmail.com>

>BTW, people here seem to have adopted the idea that if you don't like
>a particular feature of a language, you should abandon it(e.g., "You don't 
>like indentation? Use Perl!"). But many languages have evolved exactly this 
>way: users criticize, complain, praize and whine, >thereby influencing the 
>shape of the languge in question. You can > >find a very nice description 
>of this process with respect to C++ >in "The Evolution of C++ : Language 
>Design in the Marketplace of >Ideas", by Jim Waldo (Editor), James Waldo 
>(Editor).

OKi, but I think the difference is that the tabbing is a main part of the 
Python language, one of the things that makes it different.
I don't like the {}, but I won't try to change C/C++ or any other languages, 
instead I use the language I find best and most useful. I like Python mainly 
because of it's simplicity while it at the same time is so powerful, and I 
really really like the intendation.

I got into a project that's been developed in Python during 2 years before I 
ever saw it. There were over 100'000 lines of code, and talking out of 
experience with other languages, I know that I couldn't have gotten into the 
that project that easily in any other language. Reading all this code from 
many other developers was made a lot easier with Python and a big reason was 
that the intendation is always there.

Best regards
Ulf Engstr?m

(mail)  ulf.engstrom at b2b-link.com
(phone) +47 6676 5863
(fax)   +47 6676 5861
(web)   http://www.b2b-link.com

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com




From kc5tja at garnet.armored.net  Mon Jan 24 18:55:40 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 24 Jan 2000 23:55:40 GMT
Subject: tabs do WHAT?
References: <000201bf6640$d154d100$022d153f@tim>
Message-ID: <slrn895h29.rar.kc5tja@garnet.armored.net>

In article <000201bf6640$d154d100$022d153f at tim>, Tim Peters wrote:
>    and user testing.  You'd almost think programming languages
>    were an interface between people and computers.

That's because they ARE an interface between people and computers.  That's
the whole premise behind LISP and Forth, and I dare say the concept works
quite well.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From mgushee at havenrock.com  Tue Jan 18 19:49:24 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 18 Jan 2000 19:49:24 -0500
Subject: **kwargs
References: <862j09$o1k$1@nnrp1.deja.com>
Message-ID: <m3snzukhrf.fsf@havenrock.com>

sp00fD <sp00fD at yahoo.com> writes:

> Does **kwargs represent a dictionary? If so, can I have a method like
> such
> 
> def foo(self, foo=None, bar=None, **kwargs)
> 
> and then do something like
> 
> this_dict = {}
> ..insert stuff into this_dict..
> 
> my.foo(foo="something", bar="something else", this_dict)

Ah, well, you've *sort of* got the right idea here.

> ? How would I do something like this ?

No need to. **kwargs is magic! You can do this:

    my.foo(foo="something", bar="something else", food="spam",
	   thug="Dinsdale")

... and the keyword arguments 'food' and 'thug' are automatically
folded into a dictionary, so that *within* the my.foo() method:

    kwargs['food']  =>  'spam'
    kwargs['thug']  =>  'Dinsdale'
    kwargs          =>  {'food': 'spam', 'thug': 'Dinsdale'}

Note that the name 'kwargs' has no meaning for the Python interpreter
-- it's just a convention. The double asterisk is what works the
magic.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From elf at halcyon.com  Tue Jan  4 02:01:59 2000
From: elf at halcyon.com (Elf Sternberg)
Date: 4 Jan 2000 07:01:59 GMT
Subject: Re module help?
Message-ID: <84s5t7$ive$1@brokaw.wa.com>

There is a line in 'xmllib.py' that has me puzzled.  I thought I
understood regular expressions and XML, but this one has me confused.

interesting = re.compile('[]&<]')

I don't get it.  What is this looking for?  The '[]' can't mean an empty
set, can it?  That doesn't make any sense.  Is this a special case where
'[]...' allows you to include the ']' in things to search for without
escaping it?

Or is it, in fact, a bug?

		Elf

--
Elf M. Sternberg, rational romantic mystic cynical idealist
     MST3K - Help save the saving grace of television.
A.A 1493                        http://www.halcyon.com/elf/


From rickhigh at aol.com  Sun Jan 16 17:00:07 2000
From: rickhigh at aol.com (RICKHIGH)
Date: 16 Jan 2000 22:00:07 GMT
Subject: JPython getting a good IDE! NetBeans (Forte CE) to support JPython!
Message-ID: <20000116170007.25196.00001095@ng-fg1.aol.com>

NetBeans (Forte CE) to support compiling JPython code in IDE. 

NetBeans said the above at the Java Live dicussion regarding "Forte for Java
Community Edition." at JavaTM Live during the Fort? for Java Community Edition
January 11, 2000.

The speakers Christine Bodo (Christine Bodo), Ian Formanek (Ian Formanek).

http://developer.java.sun.com/developer/community/chat/JavaLive/2000/jl011
2-print.html

Partial Transcript:

Rick Hightower (Question): 
 "A few weeks back you had a poll on your NetBeans website that spoke about
adding support for JPython. 

   What kind of support do you plan to add for JPython? 
   (Do you plan) scripting for NetBeans IDE? (GOOD) 
   (Or) An extension that compiles and runs JPython? (BETTER) 
   (JPython is a free language that compiles and runs in the JavaTM Virtual
Machine1(JVM)"

Ian Formanek (Answer): 

"In response to Rick: With this poll we were evaluating what scripting
languages people use most. The addition of scripting to the IDE is planned as
an extension, which would address *both* of the usages you mentioned. Actually,
once you do one, the second is for free. "
(End Transcript)

Is that Cool Beans?

Check out the full transcript:
http://developer.java.sun.com/developer/community/chat/JavaLive/2000/jl011
2-print.html

Additional 

Rick Hightower: 

"Excellent, I am glad to hear that you are adding JPython support for Fort? CE.
I know I will use it. I think JPython is great! Having an IDE that supports it
and allows compiling and integrating is great (idea). Do have plans to support
any other languages that compile or run in the JVM? Rhino (JavaScript)?, Bistro
(Smalltalk)?, Skij (Scheme)?, etc." 
"Also there are versions of JSP that work with Python (JPython) and JavaScript.
Will your JSP support other JSP languages?"

Ian Formanek: 
"Regarding Rick's question about other languages: there are currently no plans
to provide such support in the basic version, but one of the nice effects of
the Open APIs is that any 3rd party, being it a company or a Linux opensource
team or a single developer, can very easily write new add-on modules to the
IDE. And the stuff you mentioned is very good example of things that may be
supported by other parties."

End Transcript

BTW For those who do not know Netbeans is a Java IDE maker that was bought out
by Sun. Sun is offering a free IDE via NetBeans called the Forte Community
Edition (CE) IDE (or something like that). They had a poll a few weeks ago
about scripting languages and JPython won by a landslide. To find out more
about netbeans and Forte CE go to http://www.netbeans.com/.

Does anyone know when NetBeans will add this support?

--Rick Hightower


From flight at mathi.uni-heidelberg.de  Fri Jan  7 06:47:54 2000
From: flight at mathi.uni-heidelberg.de (Gregor Hoffleit)
Date: Fri, 7 Jan 2000 12:47:54 +0100
Subject: mod python?
In-Reply-To: <Pine.WNT.4.21.0001070938001.197-100000@wks_3.e-shopwork.com>
References: <Pine.WNT.4.21.0001070938001.197-100000@wks_3.e-shopwork.com>
Message-ID: <20000107124754.A19678@mathi.uni-heidelberg.de>

On Fri, Jan 07, 2000 at 09:43:49AM +0000, Sposhua wrote:
> b) Is there an equivalent for Python?

Yep, there is. The equivalent to mod_perl is known as PyApache
(http://www.msg.com.mx/pyapache/).

> c) What the heck is an Apache module anyway? (in practical terms)

cf. www.apache.org. A module is a module. My definition of an Apache module:
Something that enhances the Apache server but doesn't belong into the Apache
core, i.e. a plugin for Apache. As you wrote, modules like PyApache and
mod_perl embed a Perl/Python interpreter into the Apache server so that the
server doesn't need to fork a separate process to handle Perl/Python CGI
scripts, i.e. you don't have to pay the fork tax.


There are alternatives to PyApache like PCGI (Persistent CGI,
http://starship.python.net/crew/jbauer/persistcgi/) and a corresponding
mod_pcgi (http://www.zope.org/Members/phd/mod_pcgi2).

    Gregor



From amused at webamused.com  Mon Jan 31 18:51:13 2000
From: amused at webamused.com (Joshua Macy)
Date: Mon, 31 Jan 2000 23:51:13 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com> <86np80$jhp$1@harvee.billerica.ma.us> <86ntlg$2tc$1@news.ycc.yale.edu> <86pi88$n3n$1@harvee.billerica.ma.us> <38918EE7.38A43687@electricorb.com> <874hhv$dnf$1@harvee.billerica.ma.us>
Message-ID: <38961DEA.6CBE3EC1@webamused.com>

"Eric S. Johansson" wrote:
> 
> Glen Starchman <glen at electricorb.com> wrote:
> > class test:
> > indent
> > def test_function():
> > indent
> > print "foo"
> > dedent
> > dedent
> 
> which is spoken:
> "class test colon new-line indent delta echo foxtrot test no-space underscore
> nospace function nospace left- paren right-paren indent new-line print open-quote
> foxtrot oscar oscar new-line dedent new-line dedent"
> 
> now, if you really want to get ugly go change the function name from test_function
> to test_subroutine when the only navigation capability you have is by counting
> lines and characters
> 
> believe me, it gets old real fast...
> 
> --
> Eric S. Johansson    esj at inguide.com    esj at harvee.billerica.ma.us
> This message was composed almost entirely using NaturallySpeaking

  You forgot a "colon" and a "close-quote", which just proves your
point.

Joshua


From shang.remove_edu at st.jyu.fi.edu  Thu Jan 27 05:46:43 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Thu, 27 Jan 2000 12:46:43 +0200
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com> <388F140B.C0AA1C99@erols.com> <388f88ca.1925884@news.erols.com>
Message-ID: <86p7nt$m08$1@mordred.cc.jyu.fi>

Roey Katz <katz at glue.umd.edu> wrote in message
news:388f88ca.1925884 at news.erols.com...
> The 'virtual classes' that Sweeney describes look to me an awful lot
> like something like this:
>
>
> class myClass:
>     def __init__( self ): pass
>     def doSomething( self ): print 'instantiating'
>
> class myClass2:
>     def __init__( self ): pass
>     def doSomething( self ): print 'instantiating (2)'
>
> class Base:
>
>     virtualClass = None
>
>     def __init__( self ):
>          self.inst = self.virtualClass()
>          self.inst.doSomething()
>
>
> class Derived1( Base ):
>     virtualClass = myClass
>
> class Derived2( Base ):
>     virtualClass = myClass2

I was thinking about something along these lines:

class SomeFramework:
    class Class1:
        def some_method(self):
            print "original class1"
    class Class2:
        pass
    def __init__(self):
        self.c1 = self.Class1()
        self.c2 = self.Class2()

class DerivedFramework (SomeFramework):
    class Class1 (SomeFramework.Class1):
        def some_method(self):
            print "derived class1"

Now DerivedFramework can inherit SomeFramework and modify Class1 how
ever it wants.

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi

Get paid to surf the Web!
http://www.alladvantage.com/home.asp?refid=BOT021




From gerrit.holl at pobox.com  Fri Jan 21 11:04:32 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 17:04:32 +0100
Subject: SUGGEST: os.getuser()
In-Reply-To: <Jylh4.3912$Ef6.1036319@news.shore.net>; from arcege@shore.net on Wed, Jan 19, 2000 at 04:24:09PM +0000
References: <20000116174004.A5995@stopcontact.palga.uucp> <000b01bf6049$36532140$f29b12c2@secret.pythonware.com> <20000116195710.A7700@stopcontact.palga.uucp> <Jylh4.3912$Ef6.1036319@news.shore.net>
Message-ID: <20000121170432.B843@stopcontact.palga.uucp>

Michael P. Reilly wrote on 948295449:
> : try:
> :     import pwd
> :     return pwd.getpwuid(os.getuid())[0]
> : except ImportError:
> :     import os
> :     for name in ('LOGNAME', 'USER', 'LNAME', 'USERNAME'):
> :         user = os.environ.get(name)
> :         if user:
> :             return user
> :     return None
> 
> And this can mislead the program.  getpwuid() usually only stops at the
> first entry matching the given userid.  There are a lot of systems
> which have multiple accounts with the same userid, for legitimate
> reasons.  If the user has changed the environment variable, there is
> usually a reason.  If your program is so paranoid, then you probably
> want your own mechanism.

So getpwuid() is never safe? But in my program, I *also* need the
comment field. So if there are multiple uids the same, I can't get
the full name?

> Not to mention that in some ways, it is far easier to "fake" out
> accounts on a MS box than it is on a UNIX box.  And then there are Mac
> "accounts".

Well, it isn't that bad; my script runs on a Linux box, on a site about
Linux. The chance it will be a MS box in the future should be small <g>.

a-sysop-letting-multiple-identical-uids-be-in-/etc/passwd-
    -is-very-paranoid-anyway-ly yours,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From ivanlan at callware.com  Tue Jan  4 14:03:54 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 04 Jan 2000 12:03:54 -0700
Subject: Grail
References: <387237F3.C35B7FA1@callware.com> <14450.17021.111722.710187@weyr.cnri.reston.va.us>
Message-ID: <3872441A.7E86CCED@callware.com>

Hi Fred--

"Fred L. Drake, Jr." wrote:
> 
> Ivan Van Laningham writes:
>  > Any clues?  I'm running on WinNT 4.0, SP5.  160MB ram, PII 400Mhz.
> 
>   That's a bug.  If you want to fund Grail development, I'll see if I
> can come up with more clues.  ;)
> 

There's a bug in grail, or it's a bug that I'm running NT?-)

I can fund Grail development in the low two figures. ...

>  > If this can't be made to run, are there any HTML widgets out there?
> 
>   Doesn't wxWindows include one?  Or are you specifically interested
> in Tk?
> 

Um, yeah, I was hoping to get a Tk HTML widget, since the entire rest of
the application is already written in Tk.  An HTML widget would let me
use the existing HTML file for help without having to fire up netscape
or IE. ...

<gotchas-to-the-left-of-me-gotchas-to-the-right-of-me>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From gmcm at hypernet.com  Tue Jan  4 21:58:08 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 4 Jan 2000 21:58:08 -0500
Subject: how to copy a dictionary
In-Reply-To: <roy-5C6597.21015404012000@netnews.nyu.edu>
Message-ID: <1265082303-1434028@hypernet.com>

dictionaries have a copy() method.

Roy Smith asks:
> If I do:
> 
> d = {'hedgehog': 'spiney norman'}
> temp = d
> temp['scotsman'] = 'earnest potgorney'
> 
> I end up changing the original dictionary, d.  It's obvious that
> what's going on is when I do temp = d I get a pointer to the same
> object instead of a new object.  My question is, how do I force a
> new object to be created, so when  modify it, I don't also modify
> the original? --
> http://www.python.org/mailman/listinfo/python-list



- Gordon



From jam at quark.emich.edu  Mon Jan 10 13:33:38 2000
From: jam at quark.emich.edu (Jeff)
Date: Mon, 10 Jan 2000 13:33:38 -0500
Subject: starting a cgi with params
In-Reply-To: <iQp6OHyH9j2y64kYNrpGPcKoXKu0@4ax.com>; from anders.eriksson@morateknikutveckling.se on Mon, Jan 10, 2000 at 05:41:04PM +0100
References: <iQp6OHyH9j2y64kYNrpGPcKoXKu0@4ax.com>
Message-ID: <20000110133338.A15914@quark.emich.edu>

On Mon, Jan 10, 2000 at 05:41:04PM +0100, Anders M Eriksson wrote:
> Hello!
> 
> I have created a Python script which takes 2 params, a filename and an
> operation code. The script works well when I run it from the command
> prompt. My question is:
> 
> How do I send 2 params to the script when I'm using it as an CGI
> script?
> 
> At the moment I'm using sys.argv[1] and sys.argv[2] to get the params
> but it doesn't work when I call it from a web page.
> 
> // Anders
> 

greetings,

I would suggest looking into the 'cgi' module, which comes with the standard
distribution. there are also a number of good websites that explain how cgi
works-- the 'protocol' is generic regardless of platform and server
platform. 

by the way, what platform are you using? linux + apache?

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From tim_one at email.msn.com  Thu Jan  6 20:19:09 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 6 Jan 2000 20:19:09 -0500
Subject: Naming conventions
In-Reply-To: <3874EBF7.8317091D@ndh.net>
Message-ID: <000801bf58ad$32db90e0$932d153f@tim>

[Stefan Schwarzer]
> are there common naming conventions for Python (functions,
> classes, variables ...)?

You can read about what Python's designer likes best at

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

disagreement-is-allowed-but-severely-punished<wink>-ly y'rs  - tim





From breiter at usf.Uni-Osnabrueck.DE  Thu Jan 13 10:15:41 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 13 Jan 2000 15:15:41 GMT
Subject: Python and MapObjects (Was: Python and ArcView GIS)
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>
    <Pine.SGI.3.95.1000112172541.19328B-100000@clavey.des.ucdavis.edu>
    <85kci6$i9l$1@newsie2.cent.net>
Message-ID: <85kq6t$4a4$1@newsserver.rrzn.uni-hannover.de>

In article <85kci6$i9l$1 at newsie2.cent.net>,
	"Norman Vine" <nhv at cape.com> writes:

> but there is the OpenSource mapserver project
> that works with shapefiles and has a SWIGed interface

> http://mapserver.gis.umn.edu/

Mapserver still has to clarify its license (though they are aiming
towards BSD style, AFAIK.)

Other free software GIS can be found at the FreeGIS pages:
	http://freegis.org

Bernhard

-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From tim_one at email.msn.com  Thu Jan 27 03:34:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 27 Jan 2000 03:34:37 -0500
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
In-Reply-To: <m12DjXD-000CxiC@artcom0.artcom-gmbh.de>
Message-ID: <000001bf68a1$58e21c60$932d153f@tim>

[Peter Funk]
> Look at the following coding alternatives often seen in generic
> python code:
>
> class A:
>     def load(self):
>         # ... do something e.g. load a file
>         try:
>             self.update()
>         except AttributeError:
>             pass
>
> class B:
>     def load(self):
>         # ... do something e.g. load a file
>         if hasattr(self, 'update'):
>             self.update()
>
> Which coding idiom is better?  The second form is one line
> shorter than the first, but it has the disadvantage, that
> the method (or attribute) must be named twice. ...

I prefer the second form.  The first form is risky, because the try/except
can hide more than the author intended (that is, if self.update does exist,
any AttributeError raised during its execution will also get ignored).  I
don't care at all about the measly line or two, or duplicating a name (an
editor with word completion renders that trivial); but *do* care that the
*intent* of the first form is easier to see at a glance.  Alas, it doesn't
really mean what it appears to say.

exceptions-are-exceptionally-exceptional-ly y'ts  - tim





From skip at mojam.com  Mon Jan 17 15:22:19 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 17 Jan 2000 14:22:19 -0600 (CST)
Subject: regular expressions
In-Reply-To: <20000117195631.A2020@gmx.net>
References: <20000117195631.A2020@gmx.net>
Message-ID: <14467.31227.790827.717384@beluga.mojam.com>

    >> Has anyone experience with regular expressions in python?

Yeah, I'm sure there are a couple people... ;-)

    >> Or is there a a howto?

How about the re module documentation:

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

or AMK's regular expression howto:

    http://www.python.org/doc/howto/regex/regex.html

?

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From JoeSmith at bogusaddress.com  Wed Jan 19 04:58:00 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Wed, 19 Jan 2000 09:58:00 GMT
Subject: Using tabs effectively. Was: Why not use tabs?
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <20000118.19182100@robert.polytopic.com> <38854A89.B3FBFA12@Lugoj.Com>
Message-ID: <38858B76.AC22E62B@bogusaddress.com>

If one is using all tabs, then putting a comment at the top of the file
indicating the tab setting seems like a waste.  If one is mixing tabs and
spaces, then it is needed.  Maybe the comment should be one of the following:
using all tabs, using all spaces or mixing with a tab setting of X.

It seems to me by using all tabs, one can change the indentation to suit their
needs.  For example, if I like two spaces, then setting the tab setting to two
gives me my preferred view of the code.  If you like 5, then you should be able
to set it to 5 and get your preferred view.  I think that mixing tabs and spaces
will normally cause problems.

The best solution is for your editor to know your preference and reformat the
text to your preference when it reads the file in (and it could write it back
out in the original format).  This would probably be difficult with python
because of the lack of an end of block indicator.  Since tools such as editors
are supposed to make our lives easier, you think that this would have been done
by now.


James Logajan wrote:

> Robert Meegan wrote:
> > Absolutely nothing is wrong with tabs, as long as you only use tabs.
> > The problem comes when people mix tabs and spaces, because it would
> > seem that every programmer on Earth uses a different default tab width
> > (I'm lucky, I got in when three spaces per tab was still available).
> >
> > Some people use eight-space tabs and type four spaces for an
> > indentation level so that their code has indentation:
> >
> > <-->
> > <- \t ->
> > <- \t -><-->
> > <- \t -><- \t ->
> >
> > etc.
> >
> > When you import this into your editor (set to three-space tabs, like
> > God intended) the effect is horrific.
>
> Under Solaris and Linux (and probably most other Unices) there exist the
> commands:
>
> expand
> unexpand
>
> Read their man pages. Note that they take an option about how wide tabs are
> to be interpreted. Also note that some versions (BSD?) of the "pr" command
> also have the "-e" option for setting tab width.
>
> The issue of "how many spaces to a tab" is an old enough problem *in all
> languages* that I'm surprised that it has never become a convention to put a
> comment at the top of all source files indicating what the tab settings for
> decent viewing should be. I've come across a lot of source code where tabs
> were intended to be every 4 characters to conclude this is a sizable number
> of authors. (I used to use 3 spaces/tab early in my career.) The fact that
> some languages like Python are sensitive to the issue detracts from them not
> a wit in my opinion.
>
> I humbly offer the following rules to a long and healthy editing experience:
>
> 1.0) If using tabs (regardless of tab setting):
>    1.1) Use nothing but tabs for indentation!
>         Do not mix tab and space indentation!
>    1.2) Indent at whole multiples of your tab settings lest you make it
>         impossible to conform to rule 1.1.
>    1.3) Place a comment at the top of your file indicating the tab
>         spacing. Even if using tab settings at 8.
>    1.4) Run the unexpand utility (if on Unix) on the file to minimize
>         the chance of accidentally violating rule 1.1.
> 2.0) If using spaces:
>    2.1) Use nothing but spaces for indentation!
>         Do not mix space and tab indentation!
>    2.2) Place a comment at the top of your file indicating that tabs
>         do not exist as indentation and should not exist.
>    2.3) Run the expand utility (if on Unix) on the file to insure
>         that you have not accidentally violated rule 2.1 and 2.2.
> 3.0) If trying to impose a particular tab setting as standard:
>    3.1) Try using a thimble to turn back the tide, as you will have better
>         luck at success at that than in imposing a specific tab spacing
>         as "correct".
>
> Rules to follow for extra credit:
> 4.0) For lines longer than 80 characters:
>    4.1) Keep all lines under 80 characters and manually insert line
>         breaks at reasonable points. Do not rely on automatic editor
>         line wraps.
>    4.2) If reasonably possible, (and it normally is) indent line
>         continuations to at least the same indentation level as the
>         continued line. (Much like the sentences in these paragraphs
>         have been "continued" at the same indentation level.)
>
> 5.) Don't expect any of these rules to be followed except this one.
>
> P.S. I sure hope Netscape Communicator didn't convert any of my spaces to
> tabs! ;)



From m.faassen at vet.uu.nl  Thu Jan 20 08:18:03 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jan 2000 13:18:03 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com> <slrn88cnb6.4i6.talon@coursen.ifu.net> <oqzou1v5rj.fsf@titan.progiciels-bpi.ca> <388673D6.192A1907@roguewave.com>
Message-ID: <8671ub$77k$2@newshost.accu.uu.nl>

Bjorn Pettersen <bjorn at roguewave.com> wrote:
>> P.S. - I begin to think he is trolling us, and rather successfully! :-)
>> OK, enough said.  Let's switch to something more productive...

> but, but, but,... I haven't posted to this thread yet.... ;-)

Argh, and I did. I'm being trollified! Argh. :)

Someone get the Python Secret Underground to do something about him. :)

Regards,

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


From emile at fenx.com  Tue Jan 18 09:06:16 2000
From: emile at fenx.com (Emile van Sebille)
Date: Tue, 18 Jan 2000 06:06:16 -0800
Subject: Interacting with external program -pty?
References: <3DB0284E2B7AD21189A400805F19DB280129E238@se-hsl-mail010.hassle.se.astra.com>
Message-ID: <00cc01bf61bd$3302d4c0$01ffffc0@worldnet.att.net>

Niklas,

Look for expect.py on the net and at telnetlib.py in
the standard distribution.

HTH

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: <Niklas.Blomberg at hassle.se.astra.com>
To: <python-list at python.org>
Sent: Tuesday, January 18, 2000 4:17 AM
Subject: Interacting with external program -pty?


> Dear all,
>
> I would appreciate some advice on how to script an external
interactive
> process(program) using python. In addition to sending commands (which
is
> straightforward) I need to be able to capture the output (which can be
> extensive) for further downstream processing. Since the program
expects a
> terminal pipes don't work - the output is buffered - I have started to
look
> at the pty module. This is uncharted territory for me and any advice
both on
> possible solutions and perhaps availiable example code would be most
> appreciated.
>
> I'll be happy to post a summary.
>
> Thanks,
> Niklas
>
> -----
> Niklas Blomberg
> AstraZeneca Structural Chemistry Laboratory
> niklas.blomberg at astrazeneca.com
>
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>





From wjk at wjk.mv.com  Thu Jan 27 23:19:39 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Thu, 27 Jan 2000 23:19:39 -0500
Subject: Charts
References: <3890C74D.786411BC@acm.org> <87n1pr6m0i.fsf@iname.com>
Message-ID: <389118DB.4B7997AE@wjk.mv.com>

Hi:
Am not sure of packages; however, as an option you can create charts using
Python
to do calculations etc. then pipe into togif or tojpg on UNIX -- thats one
possibilty..
Other would be to use Tkinter and create a canvas ( have to do math as well)
and another would be to use Tcl/TK and a create a canvas.. you would have
to have plug-in for browser... And you could also do it with Java AWT and
Swing (possibly using JPython?)... Each has it own quirks (ie. Whether you
draw rectangles from the top or the bottom)... As a crazy newbie I like to
mix
and match and glue stuff together...... So maybe the Tkinter and JPython
people
might know of something more, but there are ways.... be they evil or not....
Hopefully you will find some package that works....



David Maslen wrote:

> Thierry Thelliez <thierry at acm.org> writes:
> > Is there any packages for creating charts (histograms, scattered plots,
> > line charts,...) in Python ?
>
> Yes. There seem to be a few. Do a search on
> http://www.vex.net/parnassus/
>
> Just to quickly summarise what I discovered.
>
> Gnuplot, stable, mature and the python module to access is quite good,
> but not as feature packed as it might be. The best thing I found.
>
> gdcharts. Very pretty. I'd consider it for generating nice graphs for
> a webpage or perhaps a report, but not if I really wanted to look at
> the lines in detail.
>
> graphite. Based on the piddle library. I think this will be very nice,
> but it didn't do the financial type charts I was after.
>
> gist stuff in python_graphics. Interesting, and interactive, not quite
> what I was after.



From darrell at dorb.com  Sat Jan  1 14:59:13 2000
From: darrell at dorb.com (Darrell)
Date: Sat, 1 Jan 2000 14:59:13 -0500
Subject: pickle problem:cant reload object saved with binary option
References: <386e1de1.998928692@news-server> <YHpb4.914$Sr1.104202@typhoon.nyroc.rr.com> <386e31c2.1004019140@news-server> <386E4B0E.C9DD0BF@appliedbiometrics.com>
Message-ID: <003001bf5492$ada0faf0$0100a8c0@rochester.rr.com>

Out of curiosity I looked into this file mode thing. Sure beats cleaning out
the garage !

Looks like the '+' allows read/write.
>>> open('t','w').write('now is the time')
>>> open('t','r+').write('xx')
>>> print open('t','r').read()
xxw is the time
>>>
For append you have to use 'a'

/////////////////////////////////
gcc\libio\fileops.c function _IO_file_fopen
  switch (*mode++)
    {
    case 'r':
      omode = O_RDONLY;
      read_write = _IO_NO_WRITES;
    // 0x08
      break;
    case 'w':
      omode = O_WRONLY;
      oflags = O_CREAT|O_TRUNC;
      read_write = _IO_NO_READS;
    // 0x04
      break;
    case 'a':
      omode = O_WRONLY;
      oflags = O_CREAT|O_APPEND;
      read_write = _IO_NO_READS|_IO_IS_APPENDING;
    // 0x1004
      break;
    }
  if (mode[0] == '+' || (mode[0] == 'b' && mode[1] == '+'))
    {
      omode = O_RDWR;
      read_write &= _IO_IS_APPENDING;
    // Allows reads and writes.
    // Keeps APPENDING if it was already given.
    // Which is only the 'a'

    }
  fdesc = open (filename, omode|oflags, oprot);

///////////////////////////

#define _IO_IS_APPENDING 0x1000
#define _IO_NO_READS 4 /* Reading not allowed */
#define _IO_NO_WRITES 8 /* Writing not allowd */


Charlene is tired of lame excuses so garage here I come.
--Darrell

----- Original Message -----
From: "Christian Tismer" <tismer at appliedbiometrics.com>
To: "Marshall" <python at scoobysnacks.com>
Cc: <python-list at python.org>
Sent: Saturday, January 01, 2000 1:44 PM
Subject: Re: pickle problem:cant reload object saved with binary option
>
> No-no-no! :-)
> Please understand the following:
> Whenever you use the binary (c)Pickle format under Windows,
> you *always* must use a binary mode for your file, *both*
> for reading and writing.
>





From matthewm at es.co.nz  Tue Jan  4 23:33:48 2000
From: matthewm at es.co.nz (Matthew Miller)
Date: Wed, 05 Jan 2000 17:33:48 +1300
Subject: popen and spawnv: losing the battle with Win98.
Message-ID: <3872C9AC.C4414F27@es.co.nz>

Hi Everyone,

Problem#1) I've tried and tried but I cannot get
os.popen('command','w').write('data') under Win98 to accept data from
the pipe - it justs ignores it. HELP!

Problem#2) Likewise, I cannot get spawnv() to work under Win98. Could
someone give me a simple example.

Thanks alot. matthew.



From bparsia at email.unc.edu  Tue Jan 18 23:25:41 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Tue, 18 Jan 2000 23:25:41 -0500
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com>
Message-ID: <1e4mlks.5sj58a1jzyjhcN%bparsia@email.unc.edu>

tye4 <tye4 at yahoo.com> wrote:

[snip]
> 
> How hard can it be to add 'end' indicators to the language.. If these end
> indicators are added, the language would still be backward compatible. I
> can't believe nobody has fixed this for so long.

"Fixed" suggests that it's a bug or a design flaw. It's clearly not a
bug, and equally clearly not a straightforward, obvious design *flaw*.
It is a design decision, about which reasonable people disagree, and
wackos go on an on about ;)

Plus, if you're worried about indendation, it's odd that you focused on
"end"s. There are no "begins" either. Blocks are marked by indentation
level, which is a standard convention in most language style guides.
Python merely enforces this, and then drops the "superfluous"
delimiters.

(def, class, etc. all lack "end"s too, remember.)

Since there are no lack of languages with block level delimiters, why
not consider Python as an experiment? It probably *will* introduce
problems you're used to not having (editor choice, tab setting, etc.),
but perhaps it will eliminate problems you are used *to* having.

If so, it's a simple design decision--a trade off. And it may come down
the the emphasis different people place on certain difficulties. I
suggest that you develop a methodology for evaluting the troubles
indendation brings vs. the trouble block delimiters bring, and then
apply that methodology first to yourself. That can be the preliminary
result. After the several months of study, present the result to the
newsgroup. I'm sure it will garner much interest and people will be
eager to apply your methodology to more subjects. In two years or so, we
would have the definite evidence, all things going well, to make a fully
informed decision. Just in time for Python 2.

You might start reading back in Deja news, not just for preliminary data
(a simple...er..."tabulation" of pro and con indentation voices would be
an interesting start), but for literature references (Knuth, in
particular).

Cheers,
Bijan Parsia.


From moonseeker at my-deja.com  Mon Jan  3 05:29:56 2000
From: moonseeker at my-deja.com (moonseeker at my-deja.com)
Date: Mon, 03 Jan 2000 10:29:56 GMT
Subject: not pickling dicts?
Message-ID: <84ptmq$v65$1@nnrp1.deja.com>

Hi!

I'm trying to pickle an object with a dictionary in it, but it seems that the
data in the dict isn't pickled (or unpickled).

Another question:
Can I have shelve to use cpickle to improve performance?

Thanks,
Mike


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


From bsb at winnegan.de  Mon Jan 17 03:38:52 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 17 Jan 2000 09:38:52 +0100
Subject: Bugs in IDLE environment (??)
In-Reply-To: "Giovanni Riccardi"'s message of "Mon, 17 Jan 2000 07:05:37 +0100"
References: <85uc1m$5i8$1@pegasus.tiscalinet.it>
Message-ID: <87hfgday8z.fsf@baal.winnegan.de>

"Giovanni Riccardi" <composzioni at tiscalinet.it> writes:

> Just a repost of the message, because I had no
> replies within a week.

Mayby you didn't get an answer because you didn't give us enough
information on your environment and didn't even describe the
error you encountered ?

> ------
> I wrote a small script that uses input() to get
> the user input,get the values, make some calculation
> and return calculated value.

> The problem is that it works when the script is executed directly by the
> interpreter, ...

I doubt it does.

> ... but not in IDLE environment where after the user has entered
> some values the script doesn't return to the main loop.
> For example:
> 
> print "Enter the name:"
> x = input("--> ")

Here's what I get when pasting your commands into IDLE's shell:

Python 1.5.2+ (#4, Nov 18 1999, 01:39:08)  [GCC egcs-2.91.66 19990314 (egcs-1.1.2 release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
IDLE 0.5 -- press F1 for help
>>> print "Enter the name:"
Enter the name:
>>> x = input("--> ")
--> Siggy
Traceback (innermost last):
  File "<pyshell#1>", line 1, in ?
    x = input("--> ")
  File "<string>", line 0, in ?
NameError: Siggy
>>> 

[...]

> doesn't continue after the first input().

Sure, it raises an exception :)

> Some help??

Look into the "Builtin Functions" section of the library reference for
input vs. raw_input and work through the Tutorial.

> thanx in advance
> Giovanni

HIH
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
                  bsb at north.de - http://www.north.de/~bsb/
****** ceterum censeo javascriptum esse restrictam *******



From mhammond at skippinet.com.au  Wed Jan  5 22:16:44 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 06 Jan 2000 03:16:44 GMT
Subject: python with ASP
References: <007c01bf57f7$eb3941c0$db75fea9@desktop>
Message-ID: <wOTc4.11602$lv5.51551@news-server.bigpond.net.au>

What exactly are you asking - "why doesnt this work?"? - what are the
symptoms?  It should work...

Mark.

"Sameer Chowdhury" <sameerc1 at yahoo.com> wrote in message
news:007c01bf57f7$eb3941c0$db75fea9 at desktop...
> how could the following work.  It's supposed to either display a
form or
> display "Got Title".
>
> <%@ LANGUAGE = Python %>
> <%
> if REQUEST.Form("title"):
> %>
>  <html>
>  <b> Got Title</b>
>  </html>
> <% else: %>
>  <html>
>  <FORM ACTION="/pytest.asp" method="POST">
> <b>Title:&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</b><BR>
>  <input type="text" value="" size=60 name="title"><P>
>  <input type=submit value="Submit">
>  </form>
>  </html>
>
>
> __________________________________________
> NetZero - Defenders of the Free World
> Get your FREE Internet Access and Email at
> http://www.netzero.net/download/index.html
>




From jimbag at kw.igs.net  Thu Jan 13 11:55:01 2000
From: jimbag at kw.igs.net (JB)
Date: Thu, 13 Jan 2000 16:55:01 GMT
Subject: Dynamic module/package import ???
References: <000801bf5802$dcca7d60$0aa0143f@tim> <is66x6vng7.fsf@localhost.ps.uni-sb.de> <87u2kljrmz.fsf@den.home.net> <isbt6rhudp.fsf@grizzly.ps.uni-sb.de>
Message-ID: <387E035F.69F91BFE@kw.igs.net>

I'm trying to import a module from a package dynamically. The layout is
like so...

/DataHandler
  |- DataRow.py
  |- MyReader.py
  |- __init__.py

In class DataRow.DataRow I try...

  def setReader( self, reader )
    mod = __import__( 'DataHandler.' + reader + 'Reader' )

... but this just loads the __init__.py, so it seems that the __import__
doesn't understand packages or something.

Any help?

cheers
jim


From kc5tja at garnet.armored.net  Mon Jan 10 16:20:41 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 21:20:41 GMT
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu> <slrn87fol9.15t.kc5tja@garnet.armored.net> <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk> <slrn87imad.thc.wtanksle@hawking.armored.net> <slrn87j11s.vc0.kc5tja@garnet.armored.net> <Y$u7SGAdYae4EwJv@jessikat.demon.co.uk>
Message-ID: <slrn87kjlo.930.kc5tja@garnet.armored.net>

In article <Y$u7SGAdYae4EwJv at jessikat.demon.co.uk>, Robin Becker wrote:
>Don't get me wrong, but to attack me for saying I treat COM as too
>object oriented and then attack me for being an OO neophyte seems a bit
>silly. I'm from the fortran era, I treat COM objects as exactly that;
>blobs with interfaces, that doesn't mean I don't appreciate nice OO
>languages like Python. The reason I like Python is that it allows me to
>"just do stuff".

As indicated in the message, the text wasn't intended to be an attack on
you.  I apologize if it sounded like it did.

>As for openness I think having seen a lot of the process of DDE OLE OLE2
>COM DCOM COM+ from the outside makes me more than a little doubtful
>about commitment to open standards. I have to do other work, not become

DDE and OLE1 are jokes.  OLE2 is a known-proprietary API.  COM itself is
open.  DCOM is also open, as interoperability standards for them have been
published, and are free to be used anywhere.  The only problem with DCOM is
that they are documented as a set of changes to the existing DCE RPC
specification, which means you also need the DCE RPC specification to
understand DCOM.  Warning -- it's a BBIIGG specification.  :)

COM+ is currently a proprietary extension to COM, but are nonetheless plain
ol' COM objects.  And they document most (not all) of the specs on their
MSDN website, so it IS possible for someone to implement much of COM+'s
capabilities without any special blessing from Microsoft.  In fact,
something tells me that MS would be very happy if you did support the COM+
APIs under other operating systems.

>an expert in all areas of CS. You have a special interest and expertise;
>please don't criticise me for not having the same. I won't slam you for

I didn't.  At least, I didn't intend to.  Again, I apologize for what
sounded like an attack.  It wasn't meant to be.

>not understanding linear algebra if you complain it's complicated. The

Linear algebra...piece of cake.  Don't waste my time with such silliness.
Now, 2+3 ... THAT is something I'm having major issues with... ;-)

>fact that a document describing the 'latest' version of the API is
>available doesn't change the complexity or otherwise of the API. M$
>describes the Win32 stdcall interface in about one page to someone who
>understands a bit about assembly programming. The equivalent for COM is
>how much?

Win32 stdcall interface is a calling /convention/.  It is not an API.  Note
the difference.  And all the stdcall convention says is, "The callee shall
remote parameters from the stack, not the caller."

Now, I might refer you to the Win32 SDK, which DOES describe the Win32 API.
Note how much bigger it is than the COM 0.9 spec.  (The 0.9 spec for COM
consists only of the first six chapters, to be pedantic.  The rest of the
document is a description of the various interfaces commonly used under COM,
but which themselves are not part of COM proper).

>I don't pretend to fully understand the various aggregation mechanisms
>in COM/DCOM even though the underlying technology is fairly simple. COM
>interaction with threading models in win32 is indubitably complex as are
>events, but that's probably inevitable given the various threading
>models. 

Yup.  In fact, I'm running into _precisely_ the same problems under Linux
with GCOM.  The problem stems from the fact that not all object
implementations are thread safe.  For example, if I write a COM object for
Linux using GCOM which provides an interface to the GTK library, no matter
how thread-safe I make the COM object itself, GTK just plain ain't thread
safe.  Therefore, the GTK COM object would need to sit in a single-threaded
apartment, while everything else can sit in a free-threaded apartment.

Other object systems exhibit similar threading architectures, though they
implement them differently.  The only reason I'm preserving the "apartment"
concept under the Linux version of GCOM is because Linux's thread/process
model is similar enough to the Windows model.  Dolphin's GCOM, however,
requires that _all_ loaded binaries be thread safe, so it doesn't understand
the single-threaded apartment concept.

>As for verbiage I guess that's a matter of style. Modern practice calls
>these things interfaces etc I come from an earlier era where callables
>were functions/procedures etc.

Individual methods are functions or procedures.  The set, specified
collection of functions is known as an interface.  Example: fopen, fread,
fwrite, fclose, fprintf, etc. all belong to the ANSI C file-handing interface.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From MHammond at skippinet.com.au  Wed Jan 19 13:10:20 2000
From: MHammond at skippinet.com.au (MHammond at skippinet.com.au)
Date: Wed, 19 Jan 2000 18:10:20 GMT
Subject: olemsg32 module used in win32com exchange server example?
References: <slrn8897gs.3cm.grant@grante.comtrol.com>
Message-ID: <864ult$fot$1@nnrp1.deja.com>

In article <slrn8897gs.3cm.grant at grante.comtrol.com>,
  grant at nowhere. (Grant Edwards) wrote:
> I'm trying to run the exchange server test script that I found
> in the win32com sources.  I've two questions:
>
>  1) Where does the "olemsg32" module come from?  I've installed
>     python 1.5.2 and win32all and I don't seem to have olemsg32.

The recent samples dont include this.  This was generated by makepy,
before makepy had the ability to manage the generated files by itself.

See testExchange.py in the win32com\test directory.

>  2) Does the win32com stuff talk directly to the Exchange
>     server, or does it go through a local copy of Outlook?

It goes through the MAPI DLLs - the same DLLs that the local copy of
Outlook itself uses.

Mark.


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


From greg at cosc.canterbury.ac.nz  Mon Jan 31 22:51:51 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 01 Feb 2000 16:51:51 +1300
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com> <86np80$jhp$1@harvee.billerica.ma.us> <86ntlg$2tc$1@news.ycc.yale.edu> <86pi88$n3n$1@harvee.billerica.ma.us>
Message-ID: <38965857.F6518113@cosc.canterbury.ac.nz>

"Eric S. Johansson" wrote:
> 
> yes it is possible but then you lose the words "indent" and "dedent"
> for most uses.  One needs to be a little more careful when picking phrases
> for command use.

Non-word vocal sounds could be used,
e.g. a rising whistle for indent and
a falling whistle for dedent.

Disadvantage: possible misunderstandings
when a male programmer is dictating 
heavily-indented programs in the 
proximity of a female...

-- 
Greg Ewing, Computer Science Dept,
+--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg at cosc.canterbury.ac.nz	   +--------------------------------------+


From aahz at netcom.com  Sun Jan  9 11:07:38 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jan 2000 16:07:38 GMT
Subject: Requirements of modules?
References: <20000109145730.A26275@stopcontact.palga.uucp>
Message-ID: <85aboa$o7a$1@nntp1.atl.mindspring.net>

In article <20000109145730.A26275 at stopcontact.palga.uucp>,
Gerrit Holl  <gerrit at nl.linux.org> wrote:
>
>I'm planning to write a irclib module in the style of other *lib modules
>likt smtplib, httplib, ftplib and nntplib. I would love to see this
>module incorporated in the next Python distribution, and I've already
>read www.python.org/1.5/legalfaq.html. But are there any other
>requirements? Where can I read about them?

I would suggest that you start by designing the module and then checking
back here to see if you missed anything.  After that, WRITE the module
and submit it for general use.  Listen to the feedback.  Once it gets
established, send a note to Guido offering it to the distribution.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From gerrit.holl at pobox.com  Thu Jan 27 14:14:51 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 27 Jan 2000 20:14:51 +0100
Subject: Is there a text mode user interface for python?
In-Reply-To: <86q1bt$fkm$1@lola.ctv.es>; from aloy@ctv.es on Wed, Jan 26, 2000 at 08:17:47PM +0100
References: <86q1bt$fkm$1@lola.ctv.es>
Message-ID: <20000127201451.A1301@stopcontact.palga.uucp>

Antoni Aloy wrote on 948914267:
> Hi!
> 
> I'm looking for a text mode user interface for Python, that works with
> M$Windows.
> 
> Any help would be greatly appreciated!

Just type 'python' in a DOS box.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From wtanksle at hawking.armored.net  Thu Jan 20 00:04:46 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 20 Jan 2000 05:04:46 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com>
Message-ID: <slrn88d6g5.13f.wtanksle@hawking.armored.net>

On Wed, 19 Jan 2000 11:48:20 -0800, tye4 wrote:
>Hans-Joachim Widmaier <hjwidmai at foxboro.com> wrote in message

>> Every sane programmer indents his or her programs anyway. And all the C
>> programmers I know were bitten by this old example:

>>   if (a > 3)
>>     if (b == 4)
>>        c = 2;
>>   else
>>     c = 3;

>Perfect example.
>   if (a > 3)
>     if (b == 4)
>        c = 2;
>    else
>     c = 3;

>If you add a space, the 'else' will belong to the inner 'if'. If you delete
>one, the 'else' will belong to the outer one. Highly error prone.

As a matter of fact, it's not at all error-prone -- it's simply an
outright error.  Python won't accept that source code (and not only
because it's actually C); the indentation is not just error-prone, it's
ambiguous.  The else isn't in a sane location; nobody but the original
programmer can tell where it was supposed to be.

In C the above code would not cause an error, but would instead be
error-prone.

>There is not single programming language that forces such a bizarre
>indentation rule.

Yes, there is -- Python.

>If thousands of programmers have a problem with this, why isn't it fixed.

We _did_ have a problem with it.  That's why we chose Python, because it
fixed the problem.

>The solution is backward compatible:
>if x = 5:
>    print 'x is 5'
>end if

If you state that particular lie often enough, does that make it true?
Here's a counterexample:

end = 0
if x == 5:
   end = 1
   print 'x is 5'
print end

Surely even you can see the problem.

>-tye4

-- 
-William "Billy" Tanksley, in hoc signo hack


From sekter at matavnet.hu  Mon Jan  3 16:18:07 2000
From: sekter at matavnet.hu (Arpad Kiss)
Date: Mon, 03 Jan 2000 22:18:07 +0100
Subject: Help with Installer beta 3e
References: <3867a311@news.gvsu.edu>
Message-ID: <3871120F.98319B59@matavnet.hu>

Andrew Sterian wrote:
> 
> I'm trying to use Gordon McMillan's Installer beta 3e software to distribute
> a set of scripts as a package. This is simple stuff, just a bunch of scripts
> that need to be installed in a directory somewhere on the Python path
> (i.e., no DLL's, no Tkinter, etc.)   I need this distribution to be a simple
> EXE that the user can download and run, assuming Python has already
> been installed.
> 
> The problem is I can't figure out how to use the Installer package. I
> tried running the Simple script on my top-level script and the generated
> EXE file fails to run (it says it can't find  .PY file that is part of my
> package).
> I deleted all .PYC files as the installhelp file suggests, but nothing helped.
> 
> So...is there any documentation or tutorial that explains how to get started
> with Installer to do basic stuff? I'm afraid Gordon's work is so robust and
> flexible that the basics have been obscured.
> 
> Many thanks,
> 
> Andrew.
> 

Hi,
Examin the builder.log file or try to use the ArchiveViewer.py, they may
help...
Sometimes Simple script is enough sometimes not, I usally edit the cfg
file then use the builder.py. The installer package contains some cfg
files, i have stolen tricks from them.
Arpad Kiss



From eddy at vortigen.demon.co.uk  Tue Jan 18 19:28:47 2000
From: eddy at vortigen.demon.co.uk (Edward Welbourne)
Date: Wed, 19 Jan 2000 00:28:47 +0000 (GMT)
Subject: Suggestions for python 2
In-Reply-To: <200001171403.JAA10668@eric.cnri.reston.va.us> (message from
	Guido van Rossum on Mon, 17 Jan 2000 09:03:46 -0500)
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us>  
            <200001171159.LAA17397@lsls4p.lsl> <200001171403.JAA10668@eric.cnri.reston.va.us>
Message-ID: <m12AizP-00006lC@vortigen.demon.co.uk>

OK, examples of useful generators.
All URLs below are relative to http://www.chaos.org.uk/~eddy/dev/toy/
including refs to assorted.py files.

Harder for me to construct because, in truth, the reason for generators
is that class is a generator; if folk are going to insist on having the
ability to reorganise what class is, it can best be met by providing for
things that can replace class in the class statement, instead of by
bodging the way class itself works.  I should really try to illustrate
some of the meta-class notions that got discussed, but can't remember
them.  The type-checked class machine in check.py is an example.

The example I've already given is of the sub-namespace or submodule:
this is simply a plain way of making the namespace of a module
hierarchical; instead of defining several modules and importing them
into one container module (in the same sense as os.path is a
sub-namespace of os), construct a lightweight submodule within the
module itself - hence the planetary example in python2.html (and see
object.html).  For that, see initspace in python.py: examples of its use
abound once it's been defined; it constructs something lightweight,
simple and totally devoid of sophistication.

The planetary example could be extended, using generator classes with
well-chosen behaviours, to give lazy-lookup methods that compute derived
data from the given data, to deploy reflexes that ensure defining a
body's orbit gets the body added to the list of satellites of the
orbit's centre (used as orbit's base, below: bases needn't always be
read in the conventional way), etc.:

import python, math

python.gennie bodgen (python.fallbacks):
    def __call__(self, *bases, **what): # self will be Body
        what.update({ 'dict': { '__dict__': {} },
                      '__class__': self })
        inst, wrap = apply(python.wrapped, bases, what)
        wrap.__lookups__.insert(0, python.dictlook(wrap.__dict__))
        wrap.__lookups__.insert(1, python.bequeath(self, wrap))
        wrap.__lookups__.append(python.getatt(inst))
        wrap.__init__()

        return inst, wrap

bodgen Body:
    def __init__(self):
        self.satellites = []
        # or use a smarter object sorting entries by orbital radius
        self.name = self.__name__

    def __setattr__(self, key, value):
        if key == 'orbit':
            self.__dict__['orbit'] = value
            orbit.occupant = self

        else: self.__dict__[key] = value

python.gennie orbgen (python.fallbacks):
    def __call__(self, *bases, **what): # self will be Orbit
        what.update({ 'dict': { '__dict__': {} },
                      '__class__': self })
        inst, wrap = apply(python.wrapped, (), what)
        wrap.__lookups__.insert(0, python.dictlook(wrap.__dict__))
        wrap.__lookups__.insert(1, python.bequeath(self, wrap))
        wrap.__lookups__.append(python.getatt(inst))
        wrap.__init__(bases[0])

        return inst, wrap

orbgen Orbit:
    # Presumes it will be given centre as its one base
    def __init__(self, centre):
        self.centre = centre
    # will have .occupant set to Body in whose suite it appears
    # but not 'til after __init__

    def __setattr__(self, key, value):
        if key == 'radius':
            self.__dict__['radius'] = value # making it possible to:
            self.centre.satellites.append(self.occupant)

        else: self.__dict__[key] = value

    def __getattr__(self, key):
        if key == 'omega': # orbital angular velocity
            self.omega = 2 * math.pi / self.period
            return self.omega

from units import * # units of measurement ... another story.

Body Sun:
    """The Sun: the star at the heart of the solar system."""
    type = 'G2 V'
    magnitude = 4.79
    aliases = 'Sol',

Body Earth:
    """The blue planet, cradle of life."""
    Orbit orbit (Sun):
        """How our home moves about."""
        radius = 93 * mega * mile
        period = 1 * year
        eccentricity = .0167

Body Moon:
    """The main cause of our tides"""
    Orbit orbit (Earth):
        radius = .238855 * mega * mile
        period = 27 * day + 8 * hour
        eccentricity = .055

Body Mercury:
    Orbit orbit (Sun):
        radius = 57.9e9 * metre
        period = .241 * year
        eccentricity = .2056

# ...

and we can use all the usual machinery of classes to set up
lazily-evaluated functionality on Body and Orbit objects to, e.g.,
compute the scale of tides (though we'd need phase-of-orbit and
tilt-of-axis data to work out the actual values at given moments) and
other mechanical truths.  (The above is loosely based on a real piece of
python 1.5 which also handles units of measurement and error bars on
data - but in which the data-entry is nothing like as neat.)

The attentive will see that python.gennie isn't the ideal tool for this
job: I designed it as a means to build class.  I'd have to play with a
lot of generator classes like bodgen and orbgen before I'd know what to
replace python.gennie with ...

Here's another simple hierarchical namespace, making some use of ability
to refer to earlier names within the suite of a sub-object:

sub = python.initspace

sub Cosmos:
    G = 66.720 * pico * metre**3 / kilogramme / second**2
    Hubble = 2.5 * atto * Hertz

sub Quantum:
    h = 662.6176e-36 * Joule * second
    hbar = h / 2 / math.pi

sub Vacuum:
    c = 299792458 * metre / second
    mu0 = permeability = .4 * pi * micro * Henry / metre
    Z0 = impedance = mu0 * c
    epsilon0 = permittivity = 1 / Z0 / c

sub Thermal:
    k = Boltzmann = 13.8054 * yocto * Joule / Kelvin
    Stefan = 56.7032 * nano * Watt / (metre * Kelvin**2)**2

sub Planck:
    impedance = Vacuum.impedance # arguably / 4 / pi
    action = Quantum.h # maybe over 2 (or 4) pi
    charge = pow(action / impedance, .5) # 8.27 positrons

    speed = Vacuum.c
    _Newton = Cosmos.G / speed**3 # time/mass; arguably / 4 / pi
    momentum = pow(action / _Newton, .5) # a cat at about 7 mile/hour
    length = pow(action * _Newton, .5)

    mass, energy = momentum / speed, momentum * speed
    time, arearate = length / speed, length * speed
    current, magneton = charge / time, charge * arearate
    force, acceleration = momentum / time, speed / time

    temperature = energy / Thermal.k # very very hot
    torque = energy

Although the last should really be implemented using a generator class
which lets one set some sub-set of the units: it then infers any others
you ask for from formulae such as those after the initial few.

Sorry, the examples of generators aren't as good as I'd have liked.

	Eddy.



From eddy at vortigen.demon.co.uk  Sun Jan 16 19:15:23 2000
From: eddy at vortigen.demon.co.uk (Edward Welbourne)
Date: Mon, 17 Jan 2000 00:15:23 +0000 (GMT)
Subject: Suggestions for python 2
Message-ID: <m129zpL-00006JC@vortigen.demon.co.uk>

I have been thinking about the namespace types.
That's modules, packages, classes and instances.
I believe they could profitably be united into *one* type.
Along the way, I see some constructive uniformisations we could apply.

I have written this up at
  http://www.chaos.org.uk/~eddy/dev/toy/python2.html
and thence-linked locations.  [The page also describes a fix to grail
0.5 which may be needed when reading the associated .py files.]

My proposal contains enough slack to allow for the following things
(among others) that some folk seem to want:

  * namespaces that you can't modify once they have been initialised
  * namespaces in which attribute-modification is type-checked
  * hybrids of those with assorted degrees of control
  * certain tools the functional programmers crave, notably a truly
    faithful implementation of currie()
  * some of the truly whacky `metaclass' proposals

and, of course, the stuff we're used to (notably class, albeit demoted
to a built-in) with minimal perturbation.  The changes are, IMO,
straightforward and pythonic: they reduce the number of special
attributes the interpreter is obliged to know about and do *not* depend
on strange hacks or anomalies.  They do perturb syntax.

I intend to push this proposal during IPC8.  I would greatly appreciate
constructive comments in advance, if only to temper my enthusiasm ;^}

	Eddy.
--
Those who have seen versions of my pages prior to 2000/Jan/17 should
note that I have simplified and clarified them somewhat of late.
I believe I have eliminated all risk of exploding heads.  YMMV.



From n89553 at stud.upb.de  Sat Jan 22 07:33:45 2000
From: n89553 at stud.upb.de (n89553 at stud.upb.de)
Date: Sat, 22 Jan 2000 13:33:45 +0100
Subject: In A Pickle Over Pickle()
References: <sgaf8s4dtfq5mplbvuem1pgiscmmk9pp5n@4ax.com>
Message-ID: <3889A3A9.F4D90C4@stud.upb.de>

wolf at one.net wrote:
> 
> I'm using Python to create a text-adventure writing system, and need
> to use pickle to save/restore the program state. The problem is, I
> seem to have a mental block when it comes to Pickle--I've read all the
> documentation I can find on pickle() and can't make heads or tails of
> it. I suspect I'm missing some obvious "first clue" that would make it
> come together.
Perhaps it?s just too easy.

>>> menu = ["spam", "fries", "spam", "tatties", "lobster", "spam" , "fries" , "vanille fla", "spam" , "spam"] 
>>> import pickle
>>> jar = pickle.Pickler(open("spamfile", "w"))
>>> jar.dump(menu)
>>> del jar, menu
>>> print open("spamfile").read()
(lp0
S'spam'
p1
aS'fries'
p2
ag1
aS'tatties'
p3
aS'lobster'
p4
ag1
ag2
aS'vanille fla'
p5
ag1
ag1
a.
>>> jar = pickle.Unpickler(open("spamfile", "r"))
>>> print jar.load()
['spam', 'fries', 'spam', 'tatties', 'lobster', 'spam', 'fries',
'vanille fla', 'spam', 'spam']


> "skeleton" that has no variables of its own. Basically what I want to
> do is pickle/unpickle the *entire* mess, variables, dictionaries, etc,

What about something like:
>>> jar = pickle.dump((vars(), vars(my_module_NewZorkCity.topten)),  open("spamfile", "w"))

There?s another module, called shelve.py .
Shelves almost look like dictionaries, but they store their data on
disk.
>>> import shelve
>>> users=("NewZorkCity-users") # this shelve looks like a dictionary
>>> users["green_knight"]= userdata,NewZorkCity.state()
>>> users[""]=userdata, NewZorkCity.state()
>>> for user in users.keys():
...     print user
...
green_knight
spanish_inquisitor
>>> selected_userdata, selected_state = users["green_knight"]

shelve.py tries to guess which dbm to use. Python 1.5.2 can?t recognize
the magic cookies of some versions of bsddb , but there?s a patch
for whichdb.py at www.python.org

regards,

Mirko
-- 
M Li?, <n89553 at squid.upb.de>


From kuncej at mail.conservation.state.mo.us  Wed Jan  5 10:49:23 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Wed, 5 Jan 2000 09:49:23 -0600
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com> <386fa6b1.58568155@news.telus.net> <20000105085647.C9441@vislab.epa.gov>
Message-ID: <CJJc4.244$B5.4768@news.more.net>

> Dose of reality.  It has to take a big step in the cross-platform area
> before that can happen.  Tcl/Tk/Python builds without a hitch on a slew
> ...

Last I heard, Tcl/Tk/Python only has PostScript printing support. It's
really
hard to call a program windows-compatible if it doesn't support windows
printing.

  --Jeff




From jkahanpa at pcu.helsinki.fi.invalid  Tue Jan 11 10:43:54 2000
From: jkahanpa at pcu.helsinki.fi.invalid (Jere Kahanpaa)
Date: 11 Jan 2000 15:43:54 GMT
Subject: Newbie simple question - can't find answer
References: <85fhha$6js$1@nnrp1.deja.com>
Message-ID: <85fj3q$qb9$1@oravannahka.helsinki.fi>

Hi.

eleisonme at juno.com wrote:
: Hello all,
: I am using Python 1.5.2-7 on intel Red Hat Linux 6.1.
: I do not seem to be able to use any of the math or cmath functions
: either in a script or on the python command line.
: For example if I type:
:>>>import math
:>>>exp (4)

Try 
>>>import math
>>>math.exp(4)
OR
>>>from math import exp
>>>exp(4)


Happy hacking
	Jere Kahanp??


From pollarda at lextek.com  Tue Jan 25 14:42:21 2000
From: pollarda at lextek.com (Art Pollard)
Date: Tue, 25 Jan 2000 11:42:21 -0800
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com> <86fl0m$gmj$1@inputplus.demon.co.uk>
Message-ID: <86kqtq$6kq$1@jaffa.itower.net>

Here is my $0.02 worth on the various Python books.

Programming Python:  Not particularly good.  Not worth the $$$.  I'm sure
I'll
   find a use for my copy but I haven't found it yet.

Quick Python: Great Book!  I really like it it is well done, sections are
well explained.
  My only gripe is that it is too short.  I'd love to see the author flush
it out so that it
   is as long as Programming Python with the same clear style and
explanations.

Learning Python: Good book.  I found that it explained things well but
tended to
   gloss over some subjects and leaving me wondering....

Python Esential Reference: Excellent Book!  I highly recommend it.  Between
 Quick Python and PER, I'd say you would have the core of what it takes to
learn
  Python.

Of course, I am just learning Python myself.  It is funny to me that the
O'Reilley books
seemed to come up short on this one.  Usually, they are pretty good.   I'd
also like to
see a good explanation somewhere about how to use IDLE and techniques for
writing / debugging Python programs since Python's development tools are a
tad
rougher to the nice IDE's I'm used to.  (So I'm spoiled.... :-)

-Art




From jody at sccsi.com  Thu Jan 20 09:40:43 2000
From: jody at sccsi.com (Jody Winston)
Date: 20 Jan 2000 08:40:43 -0600
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net> <slrn87cvv9.gcu.kc5tja@garnet.armored.net> <3876893D.876A8949@rubic.com>
Message-ID: <wkn1q0erh0.fsf@sccsi.com>

Jeff Bauer <jbauer at rubic.com> writes:

> On a more serious note -- and as a matter of genuine interest -- are
> there any Python programmers using COM in mixed Unix and Windows
> environments?

I'm currently looking at doing this for one of my consulting clients.
I'm looking at using Nouveau from RougeWave to help with the CORBA and 
COM bridging.
-- 
Jody Winston



From larsde at my-deja.com  Mon Jan 17 22:59:40 2000
From: larsde at my-deja.com (larsde at my-deja.com)
Date: Tue, 18 Jan 2000 03:59:40 GMT
Subject: Update: Trouble with Python and ASP
References: <860gpo$882$1@nnrp1.deja.com>
Message-ID: <860of5$dh3$1@nnrp1.deja.com>

In article <860gpo$882$1 at nnrp1.deja.com>,
  larsde at telepost.com wrote:
> I'm experiencing disturbing problems with using Python as an ASP
> scripting engine. I've been working on a project for quite some time,
> and so far everything has been peachy.

[snip]

I also sometimes even get the following error message:

Python ActiveX Scripting Engine error '80020009'

Traceback (innermost last): KeyboardInterrupt:

?

This also happens only when I try to access the application with two or
more webclients at the same time. This can't possibly be right, right?

LT.




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


From ScrewMe at mypad.com  Mon Jan 17 01:55:30 2000
From: ScrewMe at mypad.com (TechSpood)
Date: Mon, 17 Jan 2000 00:55:30 -0600
Subject: simple question about raw_input
Message-ID: <1Zyg4.1372$Pb4.11218@newsfeed.slurp.net>

Ok I am about as new a you can get to programming and I have a question
which should be very simple.

I am trying to make a simple program which ask the user for his/her name
then says
"hello (name) what would you like to do something"
after it displays this it is supposed to take the users input and store it
in a variable... all in 2 lines (which I think is possible)

I tried this but it doesn't quite work right...can you tell me what I am
doing wrong?

name=raw_input ("What is your name?")
choice1=raw_input "hello, %s, would you like to do something?" % name


thanks






From wuyi at cyberway.com.sg  Tue Jan 25 00:47:19 2000
From: wuyi at cyberway.com.sg (wuyi)
Date: Tue, 25 Jan 2000 13:47:19 +0800
Subject: Just a testing
Message-ID: <388D38E6.254EDF11@cyberway.com.sg>

Just a testing



From echuck at mindspring.com  Thu Jan  6 05:17:02 2000
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Thu, 06 Jan 2000 05:17:02 -0500
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net> <N6Vc4.610$hJ1.7143@news.rdc1.tx.home.com> <slrn878b1i.k6b.kc5tja@garnet.armored.net>
Message-ID: <38746B9E.42AF1099@mindspring.com>

Thanks to everyone who has replied to my inquiry.

-Chuck


"Samuel A. Falvo II" wrote:

> >> Seems like many web host providers, such as MindSpring, offer CGI with
> >> Perl, but not other languages, notably Python.
> >>
> >> Does anyone know who would host a CGI Python site?
>
> Armored Internet (http://www.armored.net) supports Zope and Python CGIs.
> Zope is supported under our Unix web hosting packages, and Python CGIs are
> supported under both Unix and Windows NT web hosting packages.
>
> --
> KC5TJA/6, DM13, QRP-L #1447
> Samuel A. Falvo II
> Oceanside, CA



From matthewm at zebrasoft.co.nz  Mon Jan 24 04:35:31 2000
From: matthewm at zebrasoft.co.nz (Matthew Miller)
Date: Mon, 24 Jan 2000 22:35:31 +1300
Subject: using threads with python and winnt
References: <1264075593-4560755@hypernet.com> <38835f4c.1380925@news.iol.ie>
Message-ID: <388C1CE3.3391A5CF@zebrasoft.co.nz>

Be afraid, be very afraid! Seriously though, I head-banged with threading
under Win98 and eventually gave up. The app though works perfectly under
WinNT. Unless it is REALLY neccessary to support Win9x, spare yourself from
the pain. matthew.

Sean Mc Grath wrote:

> >Kreiner Karl or Charly or somebody wrote:
> >
> >>             does anyone have expiriences with using threads and winnt ?
> [...]
> <gmcm at hypernet.com> wrote:
> >Threads work dandy on NT
> [...]
>
> Does anybody know how good/bad/indifferent the threading stuff works
> on Windows 95/98? I am considering using threads for an app
> under development but I'm a bit worried (read "terrified") about
> the issues that might pop up in '95 and '98.
>
> regards,
> Sean McGrath
>

--
Matthew Miller
Zebra Software Ltd.

E-mail: matthewm at zebrasoft.co.nz
Ph: 025-2920625 / FAX: 03-3842397




From admin at moreclicks.com  Sun Jan  2 12:40:15 2000
From: admin at moreclicks.com (Wichtige Information!)
Date: 02 Jan 2000 17:40:15 GMT
Subject: ÖSTERREICH, DEUTSCHLAND, SCHWEIZ! 20 FREE PICS!
Message-ID: <386f8d7f$0$27620@SSP1NO17.highway.telekom.at>

20 supergeile Gratisbilder unter:
http://www.dirtnet.com/e/sex01.html

FETISH unter:
http://www.dirtnet.com/notdurft,Viel Spass!


From gawron at obop.com.pl  Tue Jan 25 10:29:23 2000
From: gawron at obop.com.pl (=?iso-8859-2?Q?Przemys=B3aw?= G. =?iso-8859-2?Q?Gawro=F1ski?=)
Date: Tue, 25 Jan 2000 16:29:23 +0100
Subject: NumPy
Message-ID: <388DC152.9624D8F5@obop.com.pl>

[ WinNT 4.0 SP 5, Python 1.5.2]

I've just downloaded Numerical-15.2 and tryied to run the setup.py
script but I get:
Tracebak (innermost last):
    File "setup.py", line 16, in ?
        from distutils.command import install
ImportError: No module named distutils.command

am'I missing something ?

--
Przemyslaw G. Gawronski UIN:8358522
mailto:gawronskip at usa.net
mailto:gawron at obop.com.pl





From bjorn at roguewave.com  Tue Jan 18 16:14:30 2000
From: bjorn at roguewave.com (bjorn)
Date: Tue, 18 Jan 2000 14:14:30 -0700
Subject: Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us> <m12AMt3-00007rC@vortigen.demon.co.uk> <al8u2kbycss.fsf@sirppi.helsinki.fi> <m3embfz6p5.fsf@greebo.nodomain.de> <s89kep3peh6124@corp.supernews.com>
Message-ID: <3884D7B5.23501D70@roguewave.com>


Evan Simpson wrote:

> Bernhard Herzog <herzog at online.de> wrote in message
> news:m3embfz6p5.fsf at greebo.nodomain.de...
> > > Edward Welbourne <eddy at vortigen.demon.co.uk> writes:
> > [snip]
> > > > Three new kinds of argument come into being (with my given spelling):
> > > >   * safe tunnel: after both * and **
> > > >   * keyword-only: after * but not **
> > > >   * positional-only: after ** but not *
> [objections to proposal]
> >
> > I'll second that. Turning a kludge that works because of an
> > implementation detail into an officially supported feature isn't a good
> > idea, IMO. Support for lexical scoping is the correct approach.
> [snip]
> > lexical scoping would take care of that, unless I'm missing something.
>
> Now that two people have asserted that lexical scoping is the right way to
> handle this, I'd like to know what they *mean* by lexical scoping.
>
> How would names get bound to (or just initialized from?) names in lexically
> enclosing scopes?  Implicitly? Explicitly?
> Would we still declare module-global variables, or would they use the same
> mechanism as names in intermediate scopes?
>
> I rather like the idea of having to explicitly bring names into an inner
> scope, and with semantics identical to parameter defaults.  Why is this so
> objectionable?  Is it just the syntax?

Yes, and it's ugly <wink>  It also leads to special casing of recursion:

    def foo():
        def bar():
            bar() # oops, I don't know who I am...

while Python's scoping rules are simple, they're definitely not intuitive.  I
think the best solution would be to make them simple and intuitive -- which
leads to implicit access to variables defined in an enclosing scope.

-- bjorn




From grant at nowhere.  Tue Jan 25 15:06:45 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 25 Jan 2000 20:06:45 GMT
Subject: Newbie question on paging..
References: <388DF473.1ED172A4@wjk.mv.com>
Message-ID: <slrn88s0il.e0a.grant@grante.comtrol.com>

In article <388DF473.1ED172A4 at wjk.mv.com>, Wm. King wrote:

>I have heard of paging but never really dug into it or used
>paging. I realize it has to do with handling large amounts of
>data on systems that might not have enough memory to handle it
>all at once? (Is this a correct assumption?)

Yes.

>Does it have something to do with swapping stuff in and out of
>memory? 

Yes.

>So in the final analysis, I am wondering what paging really is,
>under what kind of circumstances one would use paging and how
>one might use it using Python.

Paging is handled by the OS, not by the application.

Lets say you declare a really large array in C: 

unsigned char myBuffer[40000000]; 

Now you only use one part of it at a time. The OS makes sure
that the part you're using is in RAM.  The rest may be on a
swap partition of a hard disk drive.  You don't have to worry
about it, it all happens as if by magic.

>Or maybe a pointer to some on-line document or book so that I
>can get a better handle on it......

Any text on OS design should discuss it. "Operating System
Design [the Xinu Approach]" by Comer, and "Operating Systems
Design and Implementation" by Tannenbaum are the ones I see on
my bookshelf.

-- 
Grant Edwards                   grante             Yow!  Life is a POPULARITY
                                  at               CONTEST! I'm REFRESHINGLY
                               visi.com            CANDID!!


From sjoerd at oratrix.nl  Wed Jan 26 04:30:53 2000
From: sjoerd at oratrix.nl (Sjoerd Mullender)
Date: Wed, 26 Jan 2000 10:30:53 +0100
Subject: HTML Widget
In-Reply-To: Your message of Thu, 20 Jan 2000 15:06:11 -0500.
             <38876AB2.58FF3EA5@hks.com> 
References: <38876AB2.58FF3EA5@hks.com> 
Message-ID: <20000126093053.E37FB301CF9@bireme.oratrix.nl>

On Thu, Jan 20 2000 Chris Swepson wrote:

>     I've been searching far and wide for an HTML widget which can be
> used in Python/Motif.  I can't seem to find one anywhere.  It's needed
> for a very important job I'm doing.  Please respond ASAP.

If you can find the (ancient) sources for Mosaic 2.6 (try
http://www.ncsa.uiuc.edu/SDG/Software/XMosaic/) then you can use the
HTML module that is part of the Python X extension.

-- Sjoerd Mullender <sjoerd.mullender at oratrix.com>



From embed at geocities.com  Thu Jan 27 11:42:50 2000
From: embed at geocities.com (Warren Postma)
Date: Thu, 27 Jan 2000 11:42:50 -0500
Subject: newbie question: stdin, stdout, stderr?
Message-ID: <hE_j4.7222$S64.211505@nnrp1.uunet.ca>

How do you first USE, and second REDIRECT stdin, stdout, and stderr from
within a Python program?






From mlh at vier.idi.ntnu.no  Mon Jan  3 17:33:28 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 03 Jan 2000 23:33:28 +0100
Subject: Doing things in one line?
References: <20000103171702.A655@stopcontact.palga.uucp>
Message-ID: <iacso0erdjb.fsf@vier.idi.ntnu.no>

Gerrit Holl <gerrit.holl at pobox.com> writes:

> Hello,
> 
> I know there's a FAQ entry called "Can I write obfuscated one-liners
> in Python?", but for fun, I'm trying to do it now.
> 
> Can I do this in one line:
> if a: foo()
> elif b: bar()
> else: foobar()
> 
> Is it possible with an intelligent construct that looks like:
> a and foo() or a and bar() or foobar()
> 
> This does both foo() and foobar().

Just be sure that all your functions (foo, bar and foobar) return true
(i.e. 1). If they don't (or you're not sure) you must do something
like:

a and (foo() or 1) or b and (bar() or 1) or foobar()

> 
> And what about while- and forloops?

Use lambda and recursion or map/reduce etc.

> 
> regards,
> Gerrit.
> 
> -- 
> -----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
> Version: 3.12
> GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
> Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
> -----END GEEK CODE BLOCK-----
> 

-- 
--

  Magnus
  Lie
  Hetland


From ajwilson at ezy.net.au  Sun Jan 23 01:17:20 2000
From: ajwilson at ezy.net.au (Austin J Wilson)
Date: Sun, 23 Jan 2000 17:17:20 +1100
Subject: Using JDK12.2. with jPython
Message-ID: <948608363.1060599761@news.ezy.net.au>

I am trying to get JDK1.2.2 working with jPython under Win95.  The JDK is
installed under directory c:\jdk1.1.2.  What do I have to do so that jPython
can find the JDK class libraries?

Regards
Austin


From arcege at shore.net  Fri Jan 21 16:08:05 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 21 Jan 2000 21:08:05 GMT
Subject: Exiting program
References: <38873769.5048399@122.0.0.250> <20000121101100.A184338@vislab.epa.gov> <20000121102244.A184499@vislab.epa.gov>
Message-ID: <VU3i4.182$GX6.50425@news.shore.net>

Randall Hopper <aa8vb at yahoo.com> wrote:
: Randall Hopper:
:  |volumeregeling:
:  | |    sys.exit()                        <-- for some reason this also
:  |
:  |Keep in mind that this doesn't terminate your program.  It merely raises
:  |a SystemExit exception.  ... it can be caught and ignored, which is most
:  |likely what is going on:

: Should have mentioned this, but in case you have more important fish to fry
: right now, just call:

:      os._exit(0)

: to force an unfriendly, immediate exit (bypassing the exception propagation
: mechanism).

And probably bypassing such things as flushing openned files and such.

  -Arcege



From michelorengo at netscape.com  Mon Jan 24 17:19:43 2000
From: michelorengo at netscape.com (Michel Orengo)
Date: Mon, 24 Jan 2000 22:19:43 GMT
Subject: MAPI and NT Service
Message-ID: <388CD0EC.85A1D8A3@netscape.com>

Has anyone ever try to access MAPI within an NT Service?

I have a script that send messages thru MAPI and works fine when run
from the console.
I made a NT service of it, and it still runs fine with:
'python mapiService.py debug'
(not surprising since os.system is called)

BUT when I tried to start the service, I have the following error on the
line Session.Logon(...)

 File "C:\Program Files\Python\win32\lib\win32serviceutil.py", line 597,
in SvcRun
    self.SvcDoRun()
  File "G:\MyProjects\Mapi\mapiService.py", line 24, in SvcDoRun
    self.session.Logon(profileName="Michel
Orengo",profilePassword="561o1",showDialog=0, parentWindow=-1)
  File "C:\Program
Files\Python\win32com\gen_py\3FA7DEA7-6438-101B-ACC1-00AA00423326x0x1x0.py",
line 696, in Logon
    return self._ApplyTypes_(0x77, 1, (12, 0), ((12, 17), (12, 17), (12,
17), (12, 17), (12, 17)), 'Logon', None, profileName, profilePassword,
showDialog, newSession, parentWindow)
  File "C:\Program
Files\Python\win32com\gen_py\3FA7DEA7-6438-101B-ACC1-00AA00423326x0x1x0.py",
line 93, in _ApplyTypes_
    return self._get_good_object_(apply(self._oleobj_.InvokeTypes,
(dispid, LCID, wFlags, retType, argTypes) + args), user, resultCLSID),
pywintypes.com_error, (-2147221246, 'Invalid window handle', (1258,
'OLE/Messaging', ' [OLE/Messaging - [80040102]]', None, 0, -2147221246),
None).

I don't understand this com_error 'Invalid window handle'.
Any pointer will be greatly appreciated

BTW, how/where can I get an a plain english description of the com
error?

Thanks
Michel



From aa8vb at yahoo.com  Mon Jan 31 11:39:53 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 31 Jan 2000 11:39:53 -0500
Subject: anydbm - a simple question
In-Reply-To: <20000124112320.A440571@vislab.epa.gov>
References: <388E2BF3.8CCF5849@inka.de> <14478.13679.746001.81905@beluga.mojam.com> <20000121110715.A184793@vislab.epa.gov> <3diu0lbz8e.fsf@amarok.cnri.reston.va.us> <20000124112320.A440571@vislab.epa.gov> <388CE28F.D3A100AB@inka.de> <14477.47754.65118.177514@beluga.mojam.com> <20000121110715.A184793@vislab.epa.gov> <3diu0lbz8e.fsf@amarok.cnri.reston.va.us> <20000124112320.A440571@vislab.epa.gov>
Message-ID: <20000131113953.A955824@vislab.epa.gov>

Skip Montanaro:
 |Randall Hopper:
 | |   >>> db = anydbm.open( "database", "r" )
 | |   Traceback (innermost last):
 | |     File "<stdin>", line 1, in ?
 | |     File "/home/rhh/software/python-1.5.2/lib/python1.5/anydbm.py", line 80, in open
 | |       raise error, "need 'c' or 'n' flag to open new db"
 | |   anydbm.error: need 'c' or 'n' flag to open new db
 | |
 | |It created the database.dir and database.dat files in the write phase,
 | |but the read phase fails on the open.
 |
 |This comes up from time-to-time.  It is a fixed problem.  You can either
 |grab the CVS version of the entire shootin' match or visit
...
 |    http://www.musi-cal.com/~skip/python/Python/

I still get this error.

I pulled:

     http://www.musi-cal.com/~skip/python/Python/Lib/whichdb.py

replaced it in the Python tree, and removed the stale .pyc and .pyo.

What did I miss?

-- 
Randall Hopper
aa8vb at yahoo.com



From tjg at avalongroup.net  Tue Jan 18 14:16:24 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Tue, 18 Jan 2000 11:16:24 -0800
Subject: More DBI question
Message-ID: <3884BC08.CC920737@exceptionalminds.com>

Hi again, I begin to feel like a pest,

I guess I have a misunderstanding of how cursors are supposed work, or
more likely how I am supposed to work with cursors.

Current understanding:

Open Database Connection
Create a Cursor
Execute SQL queries using that cursor
Process  results

Now, I need to make all sorts of different queries using that cursor,
some are selects, some are updates, some are deletes, some deal with
functions

Now my misunderstanding comes when I do something like this

c.execute("""SELECT NEXTVAL ('id')""")

c.execute("""INSERT INTO mytable ( name, rank, serialno)
		VALUES( ?, ?, ?""", (namevar, rankvar, serialnovar))

This works as advertised

Howevever, the second time through this sequence, the first execute
returns an error along the lines of "Expecting 3 arguments received 1"

Now I understand where the three arguments part comes from, the cursor
just did something that required three arguments, but how do I "reset"
what the cursor expects, it seems clunky to call a db.commit() after
each execute.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From R.Fehrenbacher at iname.com  Tue Jan 25 12:18:16 2000
From: R.Fehrenbacher at iname.com (R.Fehrenbacher at iname.com)
Date: Tue, 25 Jan 2000 17:18:16 GMT
Subject: DCOracle on Oracle 8.1.x
Message-ID: <86klsi$ktb$1@nnrp1.deja.com>

Did anybody have success with this combination on
Solaris 2.6 (or other OS) and Python 1.52.

Got the module compile all right, and most things
I checked so far work. However some don?t:

Example:

cursor(select * from v$session)
raises ORA-01455 error, whereas
cursor(select sid,serial#,username,osuser from
v$session)
works just fine.

I had a similar problem with another select * from
a v$xxxxx table which I can't remember now. Again,
when I specified the fields explicitly things
worked.

Could anybody veryfy this problem on Oracle 8.0.5
e.g., or on other installations.

Cheers,

Roland


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


From pxl at bby.com.au  Fri Jan 21 00:38:29 2000
From: pxl at bby.com.au (Peter Larbalestier)
Date: Fri, 21 Jan 2000 16:38:29 +1100
Subject: ODBC connectivity to SQL7.0 mdb, UNIX - ODBC Bridges
Message-ID: <3887F0D5.EBC18C60@bby.com.au>

G'day All,

A number of queries:

1. Windows connection via ODBC to a SQL mdb on NT Server 4.0
    I have been making a connection to our SQL7.0 mdb via ODBC after
having set up a 'DSN' in the 'ODBC Data Source Administrator' window..
Most of the examples I have viewed so far refer to this practice.
    i.e. >>>import odbc, dbi
          >>>mydb = odbc.odbc(dsn,uid,pwd)

It worries me though that this seems to create a "permanent" connection
to the database as there is an open port listening for any queries.

Does anyone have any sample code for connecting on the fly to a SQL7.0
mdb WITHOUT using the 'ODBC Data Source Administrator' to setup a
"permanent" DSN.

Having programmed in VB I simplly conect on the fly using ADO to our
SQ7.0 mdb. Can this be done with Python using mxODBC?
    I believe so ! but can anyone help me with an example?

2. UNIX connectivity to SQL7.0 mdb on NT Server 4.0
    It seems that using ODBC - ODBC Bridge is the way to go but this is
after the NT problem.

I hope that this forum will provide some definitive answers for all us
'newbies' in regards to these important issues.

Regards

Peter Larbalestier
Down under



From ivanlan at callware.com  Fri Jan 21 16:53:54 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 21 Jan 2000 14:53:54 -0700
Subject: Python Rocks!
References: <200001211334.IAA28640@python.org> <86ajlq$661@news.or.intel.com>
Message-ID: <3888D572.2AB3FC3C@callware.com>

Hi All--

tye4 wrote:
> 
> Yes, I think I'll let you guys wallow in your ignorance.
> 

Good.

> -tye4
> 
> PSU boss <secret at email.com> wrote in message
> news:200001211334.IAA28640 at python.org...
> > Martijn Faassen wrote on 948370683:
> > > Someone get the Python Secret Underground to do something about him. :)
> >
> > He won't hurt us anymore.
> >

<python-mind-beginner's-mind>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From fdrake at acm.org  Fri Jan 21 12:06:36 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 21 Jan 2000 12:06:36 -0500 (EST)
Subject: Forcing exit(); unqualified exception catchers
In-Reply-To: <E12BhAp-0005EL-00@lsls4p>
References: <E12BhAp-0005EL-00@lsls4p>
Message-ID: <14472.37404.218296.724920@weyr.cnri.reston.va.us>

Edward Welbourne writes:
[...lots of stuff about bare except: clauses...]

Eddy,
  An excellent treatise on proper use of exceptions!  I hope all the
new Python users take the time to read through it carefully; it can
really help them create better code.
  Bare except: clauses are also a real problem when debugging, as
well.  Even if you do want to handle an exception that gets raised,
if you don't realize you're getting what you are getting but think its 
something else, your handler will probably break things in subtle ways 
that create a different problem later during execution.  This can be
*very* hard to track down in a highly stateful program!

 > try:
 >     return db.connect(user, password)
 > except (KeyboardInterrupt, SystemExit), what:
 >     raise what

  This can be simplified:

try:
    return db.connect(user, password)
except (KeyboardInterrupt, SystemExit):
    raise

 > So: the real lesson here is, use SystemExit or exit(), not _exit(), and
 > fix the wanton except: clauses that are preventing your SystemExit from
 > having its intended effect.  That way, all your finally: clauses will
 > get to do their stuff.

  There are proper uses of os._exit(), usually related to os.fork()
or running untrusted code that's still active on the stack (i.e., you
do *not* want to hand over control to except or finally clauses).


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From robin at jessikat.demon.co.uk  Sat Jan  1 19:30:28 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 2 Jan 2000 00:30:28 +0000
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
References: <841d0c$j0n$1@violet.singnet.com.sg>
 <387f6d37.25522789@news.demon.co.uk> <19991230065516.A778039@vislab.epa.gov>
 <W8TkoEATH4a4Ewpu@jessikat.demon.co.uk> <387378c0.4549271@news.demon.co.uk>
Message-ID: <xc1BqAAkwpb4EwWW@jessikat.demon.co.uk>

In article <387378c0.4549271 at news.demon.co.uk>, Andy Robinson
<andy at robanal.demon.co.uk> writes
>Robin Becker <robin at jessikat.demon.co.uk> wrote:
>
>>In article <19991230065516.A778039 at vislab.epa.gov>, Randall Hopper
>><aa8vb at yahoo.com> writes
>>>Andy Robinson:
>>> |Anyone who wants to discuss this, come join the python-finance list at
>>> |www.egroups.com/group/python-finance/.  It's been a bit quiet of late
>>> |but that's where my stuff and Kevin's is being discussed..  
>>>
>>>Will do.  You ought to get the list sited on:
>>>
>>>     http://www.python.org/psa/MailingLists.html
>>>
>>>This is the first I'd heard of it.
>>>
>>I have some good analytics for Markowitz risk/return optimisation. Does
>>this come into the area?
>
>Absolutely!
>
>- Andy
OK I have to wrap this in a python c-api extension. I guess the matrix
side is fairly easy I can use Numpy type things. I have a bit of a
problem with the old style IO handles though. The current thing uses
file descriptors to allow diversion of various messages to files rather
than stdio/stderr etc. Is there an obvious way to allow an extension
module to write messages to various fd type objects?


Also under certain extreme circumstances my code would like to report a
total failure eg attempt to divide by error. I would like to ask more
experienced pythoneers how arithmetic errors are handled using built in
IEEE exceptions and how does one gracefully crash a python script? (eg
an abend/abort).
-- 
Robin Becker


From bjorn at roguewave.com  Mon Jan 17 14:06:03 2000
From: bjorn at roguewave.com (bjorn)
Date: Mon, 17 Jan 2000 12:06:03 -0700
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com> <slrn886oku.2ac.grant@grante.comtrol.com>
Message-ID: <3883681A.DBD1B3C1@roguewave.com>

> >> I just found the http server page on Exchange server that
> >> allows access to e-mails via a web browser.  What I'd really
> >> like is to write a Python program that just sucks the mail out
> >> of the inbox and delivers it locally.
>
> My other option would be to write a MAPI client that runs under
> Win32 and then gives the mail to the SMTP server on my Linux
> machine.  There's a MAPI module for Python, isn't there?
> Perhaps that would be easier -- but I've never programmed under
> Win32 before...

If you do this, let us know (I know I would be interested ;-)

> >It might be easier to use the exchange server as either an IMAP
> >or POP server instead (both should be on by default).
>
> The admin shut off both POP and IMAP to try prevent people from
> using mail clients other than Outlook.  He's convinced
> management that people who don't use Outlook were increasing
> his workload. [...]

Seems to be the favorite pastime of sysadmin staff these days (and they wonder
why everything is so slow...)

-- bjorn





From pearu at ioc.ee  Tue Jan 25 15:15:12 2000
From: pearu at ioc.ee (Pearu Peterson)
Date: Tue, 25 Jan 00 20:15:12 GMT
Subject: f2py 1.116: The Fortran to Python Interface Generator
Message-ID: <mt2.0-2572-948831312@news.informatik.uni-bonn.de>

I am pleased to announce the first public release of f2py 1.116:

Writing Python C/API wrappers for Fortran routines can be a very
tedious task, especially if a Fortran routine takes more than 20
arguments but only few of them are relevant for the problems that they
solve.

The Fortran to Python Interface Generator, or FPIG for short, is a
command line tool (f2py) for generating Python C/API modules for
wrapping Fortran 77 routines, accessing common blocks from Python, and
calling Python functions from Fortran (call-backs).

The tool can be downloaded from

    http://cens.ioc.ee/projects/f2py2e/

where you can find also information about f2py features and its User's
Guide.

f2py is released under the LGPL license.

With regards,
     Pearu Peterson <pearu at ioc.ee>

<P><A HREF="http://cens.ioc.ee/projects/f2py2e/">f2py 1.116</A> - The
Fortran to Python Interface Generator.  (25-Jan-2000)

-- 
----------- 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 philh at vision25.demon.co.uk  Fri Jan 14 08:50:54 2000
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Fri, 14 Jan 00 13:50:54 GMT
Subject: Parser Generator for Python
References: <387E8B81.23C91488@attglobal.net> <009f01bf5e6a$f93b7170$f29b12c2@secret.pythonware.com>
Message-ID: <947857854snz@vision25.demon.co.uk>

In article <009f01bf5e6a$f93b7170$f29b12c2 at secret.pythonware.com>
           fredrik at pythonware.com "Fredrik Lundh" writes:
> Yapps and Spark appear to be among the more popular:
> http://theory.stanford.edu/~amitp/Yapps/
> http://www.csr.uvic.ca/~aycock/python/

I use SPARK in my Parrot project; I find it expressive and easy to 
use (apart from the lexical analyser's use of regular expressions;
I've come to the conclusion that REs are intrinsically difficult to
understand).

-- 
Phil Hunt - - - phil at comuno.com

 See Comuno's Linux coverage at <http://www.comuno.com/linux/> for:
* Links to useful sites * Linux-related news * Open source software *



From bitbucket at isomedia.com  Wed Jan  5 14:03:20 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Wed, 05 Jan 2000 19:03:20 GMT
Subject: Python and ASP - foo.asp (0/1)
References: <_ji34.8719$523.343739@news.chello.at> <bBk34.1288$Cl3.6689@news-server.bigpond.net.au> <slrn84t6n0.q3j.aa126@fan1.fan.nb.ca> <BZA34.2308$Cl3.11615@news-server.bigpond.net.au> <387351CA.1EA77B30@gssec.bt.co.uk>
Message-ID: <3873939b.332551623@news.isomedia.com>

Here's a simple ASP page in Python that dumps the contents of some
collections in the Request object.

'>>>
<%@ LANGUAGE = Python %>

<%
def HTMLTableFromCollection (oCollection, sCollectionName):
	sHTML = '<table border>\n<tr><th
colspan=2>%s</th></tr>\n%s</table>\n'
	sRowPat = '<tr><td>%s</td><td>%s</td></tr>\n'
	sBody = ''
	for sKey in oCollection:
		sBody = sBody + sRowPat % (sKey, oCollection (sKey))
	return sHTML % (sCollectionName, sBody)
%>

<%= HTMLTableFromCollection (Request.ServerVariables,
'Request.ServerVariables') %>
<%= HTMLTableFromCollection (Request.ClientCertificate,
'Request.ClientCertificate') %>
<%= HTMLTableFromCollection (Request.Cookies, 'Request.Cookies') %>
<%= HTMLTableFromCollection (Request.QueryString,
'Request.QueryString') %>
'<<<

On Wed, 05 Jan 2000 14:14:34 +0000, Alan Gauld
<alan.gauld at gssec.bt.co.uk> wrote:

>Mark Hammond wrote:
>> No - you can do Python in ASP.
>
>Care to elaborate?
>
>> Not really.  Python programs running under ASP make extensive use of the ASP
>> object model, and hence won't work outside an ASP environment.  
>
>Again, how is that done? Any Web pages you can reference?
>I don't see any ASP references on python.org
>

The ASP object model is, in my mind, just the "5 magic (COM) objects"
always available to the script.  They are used in Python ASP pages
just like in VBScript ASP pages, with a few exceptions, such as the
need for exact Capitalization in some instances and a Python-side bug
that makes setting certain values in the objects more difficult.  (One
cannot simply say "Session ('foo') = 5".  One must instead say
"Session.SetValue ('foo', 5)".  And Response.Cookies is still
resisting :(.... )

     -Eugene
import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From reic0024 at ub.d.umn.edu  Sun Jan 30 23:09:34 2000
From: reic0024 at ub.d.umn.edu (Aaron J Reichow)
Date: Sun, 30 Jan 2000 22:09:34 -0600
Subject: Python on Rhapsody DR2/x86
Message-ID: <Pine.SO4.4.02.10001302153550.3381-100000@ub.d.umn.edu>

Good evening, fellow Pythonistas!
	I thought I'd try this as a last resort...  

	I'm trying to get Python 1.5.2 working under Rhapsody DR2/Intel.
I used the "./configure --with-threads --with-next-framework --with-dyld"
to configure for the Makefile.  When I make, the process stops with the
following:  

./posixmodule.c: At top level: ./posixmodule.c:2695: conflicting types for
`putenv'
/System/Library/Frameworks/System.framework/Headers/bsd/stdlib.h:120:
previous declaration of `putenv' gnumake[1]: *** [posixmodule.o] Error 1
gnumake: *** [Modules] Error 2

I assume this has to do with the abandoned POSIX stuff (I think NeXT
abandoned POSIX with OpenStep, no?).  Is there any way I can get this to
work?  I've heard that Python 1.5.2 builds "right out of the box" on Mac
OS X Server-- there aren't too many differences between Rhapsody 5.1
(uname's description of Rhapsody DR2) and Rhapsody 5.3 (uname's desc. of
Mac OS X Server 1.0.2), so I'm suprised this didn't build fairly easily.

Does anyone have any ideas?  I've talked with one gent who couldn't get
Python 1.5.2 working under OpenStep 4.2, but I would appreciate it if
anyone has any experience getting (or not getting) Python 1.5.2 working
under Rhapsody.

I realize Tkinter won't work, I just want to do some Zope, and failing
that, other cgi type projects which I've started to develop under
Linux/x86, but would like to be using Rhapsody for verious reasons.

Thanks!
Aaron



From fredrik at pythonware.com  Tue Jan 11 05:15:21 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 11 Jan 2000 11:15:21 +0100
Subject: xmllib has fatal bug.
References: <slrn87kpsu.aq0.kc5tja@garnet.armored.net>
Message-ID: <001901bf5c1c$df1a36a0$f29b12c2@secret.pythonware.com>

Samuel A. Falvo II <kc5tja at garnet.armored.net> wrote:
> What is up with this?  Can xmllib not handle namespaces?

your problem is that it does handle namespaces.

> What does it do with a namespace it does find, and why does
> it invalidate the mere existance of the tag?  It doesn't even _call_
> the start_schema method.  :( :( :(

in xmllib, the element name consists of the name-
space plus the tag, not just the tag.  in other words,
there is no element named "schema" in your file.

to see for yourself, add this method to your parser
class, and run it again (you have to fix the quotes
first, of course):

   def unknown_starttag(self, tag, attributes):
       print repr(tag), attributes

the docs have more info on this:
http://www.python.org/doc/current/lib/xml-namespace.html

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From tim_one at email.msn.com  Sat Jan 29 04:28:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 29 Jan 2000 04:28:37 -0500
Subject: statements and blocks
In-Reply-To: <86tt56$d4a$1@nnrp1.deja.com>
Message-ID: <000701bf6a3b$3932aba0$5a2d153f@tim>

[rdudfield at my-deja.com]
> Speaking of blocks, is there a way to show where the block ends?

How about adding a comment <0.5 wink>?

> I have been caught a few times doing this:
> class b():
>   def asdf():
>     print 'a'
>     if bla:
>       print 'a'
>       print 'a'
>       ... more than a page of stuff.
>
>    # method is supposed to be in the classes scope.
>     def qwer():
>       print 'a'
>       print 'a'
>       print 'a'
>
> Basically because the function asdf is more than a page long.
>
> I find indentation is better in most cases except this.  This isn't too
> much trouble, as I usually find the problem pretty quickly.  However
> occasionally I waste a bit too much time on it than I'd like to :)

Quite a few years ago I dug up this quote from Donald Knuth, which has since
been widely misquoted in the Python world:

    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

A key word there is "small":  indentation sucks for functions that slobber
over pages.  The solution is obvious:  don't write functions that slobber
over pages!  You can do *so much* in so few lines of Python that there's
simply no good excuse for violating common sense here.  It's one of the ways
in which Python (not so) subtly encourages good practice.

easy-readability-is-an-author's-greatest-achievement-ly y'rs  - tim





From moshez at math.huji.ac.il  Mon Jan 31 22:20:35 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 1 Feb 2000 05:20:35 +0200 (IST)
Subject: Dynamic class construction?
In-Reply-To: <001301bf6c16$81e08ac0$3acbd9c2@peridot.optichrome.com>
Message-ID: <Pine.SOL.3.96.1000201051814.23158M-100000@sundial>

On Mon, 31 Jan 2000, Adrian Eyre wrote:

> You didn't look very hard :)
> 
> http://www.python.org/doc/current/lib/module-new.html
> 
> Ah. Be aware this module was only introduced in 1.5.2 IIRC.

Nope -- it was only documented in 1.5.2, and only at p1 or p2 (I can't
remember). I can say it with certainty, because I'm the poor guy who
documented it <wink>.

--
Moshe Zadka <mzadka at geocities.com>. 
INTERNET: Learn what you know.
Share what you don't.




From tim_one at email.msn.com  Wed Jan 19 23:54:00 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 19 Jan 2000 23:54:00 -0500
Subject: Python Rocks! - get rid of colons
In-Reply-To: <Pine.WNT.4.21.0001191125160.243-100000@wks_3.e-shopwork.com>
Message-ID: <000701bf6302$5eb88a60$31a2143f@tim>

[Sposhua]
> I'm one of the "prefers indentation to {}" ppl, and I'd
> just like to add: whe the heck do we need colons (:) at
> the end of if's 'n loops anyway?  The only loss I see is
> to these 1-liners:
>
> if a==b: eric=idle
>
> What's wrong with:
>
> if a==b
> 	eric=idle

As Michael Hudson said, the connection between colons and blocks was adopted
from Python's closest predecessor, ABC.  Formal usability studies showed
that while indentation was less confusing than assorted bracketing
constructs to newbies, the combination of indentation and an introductory
colon was even better.  You're right that the grammar has little use for it;
although, pragmatically, it's a great aid to semi-stupid tools (like Python
editor modes -- not to mention all us Python programmers <wink>).

> (oh, and wouln't it make life easier to change == to = and
> let python work out when it is assingment and when it is
> comparison?)

Before its 1.0 release, Python *did* use "=" for both binding and equality
testing.  It didn't work out well; e.g., picture

>>> x = y

in interactive mode.  It's truly ambiguous (are you trying to bind or test
for equality?).  It was taken to mean binding, and testing for equality
needed to be spelled with parens, like

>>> (x=y)

Ambiguity, and Yet Another Special Meaning for Parens, were taken as bad
omens.  So Guido exorcised those demons.

BTW, you can still find the script that was distributed at the time (to
change equality "=" to "==") in the source distribution, in
Demo/scripts/eqfix.py.  I wrote a script at the same time to do the same
thing, but in Perl.  This had the intended effect of redoubling Guido's
resolve to continue developing Python <wink>.

although-parsing-without-a-parser-is-hokey-in-any-
    language-ly y'rs  - tim





From tye4 at yahoo.com  Thu Jan 20 20:04:34 2000
From: tye4 at yahoo.com (tye4)
Date: Thu, 20 Jan 2000 17:04:34 -0800
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim>
Message-ID: <868bbv$lug@news.or.intel.com>

Tim Peters <tim_one at email.msn.com> wrote in message
news:000001bf63a1$832e5740
> I haven't used this script myself (I prefer indentation, for reasons
covered
> earlier), but there's no good reason it *can't* be used for its stated
> purpose <wink>.
>

I don't oppose indentation either. It's good for program readability, and
readability only. It shouldn't be part of the syntax. Whitespace end
delimiters are invisible and hence confusing and prone to error.
I've used some stupid Unix shell languages which fail if I add whitespace:

Without spaces (this works):
    $x=10
    $y=2+2

With spaces (doesn't work):
    $x  =  10
    $y  = 2 + 2

I thought they gotten rid of all such languages until I came across Python.
Good programming languages should be whitespace tolerant, and most of them
are.
(C++, Java, Ada, BASIC, HTML, Eiffel, Perl, Pascal, ML etc... really long
list.)

Why is my question about whitespace raise so many eyebrows? This feature is
commonplace.

-tye4





From jblazi at netsurf.de  Sat Jan 29 03:13:05 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Sat, 29 Jan 2000 09:13:05 +0100
Subject: Rational numbers
References: <388f30c7_3@goliath.newsfeeds.com> <p6qoga6j800.fsf@informatik.hu-berlin.de>
Message-ID: <38929f74_4@goliath.newsfeeds.com>

It is probably a matter of taste and may also depend on how much code you
already have...
I would very much prefer 3/4 != 0.

J.B.



Martin von Loewis <loewis at informatik.hu-berlin.de> schrieb in im
Newsbeitrag: p6qoga6j800.fsf at informatik.hu-berlin.de...
> "Janos Blazi" <jblazi at netsurf.de> writes:
>
> > What do you think, wouldn't it be nice if Pythony had the built-in data
type
> > "rational" as LISP has? (I would like it. The truth is that the packages
> > are a bit less confortable.
>
> What's wrong with
>
> >>> from yarn import Rat
> >>> Rat(4,5) * Rat(7,12)
> Rat(7, 15)
>
> I can't imagine that you can get this much more comfortable, without
> breaking existing programs. I guess you'd like 4/5 return a rational
> number, even though it currently gives you 0. I would not like that.
>
> Regards,
> Martin
>




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From gerrit.holl at pobox.com  Sun Jan 30 04:14:44 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 30 Jan 2000 10:14:44 +0100
Subject: Creating a local variable with a parameterized name
In-Reply-To: <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu>; from dworkin@ccs.neu.edu on Sat, Jan 29, 2000 at 06:19:10PM -0500
References: <38937075.18B75753@rochester.rr.com> <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <20000130101444.A1582@stopcontact.palga.uucp>

Justin Sheehy wrote on 949166350:
> Michael Ross <mross at rochester.rr.com> writes:
> 
> > I want to create a local variable with a passed in name.
> > 
> > I realize I can do the following:
> > 
> > >>> def f(varname):
> >              cmd = '%s = 123' % varname
> >              exec(cmd)
> >              cmd = 'print %s' % varname
> >              exec(cmd)
> > 
> > >>> f('mike')
> > 123
> > >>>
> > 
> > But exec needs to invoke the parser, which seems to be an extremely
> > costly thing to do.
> 

...

> my_directory = {}
> my_directory[varname] = 123
> print my_directory[varname]

...

> Directories are handy, and particularly well suited to this sort of
> thing.  Why be any messier than you need to be?

<nitpick>
 Those things are called "dictionairies", not "directories". You could
 use them for a directory structure, however :-)
</nitpick>

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From tjg at avalongroup.net  Tue Jan 18 13:24:50 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Tue, 18 Jan 2000 10:24:50 -0800
Subject: DBI spec
Message-ID: <3884AFF2.9BE7773D@exceptionalminds.com>

More questions regarding the DBI spec:

First, is it possible to walk backward through a cursor? All the
fetchXXX() functions appear to only move forwards, however, one of my
SQL books mentions the ability to move backwards as well as forwards.

Second, and very much related to the first, how are the results of a
cursor query stored, are they stored in temporary storage on disk, are
they loaded in memory, etc.

Third, at the moment, given the current size of the results that my
queries retrieve, I can implement the backward cursor movement by using
fetchall() and traversing the resultant list, however, I can forsee
that, given a large result set, that method will cause grief, how have
others resolved the bi-directional cursor traversal issue?

Thanks for your input.


-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From esj at harvee.billerica.ma.us  Wed Jan 26 16:33:52 2000
From: esj at harvee.billerica.ma.us (Eric S. Johansson)
Date: 26 Jan 2000 21:33:52 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com>
Message-ID: <86np80$jhp$1@harvee.billerica.ma.us>

Fredrik Lundh <fredrik at pythonware.com> wrote:
> Roy Smith <roy at popmail.med.nyu.edu> wrote:
>> When I was first learning programming (a quarter of a century ago!), 
>> there was the idea of the "telephone test", i.e. your code should be 
>> simple enough that if you read it to somebody on the phone they would 
>> understand it.

> hey, who the heck is reading code over the phone
> these days?

anyone trying to program using speech recognition.  Why would they 
do this you might ask?  Probably because they were injured from
programming using keyboards.

yes, it's not strictly reading code over the phone but there 
are significant similarities...

--- eric

-- 
Eric S. Johansson    esj at inguide.com	esj at harvee.billerica.ma.us
This message was composed almost entirely using NaturallySpeaking


From jcollins at pacificnet.net  Thu Jan 27 13:02:26 2000
From: jcollins at pacificnet.net (Jeffery D. Collins)
Date: Thu, 27 Jan 2000 10:02:26 -0800
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com> <slrn88pg1h.scl.wtanksle@hawking.armored.net> <m3r9f3veja.fsf@chinon.cnrs-orleans.fr> <UIYj4.75$q64.2465@weber.videotron.net> <389059A0.B37CEBBF@tismer.com>
Message-ID: <38908832.48062DA1@pacificnet.net>

My message concerning this must not have gotten through
to the python-list.  I responded via the egroups server, but
the message didn't appear:(

The release of python-1.5.1 for the PalmIII (or greater) will
be announced within the next couple of weeks.  I need to clean
up the distribution and documentation first.  The first release
will probably be the final for 1.5.1, since the porting tools have
improved and we are eager to get in sync with the latest
stuff.  With the lessons learned and code developed during the
first port, the second should go much more quickly (the first
took several part time months)...

Jeff

Christian Tismer wrote:

> Steve Menard wrote:
> ...
> >     Any timefame for the palm port? Any details? I have looked at doing a
> > port myself, and am still in the early stages. I'd like to know what
> > features of python will be supported and of course how long I'll have to
> > wait :)
>
> Howdy,
>
> Please contact Jeff concerning this.
> Also let me know wether you think it makes sense to
> set up a mailing list for palm ports. I want to make
> this happen! :-)
>
> Finally I'm considering to delve into porting myself
> if that helps.
>
> ciao - chris.stackless
>
> --
> Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
> Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
> D?ppelstr. 31                :    *Starship* http://starship.python.net
> 12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
> PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
>      we're tired of banana software - shipped green, ripens at home




From paul.robinson at quantisci.co.uk  Thu Jan 27 04:22:14 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Thu, 27 Jan 2000 09:22:14 +0000
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com>
Message-ID: <38900E46.909421F8@quantisci.co.uk>

Joel Lucsy wrote:
> 
> I have a version of python15.dll that I've compiled using the delayload
> mechanism. Basically this means that it will only load a dll when it needs
> it. Running a very basic test on my system (400Mhz celeron,128Mb,Win2K RC3)
> the load time went from over 2 seconds to under half a second. If you wish I
> can email it to you, or if someone wants, I'll upload it somewhere.

I'd be very interested in that!

If it's about the size of the standard dll then that's pushing it a bit
on our mail server, so if you could upload it somewhere that would be
great.

	Paul.



From neelk at brick.cswv.com  Tue Jan 18 20:53:33 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 19 Jan 2000 01:53:33 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net>  <8631bt$1nu@news.or.intel.com>
Message-ID: <slrn88a8jb.7qj.neelk@brick.cswv.com>

tye4 <tye4 at yahoo.com> wrote:
> 
> So you don't think this is a problem?? Hello!, read all the messages in this
> thread. Seems to me a lot of programmers have a problem with this.
> Thanks to this quirk, I now have to worry how an editor saves tabs to files.
> I have offered you many reasons why we need 'end' block support..
> You have not offered a single reason why this is not a problem..

My other language-of-choice, Dylan, has a syntax and level of
expressive power very similar to Python's, except that it has "end"
block delimiters.

The Gwydion Dylan compiler is about 56,000 lines of non-comment
non-blank lines. Of these 56 KLOC, 8700 contain only "end" statements.
That's about 16% of the code, or 1 line in 6. Presumably a similar
fraction of Python code would be end statements if Python had block
delimiters too.

It's also well-known that the average time to debug a function or
class goes up with length, with a sharp upwards kink when the length
of the program object exceeds 1 screenful, *whatever the screen size
is*.

Let's suppose that we are editing on a terminal that has 42 horizontal
lines. Then adding block delimiters to Python would mean that the
effective screen size would shrink from 42 currently to 35 plus
delimiters. This is a *painful* loss, considering that the debug time
kinks upward whenever a program exceeds one screenful in size.
Consequently, I think Guido made an excellent choice in keeping block
delimiters out of Python. The gain in readability is well worth the
slight inconvenience of dealing with tabs.

(BTW, I also think Dylan did the right thing, because the block
delimiters in Dylan are mostly there to make it possible to have a
hygienic macro facility. Since Python doesn't have this, it doesn't
need gratuitous "end" statements.)


Neel


From tim_one at email.msn.com  Mon Jan 24 19:30:40 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jan 2000 19:30:40 -0500
Subject: Python Rocks!
In-Reply-To: <388BB4E1.974E63B7@prescod.net>
Message-ID: <000f01bf66cb$6861d720$a42d153f@tim>

[Paul Prescod]
> Wouldn't it be easier if we just added an optional, (but
> compiler-recognized if present) block delimiter to Python 2?

The option was offered to Guido years ago, via a patch by Donald Beaudry to
do just that.  He declined then, and I doubt he'd change his mind now.

> Then people would have to stop complaining that Python is
> broken

No, they wouldn't.  That's the whole problem here:  people with a bug up
their butt about indentation are no different than people with a bug up
their butt about integer division (like me!), or people who hate reference
counting, or those who hate the lack of an address-of operator, or lack of
lexical closures, or lack of an ISO std, or ... I once drew up a list of
about 30 of these "crucial show-stoppers" culled from semi-hysterical c.l.py
threads.

When people say that Python's indentation drives away thousands of
programmers, they're almost certainly correct!  And so do each of dozens of
other things.  The same is also true of every other language on earth.
Guido is on the receiving end of *all* these gripes, so has a sobering
global perspective the proponents of individual features X, Y and Z lack in
their single-minded zeal.  Over the years, and only after a great deal of
thought and care, he occasionally "yields" on the few he judges most likely
to add more power than cause harm.

I don't know whether he keeps a formal priorty list of these things, but, if
he does, optional block delimiters must be scraping the bottom of it (adds
no power, at best leads to the usual flamewars about where the delimiters
"should be" placed, at worst ruins the easy visual uniformity of Python code
that so many old-timers (incl. esp. Guido!) actively love).

> and instead admit that Python programmers choose to
> program without delimiters.

The existence and utter non-use of Guido's Tools/Scripts/pindent.py is
sufficient proof of that.

that-was-indeed-his-last-word-on-the-subject-ly y'rs  - tim





From tim_one at email.msn.com  Tue Jan  4 03:09:38 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 4 Jan 2000 03:09:38 -0500
Subject: Re module help?
In-Reply-To: <84s5t7$ive$1@brokaw.wa.com>
Message-ID: <000001bf568b$0c305440$b02d153f@tim>

[Elf Sternberg]
> There is a line in 'xmllib.py' that has me puzzled.  I
> thought I understood regular expressions and XML, but this
> one has me confused.
>
> interesting = re.compile('[]&<]')
>
> I don't get it.  What is this looking for?

One of the three characters
    ] & <

> The '[]' can't mean an empty set, can it?

Right, the way to write a character class that can't match anything is

    [^\000-\377]

<arghggh!>.

> That doesn't make any sense.  Is this a special case
> where '[]...' allows you to include the ']' in things
> to search for without escaping it?

Yes, and it's common in regexp pkgs.  I think the line would be much clearer
as

    interesting = re.compile(r'[\]&<]')

There are similarly revolting special cases involving "-" in character sets,
which also ascribe meaning to what *looks* like an error.  That's a very
un-Pythonic thing to do, but Python's re package is intended to be
compatible with Perl's.

> Or is it, in fact, a bug?

Perl?  Yes <wink>.

non-judgmentally y'rs  - tim





From paul.magwene at yale.edu  Thu Jan 20 21:35:06 2000
From: paul.magwene at yale.edu (Paul M)
Date: Thu, 20 Jan 2000 21:35:06 -0500
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com>
Message-ID: <868ghe$8pl$1@news.ycc.yale.edu>

tye4 wrote in message <868bbv$lug at news.or.intel.com>...
[snippage]
>I don't oppose indentation either. It's good for program readability,
and
>readability only. It shouldn't be part of the syntax. Whitespace end
>delimiters are invisible and hence confusing and prone to error.
>I've used some stupid Unix shell languages which fail if I add
whitespace:
>
>Without spaces (this works):
>    $x=10
>    $y=2+2
>
>With spaces (doesn't work):
>    $x  =  10
>    $y  = 2 + 2
>
>I thought they gotten rid of all such languages until I came across
Python.
>Good programming languages should be whitespace tolerant, and most of
them
>are.
>(C++, Java, Ada, BASIC, HTML, Eiffel, Perl, Pascal, ML etc... really
long
>list.)
>
>Why is my question about whitespace raise so many eyebrows? This
feature is
>commonplace.
>
>-tye4


If you are going to spew criticisms (after how many days of using the
language?) please at least get your facts right.  Python IS whitespace
tolerant, your former examples:

    x=10
    y=2+2

and
    x  =  10
    y  =  2 + 2

mean exactly the same thing in Python as you would expect them to in
any of the other languages you cited.

What you were complaining about previously was the use of indentation
for block-delimiting. You suggested that this was a "bad" thing - a
few people agreed that they weren't crazy about indentation, but a
majority of people said they thought it was a good thing (readability,
studies of usability, etc.).  People have managed to write hundreds of
thousands of lines of useful and readable Python code using the
current scheme.

You've now expressed your views on the subject.  Now it is time for
you to assimilate the hard, cold fact that the use of indentation
isn't going to change no matter how much you whine and carry on.  Get
over it and move on.

Do some real programming in the language and decide whether or not
this is a language you want to continue using. If so, then the Python
community looks forward to any real contributions or suggestions you
have to improve the design of the language.  If not, then please do us
all a favor and post your protestations elsewhere.

--Paul



From anders.eriksson at morateknikutveckling.se  Sun Jan 16 10:07:55 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Sun, 16 Jan 2000 16:07:55 +0100
Subject: universal DB module (a la Perl's DBI/DBD)
References: <85qomm$91i$1@nnrp1.deja.com>
Message-ID: <5N2BOHLzIIugkLLw4VGmwkxvTHNR@4ax.com>

On Sat, 15 Jan 2000 21:26:58 GMT, Otis Gospodnetic <otis at my-deja.com>
wrote:

>Since the backend DB vendor is not yet known, I would love to use some
>modules that will make my code DB-independent

The SIG DB-API  specification for connecting to databases are
independent. Using this specification you can later select any module
that conform to the same.

ODBC would be an another way to go


// Anders
---
Enlish isn't my first or second language, 
so anything that you find offending is not there on 
purpose it's just due to the translation ;-)


From gvwilson at nevex.com  Sat Jan 15 09:32:19 2000
From: gvwilson at nevex.com (Greg Wilson)
Date: Sat, 15 Jan 2000 09:32:19 -0500
Subject: ANN: Open Source (Python) Design Competition
Message-ID: <388084F3.CD0EA1C1@nevex.com>

[Note: Python's own Guido van Rossum will be one of the judges,
and all implementations are to be written in (or at least scripted
with) Python!  The project coordinator, Greg Wilson, will be at
IPC8 in Washington with more news...]

Los Alamos National Laboratory                     CodeSourcery, LLC

                          Software Carpentry
                  http://www.software-carpentry.com

                    Open Source Design Competition

                         $100,000 in Prizes!

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

The Software Carpentry project is pleased to announce its first Open
Source design competition, with prizes totaling $100,000.  Students
and professionals from any country, working individually or in teams,
are invited to submit design outlines for:

* a platform inspection tool to replace autoconf;

* a dependency management tool to replace make;

* an issue tracking system to replace gnats and Bugzilla; and

* a unit and regression testing harness with the functionality of
  XUnit, Expect, and DejaGnu.

The best four entries in each category will be awarded $2500, and
invited to submit full designs by June 1, 2000.  The best design in
each category will then receive an additional $7500, while runners-up
will each receive $2500.  Once winning designs have been announced,
$200,000 will be available through open bidding for implementation,
testing, and documentation.

Participants may submit separate entries in one or more categories by
March 31, 2000.  Entries must be in English, and no more than 5000
words long.  For more information, see the Software Carpentry web site
at http://www.software-carpentry.com.  All of the project's work will
be Open Source; all tools will be written in, or scriptable with,
Python, and will be required to run on both Linux and Microsoft
Windows NT.

The competition will be judged by a panel that includes the following
noted software developers, authors, and computational scientists:

    Stephen Adler       (Brookhaven National Laboratory)
    Frank Alexander     (Los Alamos National Laboratory)
    Donnie Barnes       (Red Hat)
    Chris DiBona        (VA Linux)
    Paul Dubois         (Lawrence Livermore National Laboratory)
    Andrew Hunt         (Pragmatic Programmers, LLC)
    Stephen R. Lee      (Los Alamos National Laboratory)
    Josh MacDonald      (University of California, Berkeley)
    Brian Marick        (Reliable Software Technologies)
    Doug Mewhort        (Queen's University)
    Bruce Perens        (co-founder of the Open Source Initiative)
    Dave Thomas         (Pragmatic Programmers, LLC)
    Jon Udell           (author of Practical Internet Groupware)
    Guido van Rossum    (inventor of Python)
    Tom Van Vleck       (TransIlluminant)
    Phil Wadler         (Bell Labs)
    Scot Wingo          (AuctionRover)

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

The Software Carpentry project is sponsored by the Advanced Computing
Laboratory at the U.S. Department of Energy's Los Alamos National
Laboratory (http://www.acl.lanl.gov), and administered by
CodeSourcery, LLC (http://www.codesourcery.com).  The project's aim is
to encourage adoption of better software development practices by
making software tools easier to use, and by documenting design,
testing, and related activities.  For more information on the project,
or to let us know that you intend to submit a proposal, see
http://www.software-carpentry.com, or mail info at software-carpentry.com.




From esj at harvee.billerica.ma.us  Thu Jan 27 08:46:48 2000
From: esj at harvee.billerica.ma.us (Eric S. Johansson)
Date: 27 Jan 2000 13:46:48 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com> <86np80$jhp$1@harvee.billerica.ma.us> <86ntlg$2tc$1@news.ycc.yale.edu>
Message-ID: <86pi88$n3n$1@harvee.billerica.ma.us>

Paul Magwene <paul.magwene at yale.edu> wrote:

> Eric S. Johansson <esj at harvee.billerica.ma.us> wrote in message news:86np80
> [snip[
>> anyone trying to program using speech recognition.  Why would they
>> do this you might ask?  Probably because they were injured from
>> programming using keyboards.

> Yes, but isn't it possible to set the speech recognition software so when
> you say "indent" it tabs right, and when you say "dedent" it tabs left?

yes it is possible but then you lose the words "indent" and "dedent" 
for most uses.  One needs to be a little more careful when picking phrases
for command use.

But as it turns out, indent vs. outdent commands are among the more minor 
problems when programming by voice.  More major problems include navigation,
high precision spacing, lots of funny characters such as . & __ and of course
spelling.  Speech recognition is extremely good at spelling ordinary 
English words.  It gets seriously cranky when you try to force it to use
English pronunciations with non-English spellings.

a project working on developing speech driven programming technologies
can be found at: http://ai.iit.nrc.ca/II_public/VoiceCode/

--- eric

-- 
Eric S. Johansson    esj at inguide.com	esj at harvee.billerica.ma.us
This message was composed almost entirely using NaturallySpeaking


From kc5tja at garnet.armored.net  Mon Jan 10 16:30:57 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 21:30:57 GMT
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu> <3877E637.D3C9CC11@callware.com>
Message-ID: <slrn87kk90.930.kc5tja@garnet.armored.net>

In article <3877E637.D3C9CC11 at callware.com>, Ivan Van Laningham wrote:
>the engine use two's complement arithmetic.  For example, to get a
>32-bit 0xFFFFFFFF, we had to assign -1 to each individual byte.  Other
>opcodes were more complex.

To be fair, Python isn't without its numerical faults either.  For example:

>>> data = string.atol( stringArgument[0:8] )
>>> print "%08lX" % (data)

will fault on any integral value greater than 0x7FFFFFFF.  Why?  Because
Python just plain can't understand the concept of an unsigned integer.  So,
I had to cheat:

>>> print "%04lX%04lX" % ( data >> 16, data & 0xFFFF )

So much for portability and convenience.  :)

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From gmcm at hypernet.com  Fri Jan 14 10:31:47 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 14 Jan 2000 10:31:47 -0500
Subject: Python native compiler on NT
In-Reply-To: <85mjrp$ci5$1@nnrp1.deja.com>
Message-ID: <1264258586-9681410@hypernet.com>

Ben wrote:

> Thanks for the info. I tried the "Installer". However, to run
> the "compiled" .exe file, say xyz.exe, you have to copy xyz.exe and
> zlib.pyd to current directory before you can run it. Some how putting
> xyz.exe and zlib.pyd in a directory in the PATH, and run xyz.exe from
> another directory will have a error message something like:
> 
>    cannot load zlib.pyd
> 
> Do you know what's the problem?

In a very early release, it only worked when the exe was in the 
current directory. However, for some time the python path has 
been constructed from the exe's directory, (unless you're 
running from R/O media, in which case things get extracted to 
the temp dir and run from there). All pyds and dlls do have to 
be in the exe's directory (Standalone and Simple) or in the 
attached archive (Freeze - this is the only one that will work 
from a CDROM).

The help page has instructions for debugging, or I can usually 
figure it out from Builder.log and the (exact) error message.
 


- Gordon



From Helka.Folch at Wanadoo.fr  Sat Jan  8 06:57:15 2000
From: Helka.Folch at Wanadoo.fr (Helka Folch)
Date: Sat, 8 Jan 2000 12:57:15 +0100
Subject: clustering with Numeric Python
Message-ID: <857aie$lgd$1@wanadoo.fr>

Is there any implementation of a clustering algorithm with NumPy available?
Thanks
Helka Folch




From tseaver at starbase.neosoft.com  Sat Jan 29 10:04:07 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 29 Jan 2000 09:04:07 -0600
Subject: Case Sensitivity and Learnability
References: <B4191059635E81B1.8AFDD386EFACF4A3.6FF161EA0151B5EB@lp.airnews.net> <86u4uf$26lq$1@thoth.cts.com>
Message-ID: <0D57817D21B70542.AF8CFC08182C8F9C.D6594BC90A625E69@lp.airnews.net>

In article <86u4uf$26lq$1 at thoth.cts.com>, Will Rose  <cwr at crash.cts.com> wrote:
>Tres Seaver <tseaver at starbase.neosoft.com> wrote:
>: Thinking about the notion of case sensitivity as a barrier to CP4E, as
>: advocated quite glibly at IPC8, I typed up this module on the plane to
>: explore the consequences of allowing/encouraging users to enter case-
>: mangled code, especially in non-interactive uses (I think that distinc-
>: tion may be important, because readability is almost insignificatn in
>: an interactive session).
>
>
>Your arguments are reasonable, but they are arguments against systems
>which are both case-insensitive _and_ case-preserving.  The trouble with
>CI/CP is that the compiler sees a different set of tokens from those seen
>by the programmer, and it's sometimes difficult for the programmer to
>realise this.  If unskilled programmers find case too difficult or
>distracting, the cure is to force a single case, either upper or lower. 
>This could be done by an (optional?) compiler switch.

One project I have done required modifying a large body (60,000+ lines) of
case-sloppy ObjectPascal (a CI/CP language).  I tried a couple of tweaks
to aid readability, and flattening the case was the best of a bad
lot.  In general, I don't believe that "case flattened" code is more
readable than code which uses case "properly", although case-flattening
does reduce certain classes of typing errors.

The fix which I think would best help the novice typing at an interactive
prompt involves incremental compilation, coupled with an heuristic which
"suggests" the properly cased identifier (a la spell checking).

I do think that measuring readability for novices of case-mangled vs.
case-flattened vs. case-enlightened code is feasible;  I also believe that
it would be a great deal simpler than trying to measure "usability" for
those who use the language beyond the initial learning curve.  I would also
like to see data which addresses my hypothesis that learning to read code
is an essential prerequisite to achieving competency in the language.


-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From robin at alldunn.com  Fri Jan 21 19:51:36 2000
From: robin at alldunn.com (Robin Dunn)
Date: Fri, 21 Jan 2000 16:51:36 -0800
Subject: Screen depth
References: <2B1262E83448D211AE4B00A0C9D61B03BA707B@MSGEURO1>
Message-ID: <I97i4.52$wg.8192369@news.randori.com>

You can use wxDisplayDepth().


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!


"Pieter Claerhout" <PClaerhout at CREO.BE> wrote in message
news:2B1262E83448D211AE4B00A0C9D61B03BA707B at MSGEURO1...
> I'm using wxPython...
>
> Pieter
>
> -----Original Message-----
> From: Fredrik Lundh [mailto:fredrik at pythonware.com]
> Sent: Friday, January 21, 2000 1:17 PM
> To: Pieter Claerhout
> Cc: Python Mailing List
> Subject: Re: Screen depth
>
>
> Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> > I'm working on Windows NT and 95/98.
>
> using Tkinter?  wxPython?  gtk?  PythonWin?
>
> for Tkinter, the answer is here:
>
http://www.pythonware.com/library/tkinter/introduction/basic-widget-methods.h
tm
> look for winfo_depth under "window related information"
>
> (or you can use an undocumented feature in PIL:
> Image.core.display_mode() returns a tuple with mode
> and size).
>
> </F>
>




From arcege at shore.net  Wed Jan 19 08:47:59 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 19 Jan 2000 13:47:59 GMT
Subject: Best way to work in debug code?
References: <85umq4$tlr$1@nnrp1.deja.com>
Message-ID: <jgjh4.3895$Ef6.1033552@news.shore.net>

rdudfield at my-deja.com wrote:
: Just wondering how you people write debug code in your functions
: /methods?

: I personally write a "if (DEBUG_LEVEL >= 5):" around the code.  However
: this reduces readability.

: To fix the readability problem I usually delete debug code all the
: time.  Only having to rewrite it again ( and debug the debug code ).

: I sometimes use cvs to find that old debug code, but that's time
: consumming.

About 90-95% of the debugging I want to do is printing data, so most of
the statements would be just print's after the conditional.  Also,
"good debugging code" can sometimes add to the readability, just like
assert statements.

I have a module that contains a singleton class, which installs a
callable instance into the __builtin__ module.  The reason it is a
class instance and a module (which for this is a little redundant) is
because I used to write the short class into the code itself.  The class
looks something like:

import __builtin__
class Debug:
  from sys import stderr
  from string import join
  state, file = 0, stderr
  del stderr
  def on(self): self.state = 1
  def off(self): self.state = 0
  def __nonzero__(self): return self.state
  if __DEBUG__:
    def __call__(self, *args):
      if self.state:
        self.file.write(self.join(map(str, args))+'\n')
  else:
    __call__ = lambda *args: None
__builtin__.debug = Debug()
del __builtin__, Debug      # cleanup

Adding debug levels is as simple as changing the state and modifying
the __call__ method to handle that.

For my own usability, I also add:
import os
if os.path.exists('.debug'): debug.on()

If the file exists, I get debugging output - no change in code.

: The way that would be perfect ( for me ) would be for the editor I use(
: jed ) to be able to toggle on and off the viewing of debug code, at
: different levels of course.  Then have the python interpreter skip over
: the debug code, or optimise it out.  One way to do this would be to do
: this:

:   for x in range(0,100):
:     doNonDebugstuff()
:   debug 5:
:     print len(listOfObjects)
:     writeGraphOfObjectsToImage(listOfObjects,"graph.png")
:   doMoreNonDebugStuff()

: Where debug 5 means "if (DEBUG_LEVEL <= 5):".

: You could then call the python interpreter with a debug flag or maybe
: call a builtin function to change the debug level.

: It would be good to be able to choose to debug at level 4 and below, or
: just level 4.

I used to use debug levels, but found that it is easier to get all the
output.  The more complex the project, the more bugs can creep into other
parts of the code, debug levels can sometimes hide that creep.

: Additions to this might include using strings as debug tags. Like:

: debug "functionName":
:   doDebugStuff()

: Or both.

: debug "functionName" 4:
:   doDebugStuff()

: Being able to tell the interpreter to run all debug code in a particular
: function/method/object would be good too.

Sounds like you could easily handle some of this with a preprocessor.
For readability, I would not use "statement" syntax, but would use
"expression" syntax.  But that's just me. :)

  -Arcege



From cwr at crash.cts.com  Mon Jan 31 00:08:49 2000
From: cwr at crash.cts.com (Will Rose)
Date: 31 Jan 2000 05:08:49 GMT
Subject: Case Sensitivity and Learnability
References: <B4191059635E81B1.8AFDD386EFACF4A3.6FF161EA0151B5EB@lp.airnews.net> <86u4uf$26lq$1@thoth.cts.com> <86uoo7$1a4$1@vvs.superst.iae.nl>
Message-ID: <8735d1$ulf$1@thoth.cts.com>

Carel Fellinger <cfelling at iae.nl> wrote:
: Will Rose <cwr at crash.cts.com> wrote:
:> Tres Seaver <tseaver at starbase.neosoft.com> wrote:
:> : Thinking about the notion of case sensitivity as a barrier to CP4E, as
: ...
:> Your arguments are reasonable, but they are arguments against systems
:> which are both case-insensitive _and_ case-preserving.  The trouble with
:> CI/CP is that the compiler sees a different set of tokens from those seen
:> by the programmer, and it's sometimes difficult for the programmer to

: I don't think he is refering to this part of the problem. Swapping case at
: random *on its own* makes the source hard to read, let alone this confusing
: of what you see and what the compiler sees.

:> realise this.  If unskilled programmers find case too difficult or
:> distracting, the cure is to force a single case, either upper or lower. 
:> This could be done by an (optional?) compiler switch.

: this would make it impossible to import library modules that do depend on
: case distinction. Or would it be a per module switch, settable in the source?
: Now that will improve things:)

: I, for one, vehemently dislike case insensitive languages, as they force me
: to use ugly naming conventions like:

:   class tv_class: pass 
:   tv = tv_class()

: yack, instead of the much clearer:

:   class TV: pass
:   tv = TV()

: Admitted you have to know this convention of using case to distinct classes
: from variables:), but then in natural language we use this same trick to
: distinguish persons from objects like in Bill and bill. So how hard can it
: be for a newbie:) OTOH, I'm easily convinced by proper usability studies.

Matt Conway's Alice study argued that the use of this sort of convention
meant that the underlying concepts had to be explained to new programmers
so that they could follow it.  Since the aim of Alice was ease of use for
new programmers, non-case sensitivity (and possibly single case) seemed a
better bet.  (I disagree with this conclusion, btw.  I think it just stores
up trouble down the road.)


Will
cwr at crash.cts.com




From andrew at one.net.au  Wed Jan 26 23:09:53 2000
From: andrew at one.net.au (Andrew Maizels)
Date: Thu, 27 Jan 2000 14:09:53 +1000
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
References: <816010E2456BD111A48700805FBBE2EE01C60408@ex-quebec-u1.baan.com>
Message-ID: <388FC511.C776341A@one.net.au>

Gaetan Corneau wrote:
> 
> I'm really tempted to hack this guy's machine.
> Destroy everything.
> Or at least bust his mailbox.
> 
> The kind of spammer I hate the most.
> Sorry, I know this is off-topic.

Just make sure you get the right target.

Darren Mulligan is a spam fighter.  It seems he has pissed off a
spammer, and this is said spammer's method of revenge.

I wouldn't mind seeing the *spammer* hauled off to jail for a long time.

Andrew.
-- 
                    Biting weevils by moonlight,      
     \\\\\          Eating bugs by daylight,
    \\\\\\\__.      Never running from a food fight,
____\\\\\\\'/_______She is the one named Sailor Hedgehog!


From jorn.h.b.dahl at stud.jbi.hioslo.no  Mon Jan 31 05:08:12 2000
From: jorn.h.b.dahl at stud.jbi.hioslo.no (Jørn Helge B. Dahl)
Date: 31 Jan 2000 11:08:12 +0100
Subject: Why Perl do But Python can't?
References: <200001310128.UAA02858@poverty.bloomington.in.us> <8737c1$76d$1@news.cz.js.cn>
Message-ID: <m3wvoqeepf.fsf@pc-432.p52s.hioslo.no>

* "????" <jfzhou at 188.net>

> But How I do it to solve the follow quertion:

It is written qwertyon (sorry, could't help it. No offense :-))

> 1:through the WWW to manage user,the CGI need root uid.
> 2:how to read or write files that need some privilege ie. root ,admin etc...

Write a C wrapper program (or another compiled language) that has the
right privileges (setuid, setgid, or whatever) and call the script
(which has no privs) from there.

> some one tell me
> 1:use another programme to communite with CGI useing socket,but it is
> complex.
> 2:use expect mod ,but python dosn't have.

I don't know.
-- 
This goes to eleven


From super_banane at yahoo.com  Fri Jan 28 10:20:06 2000
From: super_banane at yahoo.com (super_banane at yahoo.com)
Date: Fri, 28 Jan 2000 15:20:06 GMT
Subject: vim and tabs in Python code
Message-ID: <86sc2v$8nk$1@nnrp1.deja.com>

When I want to modify somebody else's Python code with
vim, I run into problems quite easily. Say the previous
author has used blanks for indentation instead of
tabs as I do, and the following code gets a syntax
error:

<blank><blank><blank><blank>FirstLineOfCode
<one 4-blank tabulation>SecondLineOfCode

both look like:

    FirstLineOfCode
    SecondLineOfCode

The first line was initially in the file, the second
one was added by me. Since vim adds tabulations by
default, how is it possible to add blanks instead?

I guess that will not solve it: say that now my vim
configuration is adding blanks. Now I am editing somebody
else's code who uses tabulations, or a mix between
tabulations and blanks, how do I add a line without
first doing surgery on the file to check out what
the indentation convention was (if there is any, that is)?

I guess the whole problem comes from the fact that:

- The indentation is seen by the interpreter but not
  by me, e.g. a 4-blank tab is seen the same as 4 blanks
  but not interpreted the same. Hard to tell something
  invisible is different from something else invisible.
- Whatever intelligence you put in vim, will not help
  other kinds of editors, will it?

I would appreciate, though, if anyone could give an
advice about how to configure vim or preprocess
somebody else's code to make it fit any given convention
for indentation.

--
Nicolas


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


From michael at camelot-it.com  Sun Jan 23 05:09:40 2000
From: michael at camelot-it.com (Michael Tiomkin)
Date: Sun, 23 Jan 2000 12:09:40 +0200
Subject: Bug in Calendar Module
References: <22E71DAEC504D111B78100805FFE9DC72F03CAD4@pfs21.ex.nus.edu.sg> <yttaelz2ldd.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <388AD363.1D96FB45@camelot-it.com>

  Unfortunately, this is a bug.  The overflow exists only when you treat the time as a signed 32-bit
integer.  When you use time as unsigned integer, this will happen only around 1970+2**32/3e7, 2090?
It seems that the calendar time functions can have a workaround for this bug, in spite of the fact
that Python uses only signed integers.

  Michael

Ben Gertzfield wrote:

> >>>>> "Liew" == Liew Pol Yee <polyp at cyberway.com.sg> writes:
>
>     Liew> Hello! Does a fix for the Calendar Module calendar.py exist?
>     Liew> Currently, i get errors when I do the following: import
>     Liew> calendar new = calendar.monthcalendar(2038,2)
>
>     Liew> It seems to be unable to accept dates after January,
>     Liew> 2038. Thanks for any help!
>
> This is because you are on a system where the time is stored as a
> 32-bit integer, as the number of seconds since 1970.
>
> In 2038, the number of seconds since 1970 overflows the 32-bit limit,
> and a lot of programs will break.
>
> There isn't a good solution, except to use a system that has a 64-bit
> counter for time.
>
> Ben
>
> --
> Brought to you by the letters F and Q and the number 5.
> "It makes my nipples hard!"
> Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From king at calibre-dd.com  Wed Jan  5 14:16:22 2000
From: king at calibre-dd.com (Stephen J King)
Date: Wed, 05 Jan 2000 14:16:22 -0500
Subject: http POSTing
References: <8EB19A9E7newshoesmailcom@205.147.63.94> <OOJc4.245$B5.4746@news.more.net>
Message-ID: <38739886.EB1A06B1@calibre-dd.com>

Jeff Kunce wrote:
> 
> > i'm trying to create an offline form i can send to a website
> > through "POST". being new to python and minimal knowledge of
> 
> For a simple example, see
>     urlpost.py
> at
>     http://starship.python.net/crew/jjkunce/

A very nice example, but perhaps not as simple as it might be?

The requirement is stated:
	<FORM ACTION="http://www.test.com/form.html" METHOD="POST">
	<INPUT TYPE="hidden" NAME="hidd" VAL="en">
	<INPUT TYPE="text" NAME="email" VALUE="you at yourmail.com">
	<INPUT TYPE="image" SRC=post.gif NAME="POST">
	</FORM>
Although I don't know if you can post to an html page ...

Here's what I threw together:

#!/usr/local/bin/python

from socket import *

msg = 'hidd=en&email=you at yourmail.com'

ns = socket(AF_INET, SOCK_STREAM)
ns.connect('www.test.com', 80)
ns.send('POST /form.html HTTP/1.0\n')
ns.send('Accept: text/plain\n')
ns.send('Accept: text/html\n')
ns.send('User-Agent: stimpy\n')
ns.send('Content-type: application/x-www-form-urlencoded\n')
ns.send('Content-length: %d\n' % len(msg))
ns.send('\n')
ns.send(msg)

while 1:
	answer = ns.recv(1024)
	if not answer:
		break
	print answer,
ns.close()

# end

Assuming that you have the tkinter form you mention.


--  Stephen J King    Technology Manager   Calibre Digital Pictures  
--  king at calibre-dd.com               http://www.calibredigital.com


From jody at sccsi.com  Tue Jan  4 07:49:20 2000
From: jody at sccsi.com (Jody Winston)
Date: 04 Jan 2000 06:49:20 -0600
Subject: DB API
References: <3873F847@MailAndNews.com>
Message-ID: <wkogb2uhm7.fsf@sccsi.com>

The Python db-sig (http://www.python.org/sigs/db-sig/) has defined an
API to interface with several different data stores.  A listing of the 
different data stores can be found on
http://www.python.org/topics/database/modules.html.  

-- 
Jody Winston


From alex at magenta.com  Thu Jan  6 06:31:23 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 6 Jan 2000 12:31:23 +0100
Subject: lazy (?) evaluation, request for advice
References: <000701bf57e5$9aa20560$b1a0143f@tim>
Message-ID: <006001bf5839$ade66180$c02b2bc1@martelli>

Tim Peters:

> > ... I'd much rather let the user write the deal-selection
> > function in Python, than in either Tcl, or any custom,
> > ad-hoc "little language".
>
> Against that, you invented ad-hoc semantics as you went along;

Did I?  I used, e.g., spades to mean "the value of a variable named
`spades' in the current scope[s]".  Isn't that normal Python semantics?

What else are the local and global scope arguments to eval/exec for,
if not to let the controlling program determine the current scope[s]
for code being eval'd/exec'd?  Isn't that normal semantics too?

Whether in a given context the `spades' variable contains the number
of spade cards in a player's hand, the number of shovels in a pile of tools,
or the number of 3-year-old stags in the stables, well, yes, that IS
"ad-hoc semantics" -- as is, case by case, the identification of, what
hand, what pile, what stables we're talking about ("the current one",
"the one of interest").  If there is a need to examine multiple stables,
or piles, or hands, then in such contexts using "x.spades" for some
"x" will surely be better -- but you appear to be saying that, even in
contexts where there is only one pile/stables/hand of interest, using
"x.spades" for some one arbitrary "x" is "less ad-hoc"?  I don't see why.

> the only connection to Python appeared to be syntactic.

?"only"?  You call, e.g., the semantics of short-circuit logic a "syntactic"
issue?!  I would appreciate understanding how you define "syntactic" in
this context to justify this extremely peculiar assertion.

> Let the user write real
> Python!  Then as the program gets fancier they'll have the full language
to
> exploit; with "spades" etc as magical global names, what chance is there
to
> extend the scheme to a user-defined function that e.g. wants to examine
> *two* hands at a time?  (I know! "spades1" and "spades2" <wink>)

Where more than one hand is of interest, object notation will be used;
and where more power is needed, statements will be used rather than
just an expression.  The fact that a statement suite is more powerful
(because, e.g., assignments can be done, if/else used rather than just
short-circuit logic for flow control, loops can be variously expressed
rather than just squeezed into map/reduce kinds of molds, etc), is no
reason to outlaw using just an expression where an expression will
suffice; and similarly for object vs plain notation.  Surely "a>2" is
"real Python", just as "x.a>2", or "for b in a: splotz(b)", etc -- a simpler
case, but no less real than the less-simple ones!


> def userfunc(x):  # they get to make up the argument name
>     ... (x.spades >= 6 or x.hearts >= 6) ... etc
>
> Provided your users can pass a mirror test (i.e., you hold a mirror under
> their nose, and they pass if it fogs up <wink>), they can learn this.
After

They can learn, and they will surely appreciate being able to use
more powerful constructs (functions, statements, object notation,
etc) _when needed_ -- but surely will not appreciate having to
use them, willy-nilly, even when their needs are much simpler.

> which, playing __getattr__ tricks with the x object gives you the "lazy"
> behavior you want.

Yep, I said that -- if object notation is used, then lazy evaluation can
be had through __getattr__.  I also said that object notation can be
avoided (for simple cases) if lazy evaluation is not needed.  I would
prefer to offer my users programming interfaces with the "right"
amount of power/complexity, based on my analysis of "right", rather
than making such an interface decision solely on the basis of hassle
of implementation of a "third way" combining simple notation and
lazy evaluation; letting implementation problems constrain interfaces
is not a good general strategy.


> > I've thought about doing the eval repeatedly, within
> > a try/except block to catch the NameError's, inserting
> > the missing names in the dictionary one by one, but
> > I'm suspicious about the performance characteristics
> > of such an approach.
>
> You should be more suspicious of the sheer convolution <0.5 wink>.
>
> Anyway, what's expensive there isn't so much the try/except, but the
> repeated eval *if* the user's expression is carried around as a string
(eval
> will recompile it to byte code each time).  If you do go this route, see
the
> builtin "compile" function for a way to do the compilation to byte code
just
> once (eval accepts compiled code objects too).

Yes, I'm aware of that, although I failed to mention it; I do plan to
compile the user-supplied text before running it over thousands of
candidate deals.

> Note also that if your users
> truly are morons, there's nothing to stop *you* from marching over their
> string and replacing e.g. "spades" with "current_hand.spades" before
> compilation.

Is there a way I could use the existing parsing tools to help me
with this?  I've read through the library reference, but I still find
that part a tad obscure. 3.14.6.1 and 3.14.6.2 are the parts I
am thinking about.  If I could examine in the AST the variable
names used in the expression, matching them with the dictionary
of the hand features I know how to evaluate, I could pre-populate
the dictionary to be used as the scope in the eval; or, I guess,
change the variables as you suggest, to get even lazier (for those
cases where short-circuit logic lets me evaluate a certain feature
only in a subset of cases).  But the logic for examining, and even
more for modifying, the AST, appears to be somewhat complex.
Are there more online examples that I could use?

Or, were you suggesting a similar operation on a purely lexical
level?  E.g., with tokenize?  Or, a string-manipulation approach?

I will appreciate further advice about this, and tolerate the
sneers and name-calling that seem to go with the advice
(although rarely have I seen discussion quite as sneering as
this on this group -- reminds me a lot of the ambiance on
the Perl group -- this one had so far seemed friendlier, though
I guess that may have been an artefact due to a frequentation
started more recently).


> treat-users-like-grownups-and-they-won't-demand-you-give-
>     them-an-allowance<wink>-ly y'rs  - tim

Grown-ups, too, appreciate convenience -- which, inter alia,
means "appropriate level[s] of simplicity/power trade-off[s]".

Rather than offering just one (rich, powerful, complex) way
to express deal selection, I would prefer to offer users the
convenience of choosing different points along the curve of
power vs simplicity according to current needs -- from just
clicking a few radiobuttons on a GUI, to entering a plain and
simple selection expression about one hand, all the way to
subclassing the deal-generator class to tweak and customize
things to the fullest.

The "plain and simple selection expression" appears to me
(from reasonably vast previous experience with this specific
application domain) to be a rather "key" point along the
trade-offs curve, one which will be very often "visited" for
many kinds of simple deal-generation needs.  Therefore, I
judge it worthwhile to expend some thought, and work, in
optimizing usability and convenience at that point in the
curve.


Alex





From nnever50 at yahoo.com  Wed Jan 19 09:49:39 2000
From: nnever50 at yahoo.com (nathan)
Date: Wed, 19 Jan 2000 15:49:39 +0100
Subject: tkinter and stdout redirection
Message-ID: <864iv0$r6o$1@news1.sunrise.ch>

Hi all,
newbie here..

if I redirect the output stream in a file with

sys.stdout = open ("file", "w")

direct from the command prompt, I get the output  stream in my file without
any problems.

But if I do the same from a tkinter application , I get the output stream
only when I close the application.
I think  that there's a buffer between the output stream and my file, but
I've not found anywhere how to set it.
Someone can help me?
Thank you

N.





From gerrit.holl at pobox.com  Wed Jan 26 16:16:06 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 26 Jan 2000 22:16:06 +0100
Subject: Tuples
Message-ID: <20000126221606.A962@stopcontact.palga.uucp>

Hello,

With tuples, some things aren't really Pythonic. In some cases, even
the Perl-like TMTOWTDI thing is true!
a = {}
a[1,2] = 'f'

is equivalent to:
a = {}
a=[(1,2)] = 'f'

And:
"aaa %s bbb" % 'foo'

is equivalent to:
"aaa %s bbb" % ('foo',)

Why's that? TMTOWTDI does not improve readability...

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From malcolmt at smart.net.au  Wed Jan  5 07:36:39 2000
From: malcolmt at smart.net.au (Malcolm Tredinnick)
Date: Wed, 5 Jan 2000 23:36:39 +1100
Subject: Sydney Pythonistas/Pythoneers (was Re: Sydney scoreboard?)
In-Reply-To: <3872CEA2.38C8E5A0@bby.com.au>; from Sarah Burke on Wed, Jan 05, 2000 at 02:55:02PM +1000
References: <385DB48C.7A7A52D6@bby.com.au> <8On74.6414$Dh3.85256@ozemail.com.au> <YCU74.7437$NV6.31360@news-server.bigpond.net.au> <38601AF6.7D39F091@bby.com.au> <D0da4.3285$lv5.19947@news-server.bigpond.net.au> <3872CEA2.38C8E5A0@bby.com.au>
Message-ID: <20000105233639.A1855@Ridcully.home>

On Wed, Jan 05, 2000 at 02:55:02PM +1000, Sarah Burke wrote:
> I would be very interested in doing this.
> Would people be interested in a more formal or less formal gathering, do
> you think? (I am Sydney based by the way.)
> Does anyone have any idea of a rough number of python users in Sydney?
> 

Let me volunteer to do something:
	Anybody from the greater Sydney area who would like to get together
	for a Python-type gathering is welcome to send me private email and
	I'll coordinate something. Let me know your preference for either:
	
	(A) Restaurant meal somewhere (cheapish, probably)

	(B) Picnic (Saturday afternoon or whatever)

	(C) Something more formal (meeting with presentations, etc)

	(D) Other? (my brain is on the fritz at the moment, so I've run out of
	options)

In a few days I'll see where the preferences lie and organise the details.
I guess I'll also post this to the mailing list Mark Hammond mentioned in an
earlier post, just in case they don't follow these discussions.

Cheers,
Malcolm Tredinnick

--
Borrow from a pessimist. They don't expect it back.




From m.faassen at vet.uu.nl  Wed Jan  5 09:06:28 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Jan 2000 14:06:28 GMT
Subject: Python Database Connectivity
References: <386AA8C1.EAD7DD3B@exceptionalminds.com>
Message-ID: <84vj54$pjk$1@newshost.accu.uu.nl>

Timothy Grant <tjg at avalongroup.net> wrote:
> Hi,

> Just wondering if anyone has used Python on Windows to connect to an
> ODBC database?

The PythonWin extensions contain an ODBC package that conforms (last
I checked) to the Python DB spec version 1. So if you have installed
those you can use ODBC; a basic example:

import odbc, dbi

db = odbc.odbc("databasename")
cur = db.cursor()
sql = "select * from foo"
cur.execute(sql)
data = cur.fetchall()
print data
cur.close()
db.close()

I imagine mxODBC works much like that too (though it may implement the DB
spec version 2?).

Regards,

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


From fdrake at acm.org  Fri Jan 21 11:54:58 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 21 Jan 2000 11:54:58 -0500 (EST)
Subject: SUGGEST: os.getuser()
In-Reply-To: <20000121170432.B843@stopcontact.palga.uucp>
References: <20000116174004.A5995@stopcontact.palga.uucp>
	<000b01bf6049$36532140$f29b12c2@secret.pythonware.com>
	<20000116195710.A7700@stopcontact.palga.uucp>
	<Jylh4.3912$Ef6.1036319@news.shore.net>
	<20000121170432.B843@stopcontact.palga.uucp>
Message-ID: <14472.36706.632509.23381@weyr.cnri.reston.va.us>

Gerrit Holl writes:
 > So getpwuid() is never safe? But in my program, I *also* need the
 > comment field. So if there are multiple uids the same, I can't get
 > the full name?

  It sounds like you should use os.getlogin() (in the CVS version) to
get the actual login name and perform the lookup that way.  To be
*really* safe, use both getlogin() *and* getuid().  ;)  Seriously,
there should be only one line in /etc/passwd that matches the result
of os.getlogin(), but the existing getpass.getuser() should be fine
for most everyone.
  Perhaps *that* should be modified to use getlogin() when available,
in preference to pwd.getpwuid(os.getuid())[0] ?


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From aahz at netcom.com  Sat Jan  8 10:18:07 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jan 2000 15:18:07 GMT
Subject: Error using string.strip
References: <38775227.2CC0D685@worldnet.att.net>
Message-ID: <857kff$28i$1@nntp9.atl.mindspring.net>

In article <38775227.2CC0D685 at worldnet.att.net>,
Eric Hausig  <codeslave at worldnet.att.net> wrote:
>
>I'm very new to Python.  (In fact, I'm walking through the tuturial
>right now!)

Congratulations!

>I'm getting the following message:
>
>>>> string.strip('str') + 'ing'
>Traceback (innermost last):
>  File "<stdin>", line 1, in ?
>NameError: string
>>>>

Try "import string"; you have to import modules before you can use
them.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From vincent_berg at hotmail.com  Sun Jan 23 08:03:27 2000
From: vincent_berg at hotmail.com (Vinnieboy)
Date: Sun, 23 Jan 2000 13:03:27 -0000
Subject: CGI?
Message-ID: <OtXZUoZZ$GA.275@net025s>

Hi,

Does anyone know where I can find more about, Python-programming and
especcially cgi-script!

Greetz from Vinnieboy




From gerrit.holl at pobox.com  Sun Jan  9 15:44:20 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 9 Jan 2000 21:44:20 +0100
Subject: Python Host
In-Reply-To: <bJ3e4.8$78.764@news.uswest.net>; from shawnball@uswest.net on Sun, Jan 09, 2000 at 12:04:39PM -0600
References: <bJ3e4.8$78.764@news.uswest.net>
Message-ID: <20000109214420.A2802@stopcontact.palga.uucp>

Andrew Ferk wrote on 947415879:
> Are there any web hosts that support Python?

For personal and commercial use:
www.vex.net, I think.

For things maarten at nl.linux.org likes:
www.nl.linux.org

mainly opensource projects and community websites about GNU/Linux,
Unix, opensource. Mail me for more information.

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----
If you think Linux should get more hardware drivers, _please_ sign this:
http://www.libranet.com/petition.html



From nic at csee.uq.edu.au  Sun Jan 30 19:10:17 2000
From: nic at csee.uq.edu.au (Nic Williams)
Date: Mon, 31 Jan 2000 00:10:17 GMT
Subject: Dynamic class construction?
References: <389309F7.4A35371F@math.okstate.edu> <8707c0$ua2$1@nnrp1.deja.com> <38948F0E.36AA2548@math.okstate.edu>
Message-ID: <872jt9$gq4$1@nnrp1.deja.com>

In article <38948F0E.36AA2548 at math.okstate.edu>,
  "David C. Ullrich" <ullrich at math.okstate.edu> wrote:
>
[cut]
>     I don't know that I want to - I was wondering if I needed to.
> Couldn't hurt.

Class versioning might be one reason: many classes with the same
internal name that represent different versions of a conceptually single
class. Instead of storing them by individual external name, perhaps a
list could be used.

Account=[newClass('Account',...), newClass('Account',...), ...]
# and a version of the Account class is obtained by indexing the
# global Account list
account0=Account[0](<parameters of constructor>)
account1=Account[1](...)
account1.__name__ == account0.__name__
1

Or whatever. If it makes some sense to have the internal and external
names be different, go for it.
>
>     I take it that simply putting a "pass" in the inner
> class definition and then setting attributes _is_ the
> "right" thing? It wasn't the first thing I thought of,
> but it was the first thing that worked.

You could also generate a large string that represents the class
definition and then exec(...) the code, but that is just plain
silliness.

I think using pass, and then setting previously defined functions as
methods to the class is ok. Just use lots of comments :-)

Nic


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


From matt_winterroth at hotmail.com  Sun Jan 23 18:42:17 2000
From: matt_winterroth at hotmail.com (Matt Winterroth)
Date: Sun, 23 Jan 2000 23:42:17 GMT
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <3dzouke2v7.fsf@amarok.cnri.reston.va.us><200120002033497688%zippy@cs.brandeis.edu>
Message-ID: <B4B0FA22.196%matt_winterroth@hotmail.com>

lets not forget that lutz's "programming python" is quite old - dating back
to october of 1996...i'd much rather wait for a new edition to be printed
up, but i'm not holding my breath with o'reilly...i love them to death, but
how long did it take the java 1.2 nutshell book to come out?

another one of my favorite python books is "the quick python book" by
kenneth mcdonald, daryl d. harms...good tutorial, code examples and pros and
cons over other application and scripting languages...

hope this helps... 

> From: Patrick Tufts <zippy at cs.brandeis.edu>
> Organization: SBC Internet Services
> Newsgroups: comp.lang.python
> Date: Thu, 20 Jan 2000 20:33:49 -0800
> Subject: Re: Which Python book?
> 
> In article <3dzouke2v7.fsf at amarok.cnri.reston.va.us>, Andrew M.
> Kuchling <akuchlin at mems-exchange.org> wrote:
> 
>> Bala <engbpa at liquidinformation.com> writes:
>>> Python Essential Reference (OTHER NEW RIDERS)
>>> by David M. Beazley, Guido Van Rossum
>>> or
>>> Learning Python (Help for Programmers)
>>> by Mark Lutz, David Ascher, Frank Willison (Editor)
>> 
>> The Essential Reference is just what its title says: a reference
>> guide.  You'd have to be *really* motivated to learn Python from it.
> 
> 
> The Essential Reference is an excellent reference.  Get it anyway.  But
> you won't learn the language from it -- it's too terse for that.
> 
> Oh, and don't bother with "Programming Python" (also by Mark Lutz).
> It's not a good for learning or reference.  I've read reviews that say
> "Learning Python" is much better.
> 
> --Pat



From Alex.Martelli at think3.com  Fri Jan 21 11:32:15 2000
From: Alex.Martelli at think3.com (Alex Martelli)
Date: Fri, 21 Jan 2000 17:32:15 +0100
Subject: function "overloading"
Message-ID: <6D8A17398E28D3119F860090274DD7DB4B3DF2@pces.cadlab.it>

Janos Blazi writes:

: I have 3 classes called Tline, Tsegment and Tpoint, repectively. I'd like
to
: write a function "subset" that cheks if one of the instances of the above
: classes  is a subset of another instance (as long as this is sensible).
:
: I am afraid I cannot overload the name "subset" and additionally when
subset
: has control it cannot decide the type of its arguments: Does this mean I
: have to write several *different* subset functions having different names

The built-in isinstance() function will let you check whether an object is
an instance of a given class (or, a given type-object).  You could use
that, plus, ideally, some polymorphism in your above-mentioned classes, 
to get exactly what you want, it seems to me.

I.e., suppose you have:

class Tpoint:
	def liesAlongLine(self,line):
		# gives desired boolean result if line is a Tline
	def liesOnSegment(self,segm):
		# gives desired boolean result if segm is a Tsegment

class Tsegment:
	def oneExtreme(self):
		# returns a Tpoint that's one extreme of the segment
	def otherExtreme(self):
		# guess what it returns
	def hasPoint(self,point):
		return point.liesOnSegment(self)
	def containsSegment(self,otherSegment):
		return self.hasPoint(otherSegment.oneExtreme()) and
self.hasPoint(otherSegment.otherExtreme())

class Tline:
	def hasPoint(self,point):
		return point.liesAlongLine(self)
	def containsSegment(self,aSegment):
		return self.hasPoint(aSegment.oneExtreme()) and
self.hasPoint(aSegment.otherExtreme())


def subset(haystack,needle):
	if isinstance(needle,Tpoint):
		return haystack.hasPoint(needle)
	elif isinstance(needle,Tsegment):
		return haystack.containsSegment(needle)


Basically, you need to double-dispatch on the types of both haystack and
needle, and this approach is based on a type-test for needle plus some
polymorphism in the haystack.  Other approaches would clearly also be
possible; in particular, I have recently seen on this list (==newsgroup)
an announcement of a general purpose multiple-dispatch package -- I
have not looked at it further, but that might certainly be a way to solve
your
problem, for example.

A "trick" solution to this particular case might come from considering a
point equivalent to a segment where both extremes coincide (==return
the same point...), of course.  But this a pretty specific case:-).


Alex




From scarblac-spamtrap at pino.selwerd.cx  Mon Jan  3 19:37:26 2000
From: scarblac-spamtrap at pino.selwerd.cx (Remco Gerlich)
Date: 4 Jan 2000 00:37:26 GMT
Subject: Python Conference -- Early Bird Registration ends Jan 5!
References: <200001031737.MAA24342@eric.cnri.reston.va.us>
Message-ID: <slrn872g4o.39e.scarblac-spamtrap@flits104-37.flits.rug.nl>

Guido van Rossum wrote in comp.lang.python:
> Make the Python conference the first conference you attend in the new
> millennium!

But doesn't the new millennium staD^&**FGSF&())KJDFKDSSAACNSD28947

NO CARRIER

-- 
Remco Gerlich,  scarblac at pino.selwerd.cx
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From james+news at daa.com.au  Mon Jan 31 02:32:20 2000
From: james+news at daa.com.au (James Henstridge)
Date: Mon, 31 Jan 2000 07:32:20 GMT
Subject: Iterative argument conversion in C extensions
References: <389510DC.6C67B965@digicool.com>
Message-ID: <38953A3C.B04EA933@daa.com.au>

If all you want to do is get at the arguments directly, remember that
the arguments are passed in as a tuple, so you can use the
PyTuple_GetItem function to get the arguments, and use PyInt_Check,
PyString_Check, etc to check the type of an argument.

If you are trying to wrap a C varargs function, this is a little more
difficult.  Here is a bit of info that may help though:

There is no portable way to do this such that it will work on all
platforms (that I know of).  If you don't need platform independence,
you could find out about the calling convention of the platform you are
using and write a bit of assembly code to construct an argument vector
on the stack.  You would have to do this for all platforms you want to
use.  This is the sort of solution Netscape is using to allow javascript
to call XPCOM methods in mozilla.

There are other ways of getting this sort of behaviour that will work in
most cases for a limited number of argument types though.  Just create a
prototype something like this:
  typedef void (*lotsofargs) (sometype *non_varargs_arg,
                              void *arg1,
                              void *arg2,
                              ...
                              void *argn);

(n is sufficiently large to hold the maximum number of arguments you
will ever want to pass -- not quite varargs, but it will probably do).

For each string and int argument, cast it to (void *).  Handling doubles
is more difficult though.  On 32 bit architectures, they are usually
twice as big as a pointer, so they are more difficult to support.

One possible way to handle doubles (which I haven't tested) is with a
function something like this:
  void
  _split_double(void *arg1, void *arg2, void **ret1, void **ret2)
  {
    *ret1 = arg1;
    *ret2 = arg2;
  }
  typedef void (*split_double_type)(double arg, void **ret1, void
**ret2);
  split_double_type split_double = (split_double_type)_split_double;

On machines with 32 bit alignment and sizeof(double)==2*sizeof(void*),
this should split the double argument into two args you can pass to
lotsofargs.  You would probably have to do something different on 64 bit
architectures.

Finally, cast the varargs function to (lotsofargs).  Call the function
with the set of (void *) arguments you have put together.  Just pass
NULLs for the extra arguments.

The code would look something like this:

  void *args[N];
  int i = 0;
  lotsofargs func;

  for (arg in python_args) {
    if (PyString_Check(arg)) {
      if (i > N-1) raise TooManyArgsException;
      args[i++] = PyString_AsString(arg);
    } else if (PyInt_Check(arg)) {
      if (i > N-1) raise TooManyArgsException;
      args[i++] = (void *)PyInt_AsLong(arg);
    } else if (PyFloat_Check(arg)) {
      if (i > N-2) raise TooManyArgsException;
      split_double(PyFloat_AsDouble(arg), &args[i++], &args[i++]);
    } else
      raise TypeError;
  }
  func = (lotsofargs)real_func;
  func(non_varargs_arg, args[0], args[1], ..., args[N-1]);

I have done something like this before (without the double handling code
though), and it works fairly well.  It would be nice if there was some
standard `inverse varargs' function in standard C, but since there
isn't, you will probably have to use a hack like this.

James Henstridge.

Chris McDonough wrote:
> 
> Hi,
> 
> I'm working on a C extension that must accept arbitrary positional
> arguments of Python types string, int, and float and convert those to
> their C equivalents (int, char, float).
> 
> I'm sort of stuck, however, on creating a C routine which can accept and
> deconstruct into C equivalents the elements of the passed-in Python
> tuple that may contain arbitrary Python object types of int, float, and
> string arbitrarily positioned within the tuple.  It doesn't help that I
> don't know much C, either :-)
> 
> To explain a little better, I'd like to be able to call a function
> within a C extension with arbitrary arguments like so:
> 
> mymodule.function(1,"string",2.0)
> 
> or maybe
> 
> mymodule.function("string", "another string", 2.0, 1, "wow")
> 
> or maybe even
> 
> mymodule.function("lonely string")
> 
> I imagine the best thing to do here is to construct a wrapper for the
> function that determines the number of passed in arguments and their
> types.  The wrapper would then shove what it knows about the arguments
> into a dictionary of some kind which then would be shoved into the tuple
> that is passed to the C extension module.  But deconstructing that
> Python dictionary and retrieving and marshalling its contents within C
> is sort of beyond me.
> 
> Does anyone have an example of doing such a thing?
> 
> Many thanks in advance!
> 
> --
> Chris McDonough
> Digital Creations, Inc.
> Zope - http://www.zope.org


From b2blink at hotmail.com  Wed Jan 19 14:25:26 2000
From: b2blink at hotmail.com (Ulf Engstrøm)
Date: Wed, 19 Jan 2000 14:25:26 CET
Subject: ZOPE/XML
Message-ID: <20000119132526.80282.qmail@hotmail.com>

Hello..:)
I've made a text-editing py-module to my Zope-installation and I want it to 
return the text as 'pure' xml. When I try to do this however it'll return a 
text starting with
<html><head></head>
<?xml version="1.0"?>

How do I get rid of this? I need this to be returned with <?xml?> as a start 
since the program getting the return reads pure xml.
Best regards
Ulf
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com




From gchiaramonte at ibl.bm  Mon Jan 31 15:24:30 2000
From: gchiaramonte at ibl.bm (Gene Chiaramonte)
Date: Mon, 31 Jan 2000 16:24:30 -0400
Subject: Python and Codebase
In-Reply-To: <058AFBA1CD7F1621.791F88D00D532EC1.0128BF9429776D7B@lp.airnews.net>
Message-ID: <LPBBLPGHGGJMCPPOKONLMEKCCCAA.gchiaramonte@ibl.bm>

Does anyone know of a Python Module for Codebase from Sequiter?

Thanks,

Gene



From bsb at winnegan.de  Sun Jan  2 04:42:25 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 02 Jan 2000 10:42:25 +0100
Subject: comparing text strings
In-Reply-To: "NuclearZen"'s message of "Sun, 02 Jan 2000 08:55:34 GMT"
References: <aoEb4.1909$cV.20812@news.rdc1.wa.home.com>
Message-ID: <871z804xoe.fsf@baal.winnegan.de>

"NuclearZen" <NuclearZen at home.com> writes:

> this is probably the stupidest Q's of all time but  Idont know how to
> compare text strings in python.  How?  I try to do it like C, or perl but???

Since python is an interpreted language, you can easily experiment:

Python 1.5.2+ (#4, Nov 18 1999, 01:39:08)  [GCC egcs-2.91.66 19990314 (egcs-1.1.2 release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
IDLE 0.5 -- press F1 for help
>>> s1="I'm a string"
>>> s2="I'm a string"
>>> s3="I'm different"
>>> s1 == s1
1
>>> s1 == s2
1
>>> s1 == s3
0
>>> 

HIH
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******



From colin.kong at toronto.edu  Sat Jan 15 02:44:49 2000
From: colin.kong at toronto.edu (Colin Kong)
Date: Sat, 15 Jan 2000 07:44:49 GMT
Subject: Build mxODBC.MySQL into cgipython
Message-ID: <387FE36E.3FC51B6B@toronto.edu>

Hello,

Can anyone tell me how to incorporate additional
modules into cgipython?  Thank you.

Colin


From roy at popmail.med.nyu.edu  Fri Jan 21 09:51:04 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Fri, 21 Jan 2000 09:51:04 -0500
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <20000121124834.A10685@xs4all.nl>
Message-ID: <roy-3D90FF.09510421012000@netnews.nyu.edu>

Thomas Wouters <thomas at xs4all.net> wrote:
> Yeah, right. You dont honestly believe all of those languages are
> white-space tolerant ? In C (or C++), what does this mean:
> 
> y+++x;
> 
> Is that y++ + x, or y + ++x ? If you add whitespace, the compiler will do
> something different.

That's not what people are complaining about.  C uses whitespace as a token 
delimter.  It's needed when leaving it out would result in an ambigious parse, 
and if you use it, any arbitrary contigious string of whitespace characters 
makes up a single delimter.  There is never any time (outside of a quoted 
string) when substituting any non-null combination of spaces, tabs, and newlines 
with any other non-null combination of spaces, tabs, and newlines will alter the 
meaning of a program.

That's a far cry from confusion about how many spaces a tab is equal to leading 
to syntacticaly correct but semanticaly incorrect programs.

Even SQL, which has to be the most wart-infested language syntax I've ever seen, 
gets this right :-)

-- 
Roy Smith <roy at popmail.med.nyu.edu>
New York University School of Medicine



From pmullins at gatecom.com  Sun Jan 23 17:28:38 2000
From: pmullins at gatecom.com (Patrick H. Mullins)
Date: Sun, 23 Jan 2000 22:28:38 GMT
Subject: [Newbie] Python as a frontend to other apps.
References: <iJKi4.25111$pb2.1977836@tw11.nn.bcandid.com>
Message-ID: <pgLi4.25161$pb2.1983916@tw11.nn.bcandid.com>

> import os
> password="secretstuff"
> infile="test.idea"
> outfile="test.zip"
> os.system("idea -dk %(password) %(infile) %(outfile)")

Found out that I just had to move a few things around.  :-)

os.system("idea -dk %s %s %s" % (password, infile, outfile))

Thanks for the help!

Patrick Mullins
pmullins at gatecom.com




From skip at mojam.com  Wed Jan 26 23:50:10 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 26 Jan 2000 22:50:10 -0600 (CST)
Subject: Another newbie question
In-Reply-To: <_nOj4.894$yY6.13534@news.goodnet.com>
References: <000001bf651d$f05ac240$132d153f@tim>
	<_nOj4.894$yY6.13534@news.goodnet.com>
Message-ID: <14479.52866.523964.498964@beluga.mojam.com>

    Scott> I've read the tutorial and I don't get lambda.

Scott,

You're probably not alone. ;-)

The result of executing a lambda is nothing more than an unnamed function
object.  It has some rather peculiar syntactic limitations that make it
appear very odd to the untrained eye.

You can safely ignore lambda for the time being and just use named
functions.  If/when you need it, you'll probably realize it.

(Fred, maybe discussion of lambda doesn't belong in the tutorial...)

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098
"Languages that change by catering to the tastes of non-users tend
not to do so well." - Doug Landauer



From jasont at wellmed.com  Wed Jan 26 17:15:59 2000
From: jasont at wellmed.com (Jason Taylor)
Date: Wed, 26 Jan 2000 14:15:59 -0800
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim>
Message-ID: <MPG.12f911f129ce00ff989681@or.news.verio.net>

> > Under NT, it seems that starting up python to process even
> > a trivial script is pretty slow.
> 
> Under all installations of NT, or some specific one that you tried?  Don't
> laugh <wink>.  One thing to watch out for is fragmentation under NT.  Get a
> defrag utility (e.g., Diskkeeper Lite is free) and check that out.

First, thanks guys for the help!

I have checked a couple other installs and they *don't* seem to have the 
problem.  So there appears to be something fishy with my machine.  I 
tried all the suggestions posted so far: defragged, kept my machine 
idle, used -S and -X options to trim startup load.  Put python before 
perl in my PATH.  It's a little bit faster but nowhere near what I am 
now seeing from python on other NT machines.

Just to simplify, I uninstalled all the Python packages (Base, 
Win32/com, Tcl) and just installed py152.exe.  No change.  All the 
registry keys look fine, at least the obvious ones I know to check such 
as HKEY_LOCAL_MACHINE/SOFTWARE/Python and the .py association.

I will keep playing with it.  If anyone else has seen this, speak up so 
we can identify commonalities.

Thanks!!
Jason Taylor

  



From nemeth at iqsoft.hu  Wed Jan  5 16:39:27 2000
From: nemeth at iqsoft.hu (Nemeth Miklos)
Date: Wed, 05 Jan 2000 22:39:27 +0100
Subject: Python plugin
Message-ID: <3873BA0F.1568A63A@iqsoft.hu>

Is there a (webbrowser) plugin for executing Python scripts?
I'm having a lot of problem with Java, and I am seeking alternatives.

Miklos Nemeth






From R.Brodie at rl.ac.uk  Mon Jan 24 06:06:03 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Mon, 24 Jan 2000 11:06:03 -0000
Subject: Living without static member functions
Message-ID: <86hbmt$g8s@newton.cc.rl.ac.uk>

There's a programming quirk of mine that I haven't seen elsewhere,
although probably well-known to some of you. Here's a rough outline
of the code:

class Milkman:

    allMilkmen = {}

    def __init__(self, name):
        ...
        allMilkmen[name] = self

    def find(self, name):
        return(allMilkmen[name])

It's handy for debugging or garbage collecting like activities. Also
for distributed applications, where it's harder to keep object references
around.

Firstly, I'm curious if this is well-known, useful or boneheaded design
on my part.

Secondly, it doesn't work in Python, of course, because I need a
Milkman to find all the others (i.e. Milkman.find['Phil'] doesn't work).
I can do something similar by taking allMilkmen and find to module
scope; that seems kind of ugly, and worries me that I'm abusing the
language. Can I do better?

Thirdly, is there some neat way that I can find all members of a class
by inquiring of the Python internals?

Introspectively yours.
Richard Brodie.






From mvh at ix.netcom.com  Mon Jan 10 19:22:53 2000
From: mvh at ix.netcom.com (mvh at ix.netcom.com)
Date: Mon, 10 Jan 2000 16:22:53 -0800
Subject: Embedding python in custom hardware
Message-ID: <t4td58.2j.ln@netcom1.netcom.com>

Can I embed Python in some custom hardware where I have no OS, just
some C libs?  How small of a memory footprint can I strip Python down
to?  Can I then run imported code in a 'safe' environment?

Thanks,

Mike Harding



From sp00fD at yahoo.com  Tue Jan 11 17:15:10 2000
From: sp00fD at yahoo.com (sp00fD)
Date: Tue, 11 Jan 2000 22:15:10 GMT
Subject: strange CGI problem
Message-ID: <85ga10$p8u$1@nnrp1.deja.com>

I'm creating a little web-site; for some reason when I view the page in
Internet Explorer, it's fine, but Netscape shows me the raw html as if
it's a text file.  What would cause this?


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


From dworkin at ccs.neu.edu  Tue Jan 11 20:40:46 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 11 Jan 2000 20:40:46 -0500
Subject: a small problem
In-Reply-To: knotwell@my-deja.com's message of "Wed, 12 Jan 2000 00:06:50 GMT"
References: <85ggif$u8t$1@nnrp1.deja.com>
Message-ID: <vndiu10dq3l.fsf@betelgeuse.ccs.neu.edu>

knotwell at my-deja.com writes:

> Is the following a bug?
> 
> >>> q = [(1,'dd'),(2,'ddd'),(3,'ddddd')]
> 
> >>> reduce(lambda x,y: x[0] + y[0],q)
> 
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 1, in <lambda>
> AttributeError: __getitem__

If you expand the reduce call, the problem becomes obvious:

((1,'dd')[0] + (2,'ddd')[0])[0] + (3,'ddddd')[0]

In the first step, you do (1,'dd')[0] + (2,'ddd')[0], which is fine.
It evaluates to the integer 3.

The next step, you do 3[0] + (3,'ddddd')[0], which will fail because
the result of the previous step (3) cannot be subscripted.

-Justin

 



From xyzmats at laplaza.org  Mon Jan  3 10:44:11 2000
From: xyzmats at laplaza.org (Mats Wichmann)
Date: Mon, 03 Jan 2000 15:44:11 GMT
Subject: Python Books & improving PyAwareness - How to ?
References: <01bf5597$3dfee1c0$ddc854a6@brancusi.pro-net.co.uk>
Message-ID: <3870c2a9.8881691@news.laplaza.org>

On 3 Jan 2000 03:09:34 GMT, "Jason Cunliffe" <jasonic at nomadicsltd.com>
wrote:


>Q1: How to improve the bookstore/publisher handling of Python book
>placement?
>
>Q2: Does it matter? 

And on a different-but-related topic, who offers Python training?  I
checked with a training provider I do some work with and Python is,
umm, below the scope of their radar :-(

Another language visibility factor these days is the availability of
training....

Mats Wichmann

(Anti-spam stuff: to reply remove the "xyz" from the
address xyzmats at laplaza.org. Not that it helps much...)


From news at dorb.com  Sat Jan  1 19:56:03 2000
From: news at dorb.com (Darrell)
Date: Sat, 1 Jan 2000 19:56:03 -0500
Subject: Finding where Python is on a Unix server through Telnet?
References: <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com>
Message-ID: <pkxb4.1245$Sr1.119203@typhoon.nyroc.rr.com>

iris:~$ which python
/home/dorb/usr/local/bin/python

--
--Darrell
"Tyler Eaves" <tylere at greenvillenc.com> wrote in message
news:V8xb4.85295$Ym1.1550463 at tw11.nn.bcandid.com...
> What is the command?
>
> Tyler
>
>
> --
> http://www.python.org/mailman/listinfo/python-list




From fredrik at pythonware.com  Wed Jan 19 06:30:30 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 12:30:30 +0100
Subject: procmail-like spamfilter ?
References: <3884EA56.4A3C29C8@lemburg.com>
Message-ID: <001301bf6270$9bc08d50$f29b12c2@secret.pythonware.com>

M.-A. Lemburg <mal at lemburg.com> wrote:
> Does anyone know of a procmail like spam filter that I could
> use to prevent all these spam mails arriving at my mail box ?
> 
> Ideally it should be written in Python, of course :-) Snakes
> eating spam... you know.

here's one:
http://www.nightmare.com/software.html => SpamWall

</F>




From brian.pedersen at mail.danbbs.dk  Fri Jan 21 03:44:57 2000
From: brian.pedersen at mail.danbbs.dk (Brian Dam Pedersen)
Date: Fri, 21 Jan 2000 09:44:57 +0100
Subject: Configure EMACS for Python
References: <3887CF58.BEF947CA@quiknet.com>
Message-ID: <3g%h4.71$el3.1310647333@news.euroconnect.net>

There is a python module and some well-written install notes on

http://www.python.org/emacs/

--
Brian Dam Pedersen
Software Engineer
George Palma <gapalma at quiknet.com> wrote in message
news:3887CF58.BEF947CA at quiknet.com...
I have recently begun to use Linux (moving fromm OS/2) and want to continue
to use
Python as a scripting language. Can anyone give me advice on how I can
configure EMACS
to format Pyhton code? I reasonably sure that someone has figured this out
and I do
not want to reinvent the wheel.
--
George Palma
Loomis, CA
Running on Linux, OS/2, and DOS





From shayman at uniserve.com  Sun Jan 30 00:45:09 2000
From: shayman at uniserve.com (Mattew)
Date: Sat, 29 Jan 2000 21:45:09 -0800
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net>
Message-ID: <949210788.364536@neptune.uniserve.ca>

Recently installed Py1.52 on a Win2K pro  Cell 466.   Suprised to see Tcl/tk
boot  so fast.  and Python taking way longer.
Don't have any explantion


Jason Taylor <jasont at wellmed.com> wrote in message
news:MPG.12f911f129ce00ff989681 at or.news.verio.net...
> > > Under NT, it seems that starting up python to process even
> > > a trivial script is pretty slow.
> >
> > Under all installations of NT, or some specific one that you tried?
Don't
> > laugh <wink>.  One thing to watch out for is fragmentation under NT.
Get a
> > defrag utility (e.g., Diskkeeper Lite is free) and check that out.
>
> First, thanks guys for the help!
>
> I have checked a couple other installs and they *don't* seem to have the
> problem.  So there appears to be something fishy with my machine.  I
> tried all the suggestions posted so far: defragged, kept my machine
> idle, used -S and -X options to trim startup load.  Put python before
> perl in my PATH.  It's a little bit faster but nowhere near what I am
> now seeing from python on other NT machines.
>
> Just to simplify, I uninstalled all the Python packages (Base,
> Win32/com, Tcl) and just installed py152.exe.  No change.  All the
> registry keys look fine, at least the obvious ones I know to check such
> as HKEY_LOCAL_MACHINE/SOFTWARE/Python and the .py association.
>
> I will keep playing with it.  If anyone else has seen this, speak up so
> we can identify commonalities.
>
> Thanks!!
> Jason Taylor
>
>
>




From a.eyre at optichrome.com  Mon Jan 31 06:45:22 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 11:45:22 -0000
Subject: sending keystroke to a dos application
In-Reply-To: <3.0.6.32.20000131123003.009693c0@orion.dgfz.de>
Message-ID: <000501bf6be0$a8397f50$3acbd9c2@peridot.optichrome.com>

> I have a dos application that runs in a dos window under NT or win98 that
> takes interactive input. Basically numbers to choose from menu items and
> return keys. I have not found a method to script this application by
> sending keystrokes to the dos window. Is there a any possibility 
> to do this with python?

Providing you don't actually need this app to run on MS-DOS, but just
98/NT in a DOS box, you can use:

import msvcrt
key = msvcrt.getch()

...to get a single keystroke.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From bovinesoft at my-deja.com  Mon Jan 31 22:40:48 2000
From: bovinesoft at my-deja.com (bovinesoft at my-deja.com)
Date: Tue, 01 Feb 2000 03:40:48 GMT
Subject: How do I create a new file?
Message-ID: <875kjv$m8r$1@nnrp1.deja.com>

What code would I use if I wanted to create a new file in python?  If
you could help me with this, I would be greatly appreciative.  Thanks
again.


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


From helge at mdlink.de  Thu Jan  6 16:14:18 2000
From: helge at mdlink.de (Helge Hess)
Date: Thu, 06 Jan 2000 22:14:18 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <14452.56388.442949.799702@weyr.cnri.reston.va.us> <3874E118.1C062D2C@mdlink.de> <200001062019.MAA30417@garnet.armored.net>
Message-ID: <387505AA.9E442E9A@mdlink.de>

"Samuel A. Falvo II" wrote:
> 
> Because __init__() is a constructor; it is called _after_ the object is
> already allocated and minimally initialized.  So when __init__( self, ... )
> is called, "self" already contains a reference to the object which you are
> initializing.
>
> It is not __init__()'s job to allocate the object.

I can call __init__ like any other method and in fact it is a usual
method. So the only specialty about a constructor in Python is that it
is automatically invoked at class execution. 
In fact is see '__init__' as practically the same as the '__call__'
special method, only that __call__ works on instances while __init__
works on classes.

Only to be sure that we don't misunderstand: it's absolutly clear to me
that returning a new object from __init__ will instantly deallocate the
current object if it isn't stored by a side-effect. I don't consider
this very problematic since: a) this will be used in a restricted
problem domain only, b) loosing programmer convenience is worse (he has
to learn two objects instead of one) and c) the overhead for object
allocation/deallocation is very low compared to other operations in
Python.

best regards
  Helge



From mikael at isy.liu.se  Thu Jan 20 11:00:05 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 20 Jan 2000 17:00:05 +0100 (MET)
Subject: Sucking e-mail from Exchange Server vis HTTP?
In-Reply-To: <slrn88ebbe.67b.grant@grante.comtrol.com>
Message-ID: <XFMail.000120170005.mikael@isy.liu.se>

On 20-Jan-00 Grant Edwards wrote:
[A detailed description about a stubborn guy at work and some
 lame bosses]

Sounds like you should do what the rats are known to do: Leave the ship 
before it sinks. I mean, if this is how your company is run it will sink
sooner or later. Sooner rather than later.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    16:54:54

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From echeverria at interactiva.cl  Sat Jan  1 19:14:22 2000
From: echeverria at interactiva.cl (Cristian Echeverria)
Date: Sat, 1 Jan 2000 20:14:22 -0400
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com>
Message-ID: <386e8a42@omega>

I work with Python on Windows 95/98 and I think that the best option
for gui is wxPython: http://alldunn.com/wxPython/.

Apart from windows wxPython run on Linux/Gtk and the two versions
are very sincronized. wxPython has all the controls that Tkinter does not
have,
Printing support and a very good html windget.

The Win distribution has a very good demo, very good documentation
and there is a wxPython user list with people that really like to help
each others.

Bye

Cristian Echeverria




From throwaway1 at my-deja.com  Mon Jan 24 09:40:56 2000
From: throwaway1 at my-deja.com (throwaway1 at my-deja.com)
Date: Mon, 24 Jan 2000 14:40:56 GMT
Subject: Replacement for ConfigParser (was Re: ConfigParser and .ini files)
References: <86a1s2$7ha$1@nnrp1.deja.com> <14472.39533.223900.313932@weyr.cnri.reston.va.us> <388C1FEB.D4CDAD6E@murl.com>
Message-ID: <86ho9o$fai$1@nnrp1.deja.com>

> "Fred L. Drake, Jr." wrote:

> > super_banane at yahoo.com writes:
> >  > The ConfigParser class in the default
Python libraries
> >  > does not support comments given on the
value lines, separated
> >  > by a semicolon, such as:
> >  >
> >  > [Section]
> >  > keyword = value ; put your comments here
> >
> >   You're right, it doesn't.
> >   Does anyone know of a *real* specification
for the .ini format?  I
> > didn't know this was allowed.

I've written a GPL'ed replacement for
ConfigParser, called ConfParser.  It doesn't allow
the dictionary formatting, but it does sensibly
allow comments just about wherever you'd like.  It
also removes most of the restrictions on what
characters can be in a section or option name.

It's available at:
http://www.qcc.sk.ca/~charlesc/software/

It mostly follows the interface from ConfigParser,
but doesn't have some of the convenience methods
of the original.

Charles Cazabon
<software at discworld.dyndns.org>


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


From toberstein at my-deja.com  Sat Jan 29 15:45:35 2000
From: toberstein at my-deja.com (toberstein at my-deja.com)
Date: Sat, 29 Jan 2000 20:45:35 GMT
Subject: Problem : using ADO in ASP
Message-ID: <86vjhe$grb$1@nnrp1.deja.com>

I am using Python as an ActiveX Scripting Engine
in ASP/IIS and want to use ADO to connect to a
ODBC-compliant DB (in fact Access).

VBScript Code works - so there should be no setup
problem. Python in ASP Pages also works.

ASP Script :

<HTML><BODY>
<% @LANGUAGE=Python %>
<SCRIPT Language="VBScript" RUNAT=Server>
set db = Server.CreateObject("ADODB.Connection")
db.Open("DSN=MW")
set rs = Server.CreateObject("ADODB.RecordSet")
set rs = db.Execute("select * from water_a")
Response.Write( rs( "name" ) )
Response.Write( rs.Fields.Count )
db.Close
</SCRIPT>
<BR><BR>
<SCRIPT Language="Python" RUNAT=Server>
db = Server.CreateObject("ADODB.Connection")
db.Open("DSN=MW")
rs = db.Execute("select * from water_a")
#Response.Write( rs.Fields("name") )
#Response.Write( str(rs.Fields.Count) )
db.Close
</SCRIPT>
</BODY></HTML>


Output from the VBScript Code (OK) :

Balthasar-Neumann-Quelle33


Output from Python :


Python ActiveX Scripting Engine error '80020009'

Traceback (innermost last): File "<Script Block
>",
 line 7, in ? Response.Write( rs.Fields("name") )
AttributeError:
 'tuple' object has no attribute 'Fields'

/PythonASP_ADO.asp, line 17

Response.Write( rs.Fields("name") )


Python ActiveX Scripting Engine error '80020009'

Traceback (innermost last): File "<Script Block
>", line 8, in ? Response.Write(
str(rs.Fields.Count) ) AttributeError: 'tuple'
object has no attribute 'Fields'

/PythonASP_ADO.asp, line 18

Response.Write( str(rs.Fields.Count) )



Platform :

        NT40 Workstation SP4
        IIS4 (PWS Edition)
        Python 1.5.2 for Win32 Binary Distr.
        Pythonwin Build 125
        ADO 2.0, Access



Any useful Info? Many thanx in advance .. Tobias


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


From ScrewMe at mypad.com  Mon Jan 17 02:04:54 2000
From: ScrewMe at mypad.com (TechSpood)
Date: Mon, 17 Jan 2000 01:04:54 -0600
Subject: Very Easy question concerning input
Message-ID: <D2zg4.1375$Pb4.11132@newsfeed.slurp.net>

can someone tell me how to use a variable from a previous line and get a
variable from the currentline using raw_input, for example

I am trying to make a simple program that ask the user his name then says
"welcome (name) would you like to do something"

I tried this...but it doesn't work right...

name=raw_input ("What is your name?")
choice1=raw_input "welcome, %s, would you like to do something?" % name

doing this displays a %s instead of the persons name, can you tell me how to
fix this?

thanks





From thamelry at vub.ac.be  Tue Jan 18 06:44:05 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 18 Jan 2000 11:44:05 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <8610or$nsu$1@news.udel.edu>
Message-ID: <861jm5$nrj$1@mach.vub.ac.be>

Terry Reedy <tjreedy at udel.edu> wrote:

: "tye4" <tye4 at yahoo.com> wrote: 

:> Indentation is a confusing and lame way to separate what is inside a block
:> and what's outside it.

: Prediction: within three months you will have ceased either to believe this
: xor to use Python.

Not necessarily. I've been using python for about a year or so and I still
dislike indentation. Indentation can make your code less readable and leads to
mysterious errors. I'm sure the strict use of indentation keeps a lot of potential users
away from python. However, I feel that the other features of the language 
outweigh this disadvantage.

Cheers,

---
Thomas Hamelryck
VIB-Free University Brussels (VUB), Belgium
Current address:
Institute of Molecular and Structural Biology
Aarhus University
Gustav Wieds Vej 10C
DK-8000 Aarhus C
Denmark                                                                                            




From tim_one at email.msn.com  Wed Jan 19 01:55:10 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 19 Jan 2000 01:55:10 -0500
Subject: Suggestions for python 2
In-Reply-To: <863ka6$7ej$1@nntp9.atl.mindspring.net>
Message-ID: <000f01bf624a$20b68e00$9a2d153f@tim>

[Evan Simpson]
> I rather like the idea of having to explicitly bring names
> into an inner scope,

[Tim]
> Me too -- except in the cases that bite most often, where a 
> trivial little lambda wants to refer to one or two measly
> little names from its context.  People stumble on that so
> often it's a FAQ.

[Aahz Maruch]
> ....and there's an easy suggestion for Python 2 that fixes
> the problem, right?  ;-)

The easiest:  throw away lambda <wink>.

> ...
> Have a *HAPPY* day!!!!!!!!!!

Oh, Aahz.  You used to be so ... *normal*!

i'll-have-a-happy-day-when-i-damn-well-feel-like-it-ly
    y'rs  - tim





From srenner at lycosmail.com  Thu Jan  6 22:17:58 2000
From: srenner at lycosmail.com (srenner at lycosmail.com)
Date: Fri, 07 Jan 2000 03:17:58 GMT
Subject: Python not a Very High-Level Language?
References: <slrn86ur6a.fph.neelk@brick.cswv.com> <000801bf5802$dcca7d60$0aa0143f@tim>
Message-ID: <853lss$tj8$1@nnrp1.deja.com>

> >> Languages ... that should also be mined for inspiration are
> >> Prolog, Dylan and SETL

[Mention of Mozart/Oz]

Tim et al:

	Mozart/Oz does look interesting. It suffers from the same problem as
Python, as far as I am concerned: it is written in C. (Or maybe C++.)
UHLL or even VHLL are not what is always needed, as shown by the fact
that all of these languages seem to be written in C or C++ (or Java,
which presents different issues that I'll get to in a minute.) As you
are probably only dimly (if at all) aware, I've been using Python and
PyOpenGL to test some realtime rendering ideas. At a certain point I
began to grow increasingly frustrated my lack of control over the many
layers involved (Python, PyOpenGL, Mesa library, X windows, Linux). The
chief point of frustration was the repeated delay of XFree 4.0 and
accelerated 3d board drivers, but it was embedded in a larger
frustration, and it became clear the only way to escape it was to either
master C, C++, and lots of other stuff which I already had theoretical
reasons to avoid, or to find an alternative which I could better hope to
comprehend.
    Another consideration was speed. I wanted Python's RAD combined with
the option of dropping into a less dynamically typed language for
compiled speed. Again, I could have used C. But whether it is a
glorified assembler or a horrible joke, the only reason to use C is
because everyone else is already using it. There aren't any arguments
for it from first principles.
    So I needed a garbage collected systems-level language for which I
could hope to build RAD tools* and which had at least one accelerated 3d
driver. There is such a language. It's called S3 Native Oberon. It runs
on bare PC hardware with no other OS. There are theoretical arguments
for Oberon that I'm not qualified to give but which I buy.
    Why not Java? It doesn't run on a bare PC, as far as I know, so
C/Linux/X is still there. It's huge. It doesn't look any nicer than C++,
although it has type safety and GC. And Oberon caught my eye.
    The lone developer is obviously in an exceptional position. I never
received one piece of mail about my renderer prototype in the PyOpenGL
demo directory. This is not a complaint, merely a reason that I don't
have to worry about whether others can run my software.
    Again, Oz looks so cool. And Python is great. Just not C. Or
anything that reminds me of C.

sr

*I'm considering a Squeak ST/Oberon project because my Argentinian
programmer friend, a Squeak guru, is coming to the States.


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


From alan.gauld at gssec.bt.co.uk  Wed Jan  5 11:21:39 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Wed, 05 Jan 2000 16:21:39 +0000
Subject: How can I use tcl/tk 8.2.2 in Python
References: <82qf4h$goo$1@news2.nctu.edu.tw>
Message-ID: <38736F93.24A867B7@gssec.bt.co.uk>

> I try to use Tkinter with tcl/tk 8.1~8.2.
> And I get an error that tell me It can not find tcl80.dll .....

I don't really understand why this happens but I suspect you 
have some Tcl/Tk DLLs in your Windows\System32 directory.

Delete(or rename if your nervous) those DLLs and the error 
should go away.... Its worked (several times) for me! :-)

Alan G.

-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.


From robin at alldunn.com  Thu Jan  6 12:19:42 2000
From: robin at alldunn.com (Robin Dunn)
Date: Thu, 6 Jan 2000 09:19:42 -0800
Subject: Python and GUI options
References: <XFMail.000105210217.Vadim.zeitlin@dptmaths.ens-cachan.fr>
Message-ID: <284d4.51$%Ph.12845336@news.randori.com>

>
> > Re the other problems.  If build system changes don't do away with it (I
> > hope they do), more docs on creating the build.local file are a must.
>
>  Sorry again, but I don't even know what is it?
>

It's used by the build script (.../utils/wxPython/distrib/build.py) for
wxPython.  I created this so I wouldn't have to maintain two different build
systems for win32 and unix, (and eventually Mac) and the new distutils tools
were not far enough along at the time.


>
> > Also, wxGTK and wxPython sense different
> > compilers (SGI vs. GCC).  I had to coerce them to use the same ones.
>
>  Hmm... How did this happen? Did the value of your CXX env var changed
> in between somehow?
>

The build.py tool tries to figure out what compilers were used to build
Python and uses those.  I've had problems with extension modules before if
there were differences.  I didn't think about similar issues with wxGTK,
maybe if wx-config had some new flags that gave compiler/static
linker/dynamic linker specifications...


Basically, build.py does the following:

1. Discovers whatever it can from Python itself and the environment on the
machine about how Python was built on the system, and how to build Python
extension modules, similar to how the old standard Makefile.pre.in worked.

2. Reads information about what needs built in the current directory from
./build.cfg.

3. Looks for ./build.local, ../build.local, and ../../build.local for
overrides to the default or discovered values.

4. Builds a Makefile that uses `wx-config --cflags` and `wx-config --libs` to
handle wx configuration issues.

5. Based on command line flags runs make, make install, make clean, etc.


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!






From thomas at xs4all.net  Mon Jan 17 19:58:09 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Tue, 18 Jan 2000 01:58:09 +0100
Subject: DBI spec and parameter passing
In-Reply-To: <1263967432-11066499@hypernet.com>; from gmcm@hypernet.com on Mon, Jan 17, 2000 at 07:24:21PM -0500
References: <3883A0CA.AB0D4854@exceptionalminds.com> <14467.42391.905771.673058@beluga.mojam.com> <1263967432-11066499@hypernet.com>
Message-ID: <20000118015809.F21652@xs4all.nl>

On Mon, Jan 17, 2000 at 07:24:21PM -0500, Gordon McMillan wrote:
> Skip wrote:
> >               """ country = %s""", (city, state, country))

> but Skip _meant_:

> c.execute(""" select id from city where city = ? and state = ? 
> and country = ?""", (city, state, country))

Doesn't that depend on the DB-API module in question ? As I recall, it's
module-dependant what kind of arguments it accepts, and you can query an
attribute to find out the specifics of that module.

Or-is-that-loud-noise-my-brain-farting-ly y'rs
-- 
Thomas Wouters <thomas at xs4all.net>

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



From aahz at netcom.com  Mon Jan 10 01:53:30 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 10 Jan 2000 06:53:30 GMT
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <3876B79F.30439283@mdlink.de> <856k4d$7e9$1@nntp5.atl.mindspring.net> <3879074F.1A4EDED3@mdlink.de>
Message-ID: <85bvla$47h$1@nntp1.atl.mindspring.net>

In article <3879074F.1A4EDED3 at mdlink.de>,
Helge Hess  <helge.hess at mdlink.de> wrote:
>Aahz Maruch wrote:
>> Helge deleted his own attribution here:
>>>
>>>Hm, I wonder why it is simpler to have two separate kinds of methods
>>>instead of a consistent behaviour for all methods ?
>> 
>> Because we're looking at it from two different perspectives.  You're
>> looking at the process from the bottom and I'm looking at it from the
>> top.  Providing your capability would make it too easy for an
>> inexperienced programmer to screw things up royally.
>
>I can't see on what you base this statement on ?! This capability
>generalizes, not specializes, the environment - which leads me to the
>assumption that the *current* state will make programmers wonder.

Well, no, not in my experience.  It seems pretty obvious to me that
__init__ is a special method that is only called at object
instantiation.  Object instantiation requires that the constructor
return the object created.  __init__ gets called by the constructor to
ensure that the returned object is in a valid state, and it seems
reasonable to me that __init__ cannot substitute a different object for
the constructor to return.

>Until now I haven't heard a reason against this 'feature' which I can
>follow. I guess that it is just a matter of taste.

That is essentially correct.

>> I think one point that you're missing is that double-underscore methods
>> in general do not follow the straight standard method paradigm; they all
>> have restrictions on their interface.
>
>Really ? I was not aware of that, I thought that I could call all
>methods as usual and that the only 'special' thing is, that they are
>automatically called by the system.
>For example '__getattr__', it is a method like any other. Eg although
>the system always calls __getattr__ with a single parameter, you could
>still define the method as 'def __getattr__(self, *args, *kwArgs)' and
>use it in different context's too (whether this makes sense is another
>question, eg for __init__ it could make sense (to reinitialize the
>instance)).

OTOH, the __getslice__ method (for example) requires that the returned
object be some sort of list or tuple.  It is possible, of course, to
pervert all of these interfaces, but I think that their clarity is a
great strength of Python.

Of course, you can also call the special methods directly, but that is
strongly discouraged, and if you fail to conform to the interface you
are likely to run into problems with Python calls them automatically.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From s.schwarzer at ndh.net  Fri Jan  7 04:55:47 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Fri, 07 Jan 2000 10:55:47 +0100
Subject: bdb: which class?
References: <B9AA1D5DA95ED211900100805F6FFF9C02F3D8F6@exbhbhmb01.us.dbisna.com> <3874C7B9.106A5730@ndh.net> <38754A4D.5D8D1389@objectsbydesign.com>
Message-ID: <3875B823.465094E1@ndh.net>

Hi Stuart,

Stuart Zakon schrieb:
> My class is derived from class Bdb in bdp.py. This is the Generic Python
> debugger class. 

You mean pdb.py (Python DeBugger class, not bdp.py). I thought you were
speaking about some database module. (I might have figured out you
meant the debugger, if I had longer stared at your code ;-) .)

So I did _not_ understand your question :-)

Stefan


From mgushee at havenrock.com  Fri Jan 14 17:32:42 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 14 Jan 2000 17:32:42 -0500
Subject: ANN: SurfNotes -- a web productivity application
Message-ID: <m3bt6op9md.fsf@havenrock.com>

Greetings, Pythoneers--

SurfNotes is a web productivity application that works together
with Netscape Navigator/Communicator on Unix platforms. It pro-
vides a convenient interface for taking notes on the web sites
you visit. It allows you to save and retrieve files in a native
format (using Python's 'pickle' or 'cPickle' module), or to
export to HTML.

SurfNotes is available at:

    http://www.havenrock.com/softlab/surfnotes/index.html

The current release is a prototype. This means that, while most
of the major features are implemented, it has some serious bugs
that I am aware of, and probably others I am unaware of. I am
making it available to the public mainly in order to obtain some
constructive criticism on the concept.

So please, try it out, abuse it, see what you like and dislike
about it, and let me know!


REQUIREMENTS

Unix or Unix-like OS
Netscape browser
Python with Tkinter (required) and tkFont (optional but highly
    recommended)
Tcl/Tk

So far it has been tested only on Linux (Red Hat 5.2) with 
Python 1.5.2, Tcl/Tk 8.0, and Netscape Communicator 4.61. It
probably requires Python 1.5 or later, and Tcl/Tk 8.0 or later.
Different flavors of Unix, Linux distributions, or earlier/later
versions of Netscape should not be a problem.


By the way, for those who are curious as to why this is a Unix-only
application: I believe in creating cross-platform applications
whenever possible. The problem in this case is that Unix versions of
Netscape use a very simple protocol for remote operation (e.g. passing
URLs from external programs); as far as I know this feature is not
available on other platforms or in MS Internet Explorer.


-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From scherbi at bam.com  Mon Jan 31 11:53:41 2000
From: scherbi at bam.com (Bill Scherer)
Date: Mon, 31 Jan 2000 11:53:41 -0500
Subject: list -> dict help
Message-ID: <3895BE15.A7CF569B@bam.com>

What's the best way for me to do this:

say I have a list:

>>> l = ['a', 'b', 'c', 1]

and I want to turn it into a dictionary that looks like this:

>>> print d
{'a': {'b': {'c': 1}}}

where the length of list l is >= 3.

Basically, list[:-2] represents the nested keys, list[-2] is a terminal
key, and list[-1] is the value.

It seems I need somthing to recur over list[:-2], then add the terminal
info but I can't figure it out...

Thanks!


--
William K. Scherer
Sr. Member of Applications Staff
Bell Atlantic Mobile






From boud at rempt.xs4all.nl  Sat Jan  1 17:40:51 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 1 Jan 2000 22:40:51 GMT
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl> <386E7BD4.514764EA@ndh.net>
Message-ID: <84lvpj$sn5$1@news1.xs4all.nl>

Stefan Schwarzer <s.schwarzer at ndh.net> wrote:

> Boudewijn Rempt schrieb:
>> Yes, I read that too, a while ago, and it got me thinking about what
>> a language like he proposes would look like. He made two striking
>> suggestions - being able to draw part of the logic, in the sidebar of the
>> code, as it were, and having a sort of underlying representation of the
>> code (he said xml, but that's what everyone is saying these days),

> Because "everyone" says it, it doesn't necessarily mean that it is
> bad ;-) .

Indeed not! It's good - but whenever you hear someone extolling XML
nowadays, you've got to ask yourself whether he's just echoing the hype,
or does he know what he's talking about. I mean, XML is a great idea,
but when it's just an intermediary, you've still got to parse the original
source somewhere. And the XML will be parsed into some intermediary
format once again, before the user interface acts on the data.

[...]

> This separation of code and maintanance tool is in part already
> common: think of folding editors. An advantage of these is that you
> can still view the code with a "normal" text editor and this
> advantage should remain in future ideas. XML may be a good starting
> point.

I've tried a lot of folding editors over the years, and I've never
found one that was as pleasant to use as, say, Word's outline editor.
Perhaps that's just because I was expecting that, instead of, for
instance, Origami. What I'd like would be an editor that didn't need
to insert all kind of cruft into the source files, but could work
on a set of preferences and rules to present the source in a nice way.
An intermediary format might be handy for the developer of the editor,
but I'm not sure that it would be necessary. 

<... tuples and multidimensional arrays ...>

I've never really needed either, so don't think I really have a 
relevant opinion here. My game is interfaces, really.

-- 

Boudewijn Rempt  | http://denden.conlang.org


From skaller at maxtal.com.au  Sun Jan  9 08:39:10 2000
From: skaller at maxtal.com.au (skaller)
Date: Mon, 10 Jan 2000 00:39:10 +1100
Subject: control structures (was "Re: Sins")
References: <1264769486-16171098@hypernet.com>
Message-ID: <38788F7E.76302733@maxtal.com.au>

Gordon McMillan wrote:

> > YUKKKKKK!!! There is no natural way to do this.
> > We fall through the loop on failure, instead
> > of success. (using a flag, or testing anything
> > afterwards, is not acceptable as a solution).
> 
> That's the reason that Python has the "for
> .. else" construct that William Tanksley
> hates so much.
> 
>  for i in X:
>    if X[i]==e:
>      # do the "found" thing
>      break
>  else:
>    print "Not Found"

That doesn't work. The 'do the "found" thing'
is in the wrong place: it must be outside the loop,
since it is usually the 'normal continuation'
of the routine.

One solution is delocalisation:

	blah ...
	i = find_in_X()
	blah with i ...

where 'find_in_X()' is a subroutine, and so
the 'do the "found" thing' can be 'return i'.
This sucks: enforced delocalisation is a language design 
fault. 

Languages without lexically scoped anonymous nested function 
expressions_always_ suffer from this problem: C, C++ and Python
included.
Viper partially fixes this, in the Python case, because it provides
lexically scoped nested functions, but there's still no way to
provide anonymous function expressions, (lambda doesn't count)
due to restrictions of the syntax. Interestingly, the latest Java
supports anonymous lexically scoped method expressions.

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From quinn at myxo.ugcs.caltech.edu  Sun Jan 23 15:37:41 2000
From: quinn at myxo.ugcs.caltech.edu (Quinn Dunkan)
Date: 23 Jan 2000 20:37:41 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de> <869uel$suj$1@nntp1.atl.mindspring.net> <slrn88hnpb.brp.quinn@auto.ugcs.caltech.edu> <388B040A.F7A15B9C@connection.com>
Message-ID: <slrn88mpnc.cic.quinn@myxo.ugcs.caltech.edu>

On Sun, 23 Jan 2000 08:37:14 -0500, Colin J. Williams <cjw at connection.com>
wrote:
>> >>> map(add(5), range(5))
>> [5, 6, 7, 8, 9]
>> 
>Wouldn't it be nice if we could write this as:
>
>              5 + range(5) or 
>              range(5) + 5?
>
>Are we not complicating things by using '+' to mean addition in
>most cases but catenation in the case of a list?

No, we _are_ complicating things in that case :)  This sort of thing, while
not appropriate for lists, does make sense for (at least my definition of)
vectors, where you want "vector promotion":

>>> apply(Vec, range(5)) + 5
<5, 6, 7, 8, 9>

This is where you define __add__ and __radd__.


From sp00fD at yahoo.com  Tue Jan 18 21:40:45 2000
From: sp00fD at yahoo.com (sp00fD)
Date: Wed, 19 Jan 2000 02:40:45 GMT
Subject: **kwargs
References: <1263886656-955174@hypernet.com>
Message-ID: <86387a$7op$1@nnrp1.deja.com>

In article <1263886656-955174 at hypernet.com>,
  gmcm at hypernet.com wrote:
> sp00fD writes:
>
> > my.foo(foo="something", bar="something else", this_dict)
>
> Not quite. To call foo with the named args wrapped up as a
> dictionary, you need to use apply:
>   apply(foo, ("something", "something else"), this_dict)

Using the above apply(), how does foo know that foo(should have named
that something else) == "something" and bar == "something else", they
appear to be passed as normal args.

Also, how is this dictionary represented after the foo method receives
it?  ie. this_dict["wow"] = 1
apply(foo, (....), this_dict)

foo then sees kwargs["wow"] == 1?

> or
>   this_dict['foo'] = "something"
>   this_dict['bar'] = "something else"
>   apply(foo, (), this_dict)
>
> Note that foo cannot tell which of the above ways was used,
> (ie, "foo" and "bar" are not in kwargs when foo examines it).
>
> BTW, while foo-abuse is common, I've never seen it carried to
> such extremes.

I do like foo! ;)
Thanks
>
> - Gordon
>
>


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


From aahz at netcom.com  Sun Jan  2 11:39:58 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Jan 2000 16:39:58 GMT
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl> <386E7BD4.514764EA@ndh.net>
Message-ID: <84nv0u$bu9$1@nntp9.atl.mindspring.net>

In article <386E7BD4.514764EA at ndh.net>,
Stefan Schwarzer  <s.schwarzer at ndh.net> wrote:
>
>The other topics which the author on the O'Reilly site, Greg Wilson,
>mentions might also be worth considering. For example, I never needed
>tuples (explicitly) yet, i. e. in a sense that lists would not work.
>(If tuples would be taken out of the language or their use deprecated)
>tuple un/packing would become list un/packing and so on.)

Anyone who claims that tuples are unnecessary has never used a tuple as
a dict key.  That requirement necessitates differentiating between a
mutable and immutable list -- might as well stick with tuples.
--
                      --- Aahz (@netcom.com)

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

Ninth Virtual Anniversary -- 364 days and counting


From dnichols at sourcesignal.com  Fri Jan  7 12:09:34 2000
From: dnichols at sourcesignal.com (David Nichols)
Date: Fri, 07 Jan 2000 09:09:34 -0800
Subject: Path problem with PIL installation
Message-ID: <38761DCE.872A6944@sourcesignal.com>

Can someone give me some advice about getting PIL working.  (I'm
running Windows 95B) I installed Python (py152.exe), placing the
resulting directory tree into E:\Programs\Python.  I then unzipped
pil-win32-991101.zip into the Python directory, changing the root
directory name from py152 in the process.  From the PIL sample
directory (E:\Programs\Python\Scripts) I executed the following
command are received this error message.

============================================================
bash.exe-2.02$ Python viewer.py test.jpg
Traceback (innermost last):
  File "viewer.py", line 39, in ?
    root = Tk()
  File "E:\Programs\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__

    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:

    {} ./lib/tcl8.0 . E:/Programs/tcl8.0/library
E:/Programs/Python/library



This probably means that Tcl wasn't installed properly.
============================================================

I did do Tcl installation as prompted by the wizard, and there is an
init.tcl in E:\Programs\Python\Tcl\lib\tcl8.0 .  I suspect there is a
problem with the path Python uses to track down modules.  What do I
need to change (registry, environmental variables, ... ?) so that the
correct files are found?

I welcome any advice on how to solve this problem.

David Nichols
Source Signal Imaging
dnichols at sourcesignal.com




From tim_one at email.msn.com  Fri Jan 21 19:19:59 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 21 Jan 2000 19:19:59 -0500
Subject: In defence of the two-namespace rule
In-Reply-To: <3888D84A.DA230AA0@lanl.gov>
Message-ID: <000a01bf646e$6b060500$caa0143f@tim>

[William B. Clodius]
> Minor historical note: Curry was not the discoverer of
> currying. It was first discovered by Frege and was ten
> independently discovered, and more widely published by
> Schoenfinkel. Curry "popularized" the concept of currying,
> but, as I understand it, never used the term himself,
> generally citing the previous work of Schoenfinkel.

All that matches my understanding, although I'd give Curry credit for more
than just popularizing (while his (along with Feys) late-1950's text
"Combinatory Logic" was the key popularizer, over the preceding three
decades he published dozens of technical papers advancing the theory).

> Somehow schoenfinkeling and freging never caught on as terms,
> at least partly because the existing meaning of currying
> (removing the tangles from an animal's hair with a comb) is
> slightly analogous to the removal of tangles of logic
> introduced by the concept of currying.

Cool!  I hadn't heard that before, and it makes great sense.  However, I can
think of a few more prosaic reasons for "schoenfinkeling" not catching on
<wink>.

pondering-what-"guidoing"-will-mean-in-2037-ly y'rs  - tim





From skaller at maxtal.com.au  Mon Jan  3 09:52:49 2000
From: skaller at maxtal.com.au (skaller)
Date: Tue, 04 Jan 2000 01:52:49 +1100
Subject: "sins" (aka, acknowledged language problems)
References: <6D8A17398E28D3119F860090274DD7DB4B3D51@pces.cadlab.it>  <38656610.E4DA287B@maxtal.com.au> <slrn86ar28.4b4.neelk@brick.cswv.com> <006b01bf4f7f$b3dfe920$bf2b2bc1@martelli>
Message-ID: <3870B7C1.35D545F7@maxtal.com.au>

Alex Martelli wrote:

>     class enum:

> > So the hypothetical xreadlines() method should return an object
> > that reads one additional line from the file each time it is called.
> 
> Right; but it need not be hypothetical, see above.

	One problem with this approach is that it based around
a generic calling protocol -- which is as slow as a wet week in Python.
For example, I have extended the for statement in Viper, to accept
dictionaries and sets. It is very inconvenient to get the i'th, or
even the next, element from a set: I have a natural way to apply
a function to each element.

	As a result, Viper loops are one of the things that Viper
currently does faster than CPython -- the iteration time
is effectively zero, and that is what pystone reported on my machine.
 
> > The general problem that needs fixing is that Python really needs a
> > better iteration protocol.

> The iteration protocol implemented above seems good enough
> to me -- even though I was the one first expressing concern on
> this thread.

	I think wrappers have limited application for modelling
fundamentals. One needs to get the fundamental right, just to write
the wrappers. It's possible to write:

	for i in range(0, len(x)):

but is sucks badly. Consider instead that [1:10] is just
a notation for [1,2,3,...9], and that

	s1 s2

is the composition of two maps: a sequence is a map 
from 0 thru n-1 to some object. This explains:

	seq[1:10] <==> seq [1,2,...9]

and the result is a sequence. This means you should
be able to write something like:

	s2 = [1:10]
	s1 = ['a','b','c'..'z']
	x = s1 s2

and this can be extended to functions, if only Python
functions always accepted exactly one argument. :-)
But this interpretation also implies:

	for i in [1:10]: ..

should work. The _change_ that is required (for sequences)
is primarily syntactic (Operator 'whitespace').

> Good point.  So, we need some *inertia* in changing a language's
> definition -- at least enough to make sure that the solution space
> possible within a given language IS well explored, before coming
> to the determination that the language needs extension.
> 
> Fortunately, we have it -- and maybe we have a deeper explanation
> for Guido's apparent stonewalling about language changes...?-)

	Guido needs to maintain a lot of compatibility.
The solution space for 'upgrading' ideas of a language 
is often constrained by the initial syntax.

	Providing class based wrapper hackery to 'hook'
existing syntax falls down badly when the initial syntax
is inadequate.

	This is the case in most languages with respect
to comparators (<, <= etc) because they're usually built
on the notion of a total order. Obviously, Guido ran out
of operators, using + to concatenate strings. The nice 
slicing notation (one of Python's strong points) is not
well generalised. Etc.

	Syntax is visual appearance: the shape of code
should indicate its overall semantics, not a bunch of named
function calls like 'enum()' or 'items()'. I mean,
who needs [1:10] for a slice:

	slice(a,1,10)

is functionally equivalent. But it looks like
an _arbitrary_ function call.

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From culliton at clark.net  Mon Jan 17 23:28:40 2000
From: culliton at clark.net (Tom Culliton)
Date: Tue, 18 Jan 2000 04:28:40 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
Message-ID: <YZRg4.1292$Ve7.28796@iad-read.news.verio.net>

In article <860ftk$bas at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>I'm a python newbie.. but I like Python better than Perl (no ugly @$#@.. in
>syntax).
>
>Python does have one quirk: no end if or end loops
>
>if x == 10:
>    print 'x is 10'
>print 'Hello world'
>
># easy to get confused whether this line is in 'if' block or not
>Having an 'end' block would help tremendously.

Are you serious or are you trolling?  I can see the end of an indented
block without my glasses, even when the text is an indistinguishable
blob.  It can't get much more obvious than that.

BTW - Python has several much stranger quirks.  Things that it will
take you much longer to notice, may puzzle, bemuse, or startle you,
and when you finally understand why they are the the way they are will
give you a deeper insight.  The way "assignment" really works is one
of my favorite examples.

>if x == 10:
>    print 'x is 10'
>end if
>print 'Hello World'
>
>or
>for x in range(1, 10):
>    print x,
>end loop

Try:

# end if
# end for
# end while

This has the advantage that it will work NOW with no ugly changes to
the language for those of us who consider this rank sillyness.

>Indentation is a confusing and lame way to separate what is inside a block
>and what's outside it.

Uhm, ... I'll be polite ... Try it.  You might be suprised how fast it
grows on you.  If you still find it confusing after a couple of weeks
of real use, I'd suggest looking for another language to use.  Beyond
that, it's fairly impolite for an admitted newbie (meaning you are
admittedly ignorant of python style, idiom and philosophy) to start
throwing around fairly radical suggestions.

Have you got extensive experience at designing and implementing
computer lanaguages like Guido?  Did you spend years working on a
major language design/development project (ABC) including lots of
usability studies like Guido?  Have you had the benefit of Tim Peters,
a veritable font of language implementation experience and good taste,
whispering (or occasionally shouting) in your ear?  I rather doubt it.

Part of the reason that Python is so clean and elegant is because
Guido, with that background, discarded everything that didn't improve
clarity, conprehensibility, and ease of learning, and only kept things
that helped.


From felixt at dicksonstreet.com  Sat Jan  8 12:30:31 2000
From: felixt at dicksonstreet.com (Felix Thibault)
Date: Sat, 08 Jan 2000 11:30:31 -0600
Subject: Question about dictionaries (Question about is ?)
In-Reply-To: <m3wvpkv8ir.fsf@havenrock.com>
References: <3.0.5.32.20000105215635.0091d420@mail.dicksonstreet.com>
Message-ID: <3.0.5.32.20000108113031.007bccc0@mail.dicksonstreet.com>

At 11:21 1/8/00 -0500, Matt Gushee wrote:
<snip my question>
>I'm not sure what it means either, but what seems to be happening is
>that dict1['spam'].keys()[0] returns a reference to text[-5:-1] --
>i.e.:
>
>        'spam-a-lam-a-spama'
>                    ^^^^
>whereas dict2['spam'].keys()[0] returns a reference to text[:4], --
>i.e.:
>
>        'spam-a-lam-a-spama'
>         ^^^^
>... and the two just happen to have the same string value. 
>
>>>> dict1.keys()[0] == dict1['spam'].keys()[0]
>1
>
>> does dict2
>> take up less memory,
>
>I don't know enough about the internals of Python to be sure, but I
>can't imagine it would. dict1 references two different objects with
>the value 'spam'; dict2 references a single spammy object twice; but
>they both contain the same number of object references.
>
>> or something, or is this just a Stupid Programming
>> Trick?
>
>Hah! Your next challenge is to find a use for this interesting
>technique. If you can do that, it won't be just a Stupid Programming
>Trick ;-)

Well the use I have is building a nested dictionary out of 
some HTML...near the beginning of the program I make 
a dictionary with the tags in the document for keys, so 
it's easy to make everything downhill from that point to the
same 20 or so slices of the document instead of pointing
to hundreds ... Which is simpler for me to think about, but
I'm getting the idea that the computer doesn't care.

>
>-- 
>Matt Gushee
>Portland, Maine, USA
>mgushee at havenrock.com
>http://www.havenrock.com/
>-- 
>http://www.python.org/mailman/listinfo/python-list
>

Thanks-
	Felix




From tjg at avalongroup.net  Fri Jan 14 19:52:10 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Fri, 14 Jan 2000 16:52:10 -0800
Subject: Stubbing function
References: <387EFCA3.94C67567@exceptionalminds.com> <14463.28160.874963.826550@buffalo.fnal.gov>
Message-ID: <387FC4BA.3F012B9C@exceptionalminds.com>

Charles G Waldman wrote:
 
> Note that
> 
>      lambda: None
> 
>  works just fine.
> 
> lambdas may not contain any statements, just functions and
> expressions.  For instance:
> 
>      lambda x: print x
> 
> is illegal, wheras
> 
>      lambda x: sys.stdout.write("%s\n"%x)
> 
> is allowed.
> 
> Hope this helps,

The above was the missing piece of the puzzle for me thank you much. Now
another question: 

Those same "command=self.function" buttons do something I don't
understand:

I have a button whose command argument reads "command=self.Add" This
works as I would expect. But, if I want to pass arguments to the
function as in "command=self.Add(a,b,c)"

I see the following behaviour: The script executes the function when it
is run, NOT when the button is pushed, and actually pushing the button
does nothing.

How does one pass arguments to a callback function?

Thanks.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From schiotz1 at yahoo.com  Wed Jan  5 13:33:09 2000
From: schiotz1 at yahoo.com (Jakob Schiotz)
Date: Wed, 5 Jan 2000 10:33:09 -0800 (PST)
Subject: Changes to UserList.py
Message-ID: <20000105183309.1485.qmail@web3006.mail.yahoo.com>

Dear Python Gurus,

Apparently, classes derived from UserList can be initialized with real
lists and with instances themselves derived from UserList but not with
other list-like objects (i.e. objects defining the methods you would
expect of a list).

Example:

>>> from UserList import UserList
>>> class test(UserList):
...     pass
...
>>> test(range(5))
[0, 1, 2, 3, 4]
>>> test(xrange(5))
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/UserList.py", line 10, in __init__
    self.data[:] = list.data[:]
AttributeError: data


The problem is the constructor in UserList:

class UserList:
    def __init__(self, list=None):
        self.data = []
        if list is not None:
            if type(list) == type(self.data):
                self.data[:] = list
            else:
                self.data[:] = list.data[:]

It assumes that if list is not a real list then it must be a UserList.
In __setslice__ it is a little more clever:

    def __setslice__(self, i, j, other):
        i = max(i, 0); j = max(j, 0)
        if isinstance(other, UserList):
            self.data[i:j] = other.data
        elif isinstance(other, type(self.data)):
            self.data[i:j] = other
        else:
            self.data[i:j] = list(other)

should the same be done in __init__?  Should UserList.py be modified? 
However, this is not the ultimate solution...

We have a number of closely related list-like objects.  Some are
defined using UserList, others from scratch.  We need to be able to
"cast" lists of one type to another, and the convenient way to do that
is to initialize one object with another.  However, it does not work
because __init__ assumes that the argument to __init__ is a UserList
since it is not a list.  The trick of using list() will not work in
this case, as list() will not work on any list-like object
(list(xrange(5)) fails, for example).

This modified constructor works, but it would not do to modify UserList
in this way, as you no longer get the expected copying of the list

    def __init__(self, list=None):
        self.data = []
        if list is not None:
            if type(list) == type(self.data):
                self.data[:] = list
            else:
                self.data = list

Another possibility is to give all out objects a data attribute, it
should then be a wrapper object with a __getslice__ method returning a
real list.


I feel that the optimal solution would be a change in UserList.py, but
I cannot figure out how the optimal solution should be.  Perhaps
UserList.__init() should be changed to be similar to __cmp__, and then
the build-in function list() should be extended to work on any object
having a __getitem__ method.  The latter change should probably be done
anyway in a later version of Python.


Best regards,

Jakob Schiotz



=====
Jakob Schiotz, CAMP and Department of Physics, Tech. Univ. of Denmark,
DK-2800 Lyngby, Denmark.  http://www.fysik.dtu.dk/~schiotz/
This email address is used for newsgroups and mailing lists
(spam protection).  Official email: schiotz @ fysik . dtu . dk
When spammed too much, I'll move on to schiotz2 at yahoo.com
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://messenger.yahoo.com



From gduzan at gte.com  Mon Jan 31 10:18:36 2000
From: gduzan at gte.com (Gary D. Duzan)
Date: Mon, 31 Jan 2000 10:18:36 -0500
Subject: concurrency in Python
References: <86utmv$1va$1@nnrp1.deja.com>
Message-ID: <3895A7CC.EE462936@gte.com>

gvwilson at my-deja.com wrote:
> 
> Hi.  I'm interested in hearing from anyone who's ever built a
> concurrency toolkit for Python that used a model other than
> threads + locks --- futures, tuple spaces, active expressions,
> concurrent aggregates, actors, or anything else.

   I've done some future-ish stuff in python, built on threads.
The futures aren't transparent since (as far as I can tell) it
isn't quite possible (or certainly not straightforward) to
create a generic conditional proxy object. It does the job for
me, though, using the __call__ method to get at the actual
value, when available. By default, our future objects require
some outside agent to resolve the value; until then accessors
simply block. Optionally, a function can be passed to the future
object which is launched as a thread, with the result of the
function becoming the value of the future and unblocking any
blocked accessors. So you can do things like:

================================================================
import future

a = future.future(get_some_numeric_value, arg1, arg2)
b = future.future(get_some_other_numeric_value, arg3, arg4)

# get_some_numeric_value() and get_some_other_numeric_value()
# are off happily executing some nasty computations in their
# own threads.

z = a() + b()

# The above will block until both results are available, then
# do the addition.
================================================================

   It isn't rigorously tested or anything, but I'll attach
our implementation in the hope that someone will find it
interesting or even (*gasp*) useful.

					Gary Duzan
					GTE Laboratories


p.s. We mostly use the manual resolution, since the values
we are interested in are actually generated by external calls
to CORBA callback objects running in other threads.
-------------- next part --------------

import threading

class future:

    def __init__(self, resolver=None, *args, **kwargs):
	self._lock = threading.Lock()
	self._resolved = threading.Condition(self._lock)
	if resolver is not None:
	    apply(self.set_resolver, tuple([resolver] + list(args)), kwargs)

    def set_resolver(self, resolver, *args, **kwargs):
	self._lock.acquire()
	if self.__dict__.has_key("_value"):
	    self._lock.release()
	    raise AlreadyResolved
	if not self.__dict__.has_key("_resolver_thread"):
	    self._resolver_thread = threading.Thread(target=self.run_resolver,
						     args=(resolver,args,kwargs))
	    self._lock.release()
	    self._resolver_thread.start()
	else:
	    self._lock.release()
	    raise DuplicateResolver

    def run_resolver(self, resolver, args, kwargs):
	try:
	    value = apply(resolver, args, kwargs)
	except:
	    import sys
	    self.set_exception(sys.exc_type, sys.exc_value, sys.exc_traceback)
	    return
	self.set_value(value)

    def set_value(self, value):
	self._lock.acquire()
	self._value = value
	self._resolved.notifyAll()
	self._lock.release()

    def __call__(self):
	self._lock.acquire()
	while not self.__dict__.has_key("_value") and not self.__dict__.has_key("_exc_type"):
	    self._resolved.wait()
	self._lock.release()
	if self.__dict__.has_key("_exc_type"):
	    raise self._exc_type, self._exc_value, self._exc_traceback
	return self._value

    def set_exception(self, exc_type, exc_value=None, exc_traceback=None):
	self._lock.acquire()
	self._exc_type = exc_type
	self._exc_value = exc_value
	self._exc_traceback = exc_traceback
	self._resolved.notifyAll()
	self._lock.release()

    def invalidate(self):
	self._lock.acquire()
	self._exc_type = ResolveFailure
	self._exc_value = None
	self._exc_traceback = None
	try:
	    del self._value
	except AttributeError:
	    pass
	self._resolved.notifyAll()
	self._lock.release()

class DuplicateResolver:
    """Attempt to attach multiple resolvers to a future."""
    pass

class AlreadyResolved:
    """Attept to attach a resolver to a future which already has a value."""
    pass

class ResolveFailure:
    """Value cannot be resolved."""
    pass

From erno at iki.fi  Wed Jan 26 12:41:30 2000
From: erno at iki.fi (Erno Kuusela)
Date: 26 Jan 2000 19:41:30 +0200
Subject: os._exit(), os.kill(os.getpid(), 9)
References: <20000121145429.A4320@stopcontact.palga.uucp>
Message-ID: <87snzk4to5.fsf@fun77.kivikko.hoas.net>

>>>>> "gerrit" == Gerrit Holl <gerrit.holl at pobox.com> writes:

    gerrit> Hi, Are those two equal?

    gerrit> os._exit(127) os.kill(os.getpid(), 9)

no, the parent process can tell if you've been killed by a signal or not.

>>> def die(): 
...   if os.fork() == 0: os.kill(os.getpid(), 9)
... 
>>> def die2():
...   if os.fork() == 0: os._exit(127)
... 
>>> die()  
>>> os.wait()
(7003, 9)
>>> die2()
>>> os.wait()
(7004, 32512)

the os.W* functions can be used to extract sensible information out of
the status. it's works exactly the same way as in C, except W* aren't
macros...

  -- erno



From thomas at xs4all.net  Fri Jan 21 17:50:34 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 21 Jan 2000 23:50:34 +0100
Subject: Looking for Solaris 2.6/Python 1.5.2 make test fix
In-Reply-To: <3888DDFC.38F7BBC@sage.nelesautomation.com>; from Rod.Haper@sage.nelesautomation.com on Fri, Jan 21, 2000 at 04:30:20PM -0600
References: <86ajtj$8f8$1@srv38.cas.org> <3888DDFC.38F7BBC@sage.nelesautomation.com>
Message-ID: <20000121235034.D10685@xs4all.nl>

On Fri, Jan 21, 2000 at 04:30:20PM -0600, Rod Haper wrote:

> I'm also getting a segmentation fault on Red Hat Linux 6.1 at the same
> point in 'make test' but no core file is generated:

> rm -f ./Lib/test/*.py[co]
> PYTHONPATH= ./python ./Lib/test/regrtest.py 
> make: *** [test] Segmentation fault
> PYTHONPATH= ./python ./Lib/test/regrtest.py 
> make: *** [test] Segmentation fault

Something probably went wrong with building your Python binary. Can you
start the python executable ? do you get the prompt ? Can you execute code ?
Python should run fine on rh6.1, unless you have a buggy compiler or a buggy
computer, or you built a buggy module statically. How were you building it ?
Which version of python was it, how did you run 'configure' and what did you
change to Modules/Setup ? (You *do* need to change a few things in there.)

> lvirden at cas.org wrote:
> > 
> > When I configure then do a make all and then make test, Python core dumps:
> > 
> > rm -f /ldatae/Misc/python/Python-1.5.2/Lib/test/*.py[co]
> > PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py
> > Bus Error - core dumped
> > make: [test] Error 138 (ignored)
> > PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py
> > Bus Error - core dumped
> > make: *** [test] Error 138

The same goes for solaris, though i cannot be 100% certain that it works
there -- haven't installed Python on the one solaris box we have (but
that'll change if we start using it) but it would strike me as preposterous
if it should fail to work ;) The same goes here, however. Did you build any
things statically ? What happens if you build them dynamically ? Can you
possibly then start python, but not import that specific module ? 

Other helpful hints towards the problem: Did you get any warnings during
compilation ? (Especially the optional modules -- you may have given the
wrong library directory for linking, for instance, or linking with the wrong
library version, compared to the include files.) If either of you can't
figure it out and need a hand, feel free to mail me with the additional info
(or the list, if you think the wonderful spam is suitable for this bunch o'
vikings ;)

-- 
Thomas Wouters <thomas at xs4all.net>

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



From neelk at brick.cswv.com  Sun Jan  2 10:20:22 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 2 Jan 2000 15:20:22 GMT
Subject: Python not a Very High-Level Language?
References: <000001bf54a1$805e6140$c52d153f@tim>
Message-ID: <slrn86ur6a.fph.neelk@brick.cswv.com>

Tim Peters <tim_one at email.msn.com> wrote:
>
> In any case, it's quite rare for any given language to evolve in really
> significant ways; there are strong pragmatic and social forces opposing it.
> Fortran is a good example of an exception -- and many (I'd guess
> overwhelmingly most) old-time Fortran programmers still want nothing to do
> with the language currently called "Fortran".  So it goes!

As possible counterexamples, what about Lisp and C++?

Common Lisp is quite a different Lisp from basically all of the Lisps
that preceded it (eg, it's lexically scoped rather than dynamically
scoped). And on top of that it grew CLOS, which is (again) a pretty
major change -- it was an backward-incompatible unification of the
Common Loops and Flavors object systems. 

And C++ of course defines creeping featurism -- lots of major (and
sometimes not fully thought-out) features have been added to it as
time has gone by: operator overloading, templates, and namespaces, to
name just a few.

> People seeking revolution should look to new languages.  Languages
> Greg didn't mention that should also be mined for inspiration are
> Prolog, Dylan and SETL (the last being as close to a full-blown UHLL
> as I've ever seen).

I know (and quite like) both Prolog and Dylan, but now you've given
me another language to learn. Thanks! 

And now a favor for you: have you heard of Mercury? It's kind of like
a purely declarative Prolog with a higher-order type system designed
along the lines of Haskell or ML.

> After a mere decade <wink> of relentless salesmanship, I think
> there's some marginally slim chance that Icon's notion of generators
> (in their intersection with Sather's notion of iterators) *may* find
> their way into Python2.

Hm. Would a real metaobject protocol (as the offspring of type/class
unification) count as major change?

> IDEs for Python are a different matter, and much should be done to
> make life easier there.

Yes. Smalltalk and Common Lisp are two of the most fun languages to
program in, in large part due to the power of their interactive
environments.


Neel


From piet at cs.uu.nl  Thu Jan  6 06:11:07 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 06 Jan 2000 12:11:07 +0100
Subject: py2bat (was: How do I make a Python .bat executable file?)
References: <38722971.4B036062@mchp.siemens.de>
Message-ID: <uembv316c.fsf@cs.uu.nl>

You must be aware of the sad fact that the Windows "Command line
interpreter" mangles some arguments, notably those with an = sign in it.
This will be replaced by a space. Sometimes this can be avoided by quoting
the argument, but if you need both quotes and = signs to be passed in
arguments you are lost.

For perl there is a runperl.exe which, if copied to myscript.exe and run,
calls perl myscript.pl with all arguments passed unchanged. It would
probably be simple to clone this for python. I have never seen the source,
however, and I am too lazy to write it myself.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From tjg at avalongroup.net  Wed Jan 19 18:15:24 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Wed, 19 Jan 2000 15:15:24 -0800
Subject: Mapping Column Names to results using DBI
References: <8825686B.007D977F.00@avnet.com>
Message-ID: <3886458C.3256ED8A@exceptionalminds.com>

Lance Ellinghaus wrote:
>     dsc = cur.description
>     l = len(dsc)
>     lst = cur.fetchall()
>     rtn = []
>     for data in lst:
>      dict = {}
>      for i2 in range(l):
>          dict[string.lower(dsc[i2][0])] = data[i2]
>      rtn.append(dict)
>     return rtn

Lance,

Thanks so much for your prompt reply.

I guess the above means that the DBI spec *doesn't* call for this
behaviour. I actually find that quite surprising, I just recently had
need to parse a CSV file and found a CSV module up on the Vaults, and it
had behaviour like I was looking for, so when I couldn't figure it out
from the spec, I thought I must be missing something.

I'll give it a whirl and see what happens.

Once again, thanks.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From fredrik at pythonware.com  Sun Jan 16 14:44:31 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 16 Jan 2000 20:44:31 +0100
Subject: Breaking out of walk()
References: <38811854.F2BC197E@rl.gov> <3881f20c.844046140@news1.on.sympatico.ca> <20000116195919.A7714@stopcontact.palga.uucp>
Message-ID: <024601bf605a$1c34f8e0$f29b12c2@secret.pythonware.com>

Gerrit Holl <gerrit.holl at pobox.com> wrote:
> Robert Roy wrote on 948036970:
> > if 'cvs' in names: names.remove('cvs')
> > 
> > will skip over directories named 'cvs'
> 
> It won't:
> >>> l = ['foo', 'cvs', 'bar', 'cvs']
> >>> l.remove('cvs')
> >>> l
> ['foo', 'bar', 'cvs']

it will -- when the walk callback is called, the "names"
argument only contains the files and subdirectories
directly under "dirname"...

(if os.listdir returns duplicates on your computer,
you should fix your operating system)

</F>




From smb at bby.com.au  Tue Jan  4 23:55:02 2000
From: smb at bby.com.au (Sarah Burke)
Date: Wed, 05 Jan 2000 14:55:02 +1000
Subject: Sydney scoreboard?
References: <385DB48C.7A7A52D6@bby.com.au> <8On74.6414$Dh3.85256@ozemail.com.au> <YCU74.7437$NV6.31360@news-server.bigpond.net.au> <38601AF6.7D39F091@bby.com.au> <D0da4.3285$lv5.19947@news-server.bigpond.net.au>
Message-ID: <3872CEA2.38C8E5A0@bby.com.au>

I would be very interested in doing this.
Would people be interested in a more formal or less formal gathering, do
you think? (I am Sydney based by the way.)
Does anyone have any idea of a rough number of python users in Sydney?

Mark Hammond wrote:

> "Sarah Burke" <smb at bby.com.au> wrote in message
> news:38601AF6.7D39F091 at bby.com.au...
> > Very impressive!
> > Since python is so big here, what sort of community activities are
> there?
> > (I'm only recently
> > arrived from Chicago).
>
> Not many.  We did create a python-au mailing list, but that is very
> quiet -
> http://starship.python.net/mailman/listinfo/python-au
>
> Some sort of social gathering would be good - although it would
> obviously need to be city-by-city.  Melbourne and Sydney both have
> decent numbers of users...
>
> Mark.



From dworkin at ccs.neu.edu  Wed Jan 12 10:48:02 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 12 Jan 2000 10:48:02 -0500
Subject: converting tuple to a list
In-Reply-To: Anders M Eriksson's message of "Wed, 12 Jan 2000 13:45:18 +0100"
References: <tHZ8OL+ftbNGsmad91aonRmEC1nh@4ax.com>
Message-ID: <vnd901ve1fx.fsf@betelgeuse.ccs.neu.edu>

Anders M Eriksson <anders.eriksson at morateknikutveckling.se> writes:

> Is this the 'best' way to convert a tuple to a list?
> 
> t = (1,2,3,4,5)
> l = []
> 
> for i in range(len(t)):
> 	l.append(t[i])

No, I would suggest something more like:

t = (1,2,3,4,5) 
l = list(t)

-Justin

 



From robin at jessikat.demon.co.uk  Tue Jan 25 11:53:26 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 25 Jan 2000 16:53:26 +0000
Subject: NumPy
References: <388DC152.9624D8F5@obop.com.pl>
Message-ID: <VI1IvAAGUdj4EwTU@jessikat.demon.co.uk>

In article <388DC152.9624D8F5 at obop.com.pl>, =?iso-
8859-2?Q?Przemys=B3aw?= G. =?iso-8859-2?Q?Gawro=F1ski?=
<gawron at obop.com.pl> writes
>[ WinNT 4.0 SP 5, Python 1.5.2]
>
>I've just downloaded Numerical-15.2 and tryied to run the setup.py
>script but I get:
>Tracebak (innermost last):
>    File "setup.py", line 16, in ?
>        from distutils.command import install
>ImportError: No module named distutils.command
>
>am'I missing something ?
>
>--
>Przemyslaw G. Gawronski UIN:8358522
>mailto:gawronskip at usa.net
>mailto:gawron at obop.com.pl
>
>
>
yep you need to get the latest distutils from parnassus.
-- 
Robin Becker


From tratt at dcs.kcl.ac.uk  Thu Jan 27 04:38:17 2000
From: tratt at dcs.kcl.ac.uk (Laurence Tratt)
Date: Thu, 27 Jan 2000 09:38:17 +0000
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
References: <m12DjXD-000CxiC@artcom0.artcom-gmbh.de>
Message-ID: <38901209.2E1CD9D5@dcs.kcl.ac.uk>

Peter Funk wrote:

> Look at the following coding alternatives often seen in generic python code:
> 
> class A:
>     def load(self):
>         # ... do something e.g. load a file
>         try:
>             self.update()
>         except AttributeError:
>             pass
> 
> class B:
>     def load(self):
>         # ... do something e.g. load a file
>         if hasattr(self, 'update'):
>             self.update()
> 
> Which coding idiom is better?  The second form is one line shorter than
> the first, but it has the disadvantage, that the method (or attribute) must
> be named twice.  Both idioms are used in the standard libary.

As a general rule, I would always use the second form: your intent - that
you're not sure if update exists - is much clearer.

However, if update nearly always exists, the second form is slower because
it has to perform the lookup for update twice; if update rarely exists then
the overhead in raising the exception causes the first form to be much
slower. In practice this is rarely important unless you have a long loop,
and need to squeeze every last ounce out of the interpreter.

To give you a quick idea, I tried a little script, and the rough figures
suggest that with a small dictionary, if update exists the first form is
about x2 faster than the second; if update doesn't exist the second form is
about x4 - x6 faster. So the difference isn't that much, unless you make the
same call a lot of times.


Laurie


From charliehubbard at my-deja.com  Wed Jan 19 16:00:24 2000
From: charliehubbard at my-deja.com (charliehubbard at my-deja.com)
Date: Wed, 19 Jan 2000 21:00:24 GMT
Subject: problems setting up PIL under Redhat 6.0??
Message-ID: <8658kv$nf3$1@nnrp1.deja.com>

Hello,

I'm trying to use PIL under redhat 6.0, but I can't seem to get it to
work.  I go through the instructions for building everything, but when
I try to import _imaging I get this error.

>>> import _imaging
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: /usr/lib/libtk8.0.so: undefined symbol: XFreePixmap

Any ideas?

thanks
charlie


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


From tismer at tismer.com  Mon Jan 31 10:48:32 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 31 Jan 2000 16:48:32 +0100
Subject: Stackless Python 1.0 + Continuations 0.6
References: <200001311509.KAA18274@python.org>
Message-ID: <3895AECF.32394123@tismer.com>


mmillikan at vfa.com wrote:
> 
> Congratulations on 'stackless'!

:-))

> There appears to be a problem in handling control flow in the
> following class of constructs:
> 
> def stacklesstest():
>     try:
>         return 45
>         print 'still here' #not printed (correct)
>     finally:
>         print 'in finally' #printed (correct)
>     return 'x = 99'

Yes I know. Was found by Robin Becker since Zope won't work.
An updated version is about to be published.
I will also submit an extra test for the test suite, it
fails to check for this.

Furthermore, due to being stackless, it is now likely to
do very deep recursions (current default limit is 123456:),
which works fine. But:
If one exceeds this limit, a huge traceback is built, and
when deallocating this chain, a stack overflow occurs :-))

I've written a trashcan.c patch for all nested structures
that does smooth deallocation. At the moment, it costs
5 percent pystone performance, probably due to lack of
locality. Will change it later to be handled in all the
affected (5) modules, also with a per-thread recursion counter.
Later :-)

thank you - ciao - chris.stackless

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From themantis at trojanslair.zzn.com  Tue Jan  4 23:13:33 2000
From: themantis at trojanslair.zzn.com (Black Mantis)
Date: Tue, 04 Jan 2000 20:13:33 -0800
Subject: Importing a dicrionary from a file
Message-ID: <3872C4EC.20ADE046@trojanslair.zzn.com>

hello
I am a newbie, and i am working on a program that uses a really big
dictionary. i was wondering if i could store the dictionary in a
seperate file, let's say dict.dat. Now, at the beggining of the program
i want to import that dictionary and get it ready, how would i do this?
is there a way to read the whole input file into one variable?

thanks for your help
keagan



From a.eyre at optichrome.com  Mon Jan 31 05:07:08 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 10:07:08 -0000
Subject: Date and Time
In-Reply-To: <tVnk4.32$aw4.698@news.more.net>
Message-ID: <000201bf6bd2$eecba190$3acbd9c2@peridot.optichrome.com>

> I also have an all-python module that does a lot of date and time
> things. It might give you some ideas.
> See:
>     datetime.py
> at:
>     http://starship.python.net/crew/jjkunce/

I'm surprised that something like this doesn't exist already in the
standard Python library. Date and time manipulations are amongst the
most common things done in programs. How difficult would it be to get
it to be added in the next release?

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From robin at jessikat.demon.co.uk  Mon Jan 10 18:38:43 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 10 Jan 2000 23:38:43 +0000
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu>
 <slrn87fol9.15t.kc5tja@garnet.armored.net>
 <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk>
 <slrn87imad.thc.wtanksle@hawking.armored.net>
 <slrn87j11s.vc0.kc5tja@garnet.armored.net>
 <Y$u7SGAdYae4EwJv@jessikat.demon.co.uk> <387A04AC.886DD88C@bellatlantic.net>
 <T2FFKFAaCie4EwJn@jessikat.demon.co.uk>
 <slrn87kjtf.930.kc5tja@garnet.armored.net>
Message-ID: <kizIvIAD2me4EwPp@jessikat.demon.co.uk>

In article <slrn87kjtf.930.kc5tja at garnet.armored.net>, Samuel A. Falvo
II <kc5tja at garnet.armored.net> writes
>In article <T2FFKFAaCie4EwJn at jessikat.demon.co.uk>, Robin Becker wrote:
>>I'm sorry if I sounded too defensive. Perhaps I should regard Sam's post
>>as being enthusiastic rather than aggressive. Telling us we're in the
>>gutter of standard OO won't endear though.
>
>Of course not -- the momentum behind the current OO model is just too great.
>And it DOES have a large applicability.  It's just not suitable for use in a
>distributed object/component environment composed of heterogeneous
>programming languages. :)
>
>I will say that, upon re-reading my message, I did tend to use some pretty
>strong verbiage.  English, despite being my native (and only) language, is
>NOT one of my strong points.  However, C and assembly are.  :)
>
OK I guess we all need a bit of adrenalin now and then heh heh heh
aaaaaarrrggghhhhh the Timbot is presumably busy.
-- 
Robin Becker


From darrell at dorb.com  Sat Jan  8 11:26:16 2000
From: darrell at dorb.com (Darrell)
Date: Sat, 8 Jan 2000 11:26:16 -0500
Subject: [ANN] XMLTreeCntrol 0.1
Message-ID: <008201bf59f5$169471b0$0100a8c0@rochester.rr.com>

Hope this doesn't show up twice :-?

I mentioned my xmlP tree control as a code example and started getting lots
of hits. But it wasn't really in any shape for public consumption. I fixed a
couple of things and packaged all the required files into a zip.
http://www.dorb.com/darrell/xmlP/

Also a screen shot http://www.dorb.com/darrell/xmlP/scrn.jpg the numbers to
the right are the line number.

It will view anything with tags in it.
Things like this aren't handled: <![CDATA[         expr="age > 18" ]]>
That came from the Zope site, so I have some more work todo someday.

If you have wxPython it should work at least as good as it does for me.

--
--Darrell


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000108/3b1230f7/attachment.html>

From volumeregeling at hotmail.com  Tue Jan  4 10:40:06 2000
From: volumeregeling at hotmail.com (volumeregeling)
Date: Tue, 04 Jan 2000 15:40:06 GMT
Subject: add_radiobutton extra´s need help
Message-ID: <387213a5.18356234@122.0.0.250>

Hi,

first of all, am i the only one who find the help files for python and
tkinter difficult to understand ??

but my problem,

I am making a menu with cascaded radiobuttons 

menu
    option
           radiobutton1
           radiobutton2

something like the above.

I make these menus through
menu.add_radiobutton(label='Remote',command=Debug_remote)

Now i would like to have a default option selected on startup, how do
I do this. In the help files , i can+t really find what i am looking
for ?!


thanks, Thijs Volders


From fdrake at acm.org  Mon Jan 17 10:16:15 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 17 Jan 2000 10:16:15 -0500 (EST)
Subject: Now What? :o)
In-Reply-To: <947731623snz@vision25.demon.co.uk>
References: <n2bd4.64$%Ph.18874529@news.randori.com>
	<85iglp$qiv$1@news1.xs4all.nl>
	<947731623snz@vision25.demon.co.uk>
Message-ID: <14467.12863.135367.599432@weyr.cnri.reston.va.us>

Phil Hunt writes:
 > Python also has the rudiments of a text-based GUI designer (Parrot
 > at http://www.comuno.com/linux/parrot/intro.html). I haven't done
 > much work on it recently, as I've been busy with other things; 
 > probably I will do more work on it this month.

  Someone should probably mention that for GTK+/GNOME users, glade
works just fine with Python.  Glade can generate an XML file which can 
be loaded using libglade.  I understand that recent versions of glade
can generate Python code as well, but using libglade probably makes
more sense for most users.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From cribbsjg at earthlink.net  Mon Jan 31 10:20:23 2000
From: cribbsjg at earthlink.net (Jamey Cribbs)
Date: Mon, 31 Jan 2000 15:20:23 GMT
Subject: Local Echo using Telnetlib
Message-ID: <3895A8F6.69D0B727@earthlink.net>

Does anyone know how I can turn off local echo while in a telnet session using Telnetlib?

More specifically, I am creating a telnet session to a Linux box and then running the TALK command.  It works fine, but while I am in TALK, it looks weird because everything I type is echoed in my Talk Box.  I turned on debugging and I noticed that everything I send through the telnet session is being sent back.

However, when I manually open a telnet session (outside of Python) and run TALK, everything works fine.

Thanks in advance for any help!

--
Jamey Cribbs
Sr. Technical Consultant


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000131/667a15db/attachment.html>

From timo at bbn.com  Wed Jan 19 18:37:28 2000
From: timo at bbn.com (Tim O'Malley)
Date: Wed, 19 Jan 2000 18:37:28 -0500 (EST)
Subject: HTMLgen and Cookies
In-Reply-To: <388643c3.14755317@nntpserver.swip.net>
References: <388643c3.14755317@nntpserver.swip.net>
Message-ID: <14470.18949.31391.618926@ghost.bbn.com>

hola.

Anders Eriksson writes:
 > My problem is: How do I use HTMLgen and Cookie.py together?

Well, I've never used HTMLgen, but that won't stop me from inserting
this tasty foot into my mouth...

It appears that the CGI class of HTMLgen has an undocumented (at least I 
don't see any document of it) attribute 'http', which is a dictionary.
I would guess you could do something like (very untested):

    q = cgiapp.CGI()
    c = Cookie.SimpleCookie("a=b;")
    q.http["Set-Cookie"] = c.output(header="")

I'm treading water here, but give it a shot..

-- 
.....................................................................
.   happy daze            .   Age and treachery will beat           .
.        -tim O           .       youth and skill every time.       .
.....................................................................



From s.schwarzer at ndh.net  Thu Jan  6 11:43:53 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 06 Jan 2000 17:43:53 +0100
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com> <3874F5DA.B73A6039@starmedia.net>
Message-ID: <3874C649.67346A6F@ndh.net>

Hello,

Andrew Csillag schrieb:
> The way to make it so that the other code can access variables in the
> including files namespace is:
> exec open('filename').read()

If I understand would you mean :-) that would probably not have the
desired effect if called in a function. In this case the symbols
from the "imported" file would go into the calling function's
namespace and got inaccessible when the function is finished.

For a case when I wanted to execute a file in a specific namespace
I used

# This function allows to customize the always read HTML format from
# html_format.py .
def custom_format( format_file ):
    try:
        execfile( format_file, text_areas.html_format.__dict__ )
    except IOError, msg:
        printerr( 'Warning: can\'t use format file ' + format_file + '\n' )
    except:
        printerr( 'Error: invalid custom format file ' + format_file + '\n' )
        sys.exit( 1 )

This causes the loaded stuff to become visible in
text_areas.html_format . (I'm not sure whether this is entirely
safe, since I only _hope_ to have understood the documentation on
execfile correctly *g*.)

Stefan


From ivanlan at callware.com  Tue Jan  4 13:12:03 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 04 Jan 2000 11:12:03 -0700
Subject: Grail
Message-ID: <387237F3.C35B7FA1@callware.com>

Hi All--
Since I'm looking for an html widget in Tk, I thought I'd give grail
(grail-0.6) a try.  When I start it up, with whatever URL I give it, I
get the following traceback:

-----------start trace------------------------
[E:/grail] grail.py ./data/about.html
Traceback (innermost last):
  File ".\grail.py", line 499, in ?
    main()
  File ".\grail.py", line 153, in main
    browser.context.load(url)
  File "E:\grail\Context.py", line 487, in load
    scrollpos=scrollpos)
  File "E:\grail\Context.py", line 378, in read_page
    Reader(self, url, method, params, show_source, reload, data,
scrollpos)
  File "E:\grail\Reader.py", line 447, in __init__
    if url: self.restart(url)
  File "E:\grail\Reader.py", line 491, in restart
    BaseReader.__init__(self, self.last_context, api)
  File "E:\grail\BaseReader.py", line 62, in __init__
    self.start()
  File "E:\grail\BaseReader.py", line 85, in start
    self.checkapi_regularly()
  File "E:\grail\BaseReader.py", line 157, in checkapi_regularly
    self.callback()
  File "E:\grail\BaseReader.py", line 184, in checkmeta
    self.getapimeta()
  File "E:\grail\BaseReader.py", line 206, in getapimeta
    self.handle_meta(errcode, errmsg, headers)
  File "E:\grail\Reader.py", line 544, in handle_meta
    if not self.handle_meta_prelim(errcode, errmsg, headers):
  File "E:\grail\Reader.py", line 534, in handle_meta_prelim
    self.restart(url)
  File "E:\grail\Reader.py", line 491, in restart
    BaseReader.__init__(self, self.last_context, api)
  File "E:\grail\BaseReader.py", line 62, in __init__
    self.start()
  File "E:\grail\BaseReader.py", line 85, in start
    self.checkapi_regularly()
  File "E:\grail\BaseReader.py", line 157, in checkapi_regularly
    self.callback()
  File "E:\grail\BaseReader.py", line 184, in checkmeta
    self.getapimeta()
  File "E:\grail\BaseReader.py", line 206, in getapimeta
    self.handle_meta(errcode, errmsg, headers)
  File "E:\grail\Reader.py", line 557, in handle_meta
    bkmks.record_visit(self.url, last_modified)
  File "E:\grail\ancillary\BookmarksGUI.py", line 1000, in record_visit
    bookmarks = self.get_bookmarks_by_uri(uri)
  File "E:\grail\ancillary\BookmarksGUI.py", line 996, in
get_bookmarks_by_uri
    self.initialize()
  File "E:\grail\ancillary\BookmarksGUI.py", line 900, in initialize
    root, reader = self._iomgr.load(usedefault=1)
  File "E:\grail\ancillary\BookmarksGUI.py", line 269, in load
    fp, reader = self.__open_url_for_reading(filename)
  File "E:\grail\ancillary\BookmarksGUI.py", line 213, in
__open_url_for_reading
    fp = urllib.urlopen(url)
  File ".\grail.py", line 344, in open_url_simple
    api = self.open_url(url, 'GET', {})
  File ".\grail.py", line 339, in open_url
    api = self.url_cache.open(url, method, params, reload, data=data)
  File "E:\grail\CacheMgr.py", line 159, in open
    return self.open_get(key, url, mode, params, reload, data)
  File "E:\grail\CacheMgr.py", line 199, in open_get
    item = SharedItem(url, mode, params, self, key, data)
  File "E:\grail\Cache.py", line 74, in __init__
    self.reset()
  File "E:\grail\Cache.py", line 91, in reset
    data=self.postdata)
  File "E:\grail-0.6\protocols\ProtocolAPI.py", line 119, in
protocol_access
AttributeError: 'None' object has no attribute 'access'
[E:/grail]
------------end trace-------------------------

Any clues?  I'm running on WinNT 4.0, SP5.  160MB ram, PII 400Mhz.

If this can't be made to run, are there any HTML widgets out there?

Tnx,
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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From aahz at netcom.com  Sun Jan 30 13:32:04 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 30 Jan 2000 18:32:04 GMT
Subject: Choosing a language
References: <01bf6b45$ff345b40$412b69ce@default>
Message-ID: <872034$37d$1@nntp9.atl.mindspring.net>

In article <01bf6b45$ff345b40$412b69ce at default>,
Patrick Cooper <future at dotstar.net> wrote:
>
>I want to start learning computer programming and from what I have found so
>far I have chosen to start with Python.  I am going to be focusing on CGI
>first but eventually want to work with desktop applicates.  Python seems to
>be the most versitle and easiest to learn.  Before I go in any more depth
>does anyone think I should start with a different language?  I want to
>eventually learn C, but think Python is the best place to start.

You expect to get a different answer in a newsgroup for Python?  ;-)

Seriously, Python more than other languages TENDS to encourage good
programming practice, which makes it an excellent first language.  Note
the emphasis on "tends", though: it's easy to write crappy code in
Python if you try.  If you go through the material at
http://www.python.org/doc/Intros.html
you'll get a good handle on the Python Way.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From warlock at eskimo.com  Tue Jan  4 12:18:35 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Tue, 4 Jan 2000 09:18:35 -0800
Subject: Python not a Very High-Level Language?
References: <84lvpj$sn5$1@news1.xs4all.nl>  <84njut$t74$1@pollux.ip-plus.net>
Message-ID: <slrn874arb.ns.warlock@raven.myth>

On 2 Jan 2000 13:31:09 GMT, 
 lin li, in the persona of <linli at computer.org>,
 brought forth the following words...:

>Message by boud at rempt.xs4all.nl (Boudewijn Rempt) on: 1/1/00 10:40:48 PM
>>I've tried a lot of folding editors over the years, and I've never
>>found one that was as pleasant to use as, say, Word's outline editor.
>>Perhaps that's just because I was expecting that, instead of, for
>>instance, Origami. What I'd like would be an editor that didn't need
>>to insert all kind of cruft into the source files, but could work
>>on a set of preferences and rules to present the source in a nice way.
>
>Just an idea, how about folding on the indentation levels? We just have
>to forget about all these other languages then.
>
>Lin
Works for me:) anyone up to chewing on the vim sources for this :)

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From tseaver at starbase.neosoft.com  Fri Jan 28 20:15:53 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 28 Jan 2000 19:15:53 -0600
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <874sby5m6l.fsf@iname.com>
Message-ID: <6453DD5A03237343.2E3554C5B39D78C5.96217AE8619D1857@lp.airnews.net>

In article <874sby5m6l.fsf at iname.com>,
David Maslen  <david.maslen at iname.com> wrote:
>I didn't really want to get involved in this thread, but I have one
>comment to make.
>
>Using the python-mode in emacs, getting the code to line up with tabs
>is easy. I press tab, and it puts it in the right place. If it doesn't
>it's because I've forgotten the : after an if statement, or perhaps
>it's because of where a #comment is directly above it.
>
>It is possible to get it wrong, but agian when running python from
>inside emacs, emacs jumps to the line with the exception and it's
>quickly fixed.
>
>However, not everyone uses emacs. As an emacs user I wouldn't wish it
>on you for all it's virtues. I think the python form of indenting
>might be rather annoying if it weren't for the programing tool I
>use. I suspect were I inclined to code with vim, I might get the
>identation wrong, more often than I do now, and that might make me
>complain that {} are a good thing, like quite a few people here have
>been doing :)

Actually, vim's python mode Does the Right Thing about indenting -- I
find it quite pleasant to use (the only problem with it out of the box
is that it insists on yanking comments "#...." back to column 0).
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From jbhr at my-deja.com  Mon Jan 17 07:00:39 2000
From: jbhr at my-deja.com (jbhr at my-deja.com)
Date: Mon, 17 Jan 2000 12:00:39 GMT
Subject: Tkinter hangs again under Windows
Message-ID: <85v092$3n4$1@nnrp1.deja.com>

By searching the FAQ and Deja News, the last time I saw this
mentioned was May 1999.

Python 1.5.2, newest Windows Installer with Tcl/Tk 8.0.5, Windows 95b

Well, now I have the problem: tk.mainloop() seems to just ignore
any event, but only under certain circumstances.

My program opens a file and displays some information of it in a
Listbox. If the program is started with a filename argument, this
file is read, otherwise a filename is prompted via
tkFileDialog.askopenfilename(). The info is then shown.
Now, if the program is started with a filename, everything (well,
almost :-)) is working fine - I can select items in the listbox,
use the menu or the button. If the program is started without a
filename, the file select box comes up, lets me select a file, and
the info is then displayed as well. The difference starts when I try
to select something - nothing happens. A Ctrl-C always breaks in
Tkinter.py, line 492: mainloop.
This problem doesn't exist under Linux. And it's the second Windows PC
with the very same problem.

There's another (small) problem: I want the last entry in the listbox
to be shown and selected by default. The selection works, but the line
is only seen for a brief moment, then the list scrolls down one line for
no apparent reason ... Just to give complete information: I'm using
Fredrik Lundh's very handy AutoScrollbar.

This happened just when I wanted to show my boss how easy it is to
write a nice little program with python ... ;-)

Hans-Joachim Widmaier


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


From ehrlich at EMBL-Heidelberg.DE  Mon Jan 31 11:44:33 2000
From: ehrlich at EMBL-Heidelberg.DE (Lutz Ehrlich)
Date: 31 Jan 2000 17:44:33 +0100
Subject: Multidimensional Dictionaries: Possible?
References: <slrn88ppsg.49t.w6uv@voyager.home.com>
Message-ID: <vqs4sbufaxa.fsf@snowy.EMBL-Heidelberg.DE>

G'day,

please check out the code below. Might be what you're looking for.

Cheers,
	Lutz



"""
$Id: GrowDict.py,v 1.7 1999/11/16 17:23:08 ehrlich Exp $

This class implements a growing dictionary that you can index
however you want. It effectively catches all KeyError occurrences.
Some features include saving/loading from a stream (file or url or whatever
looks like a file in Python) and dumping the thing as XML. You can add two dictionaries,
thereby merging their respective trees.
A TimeStampGrowDict can keep a list of modification timestamps and can be sorted according 
to the latest timestamp date.

These classes can come in handy when you need a hierarchical container for your
data but you don't know the layout beforehand. You could even emulate a file system ;-)




Usage:

import GrowDict
reload(GrowDict)
gd = GrowDict.GrowDict()
gd[0]['ppp'][3455]=  [1,2,3]
gd['comment'] = 'test this'


gd.pretty()
out = open('gd.pretty','w')
gd.pretty(out)
out.close()

GrowDict.gd_save(open('gd1.pickle','w'),gd)
gd2= GrowDict.gd_load(open('gd1.pickle'))

gd3 = GrowDict.GrowDict()
gd3[100]['ff'] = 'test data'

# merging of dicts
gd4 = gd1 + gd3

# dumping as XML
gd4.toxml()
f = open('gd.xml','w')
gd.toxml(f)
f.close()


# dicts with timestamps 
t1 = GrowDict.TimeStampGrowDict('first timestamp text')
t1[0]['gaggag'] = [1,2,43]
t1.stamp('added list data')

t2 = GrowDict.TimeStampGrowDict('comparison data')
t2['data'] = 'whatsoever'

if t1 > t2: print 't1 has been modified later than t2'

"""
import sys
import copy
import DictWithDefault

class GrowDict(DictWithDefault.DictWithDefault):
    def __init__(self):
	self.data = {}
	self.defaultClass = GrowDict

    def __getitem__(self,key):
	try:
	    item = self.data[key]
	except KeyError:
	    item = copy.copy(self.defaultClass())
	    self.data[key] = item
	return item


    def clean(self):
	"""deleting empty leaves from the growDict"""
	for (kk,vv) in self.items():
	    if type(vv) == type(self):
		if len(vv.items()) == 0:
		    del self[kk]
		else:
		    vv.clean()
    def depth(self):
	"""
	return how many layers I contain, only makes sense for balanced treelike structures.
	"""
	res = 0
	for kk,vv in self.items():
	    if type(vv) == type(self):
		res = vv.depth() + 1
	    else:
		res = 1
	return res



    def pretty(self,f=sys.stdout,indent=0):
	
	for kk,vv in self.items():
	    f.write('\t' * indent  +  _encode(kk)+'\n')
	    if isinstance(vv,GrowDict):
		vv.pretty(f,indent + 1)
	    else:
		f.write( '\t' *(indent+1)+ _encode(vv)+'\n')

    def toxml(self,f=sys.stdout):
	self.doc = convert(self)
	dumpAsXML(self.doc,f)

    def __deepcopy__(self,memo):
	from copy import deepcopy
	res = self.__class__()
	for kk, vv in self.items():
	    res[deepcopy(kk,memo)] = deepcopy(vv,memo)

	return res
	
    def __add__(self,other):
	"""
	addition happens on all levels. The addition semantics are:
	
	0) When a key of 'other' is not found in 'self', the entire 
	subtree is inserted. No problem here.

	1) A a key of 'other' is found in 'self' as well: Three choices 
	here:
	
	  a) Both values are not GrowDicts: Addition is done by catenation
	  a[0] = 'rrr'  b[0] = 5 then (a+b)[0] = ['rrr',5]
	  
	  b) Both values are GrowDicts: Addition is done recursively for the 
	  values

	  c) One value is a GrowDict (or subclassed), the other is not:
	  This raises an error until I can think of a decent solution.
	

	"""
	from copy import deepcopy
	res = deepcopy(self)
	for kk,vv in other.items():
	    newkk = deepcopy(kk)
	    if not res.has_key(newkk): # insert the entire subtree
		res[newkk] = deepcopy(vv)
	    else:
		resGD = isinstance(res[newkk],GrowDict)
		othGD = isinstance(vv,GrowDict)
		if not resGD and not othGD: # non-GrowDicts, catenate them
		    res[newkk] = [res[newkk],vv]
		elif resGD and not othGD:
		    raise IOError,'merging subtree GDs of different depth'
		elif not resGD and othGD:
		    raise IOError,'merging subtree GDs of different depth'
		elif resGD and othGD: # both are GDs, add them
		    res[newkk] = res[newkk] + deepcopy(vv)

	return res

	
		
def _encode(it):
    return str(it)

def gd_save(stream, item):
    import cPickle
    if type(stream)==type(''): stream = open(stream,'w') # accepting fileNames as well
    p = cPickle.Pickler(stream)
    p.dump(item)
    stream.close()

def gd_load(stream):
    import cPickle
    if type(stream)==type(''): stream = open(stream) # accepting fileNames as well
    p = cPickle.Unpickler(stream)
    return p.load()


"""
XML conversion
"""
try:
    import DOM
    import DOM.Ext
    import DOM.Ext.Factory
    import DOM.Ext.Factory

    PP = DOM.Ext.PrettyPrint
    localCreateDocument = DOM.Ext.Factory.CreateDocument
    
except ImportError:

    try:
        from xml.dom.core import createDocument
    except ImportError:
        raise ImportError,"""Couldn't find any XML support for your Python. Please"""\
        """ install 4DOM from http://www.fourthought.com/opentech/projects/4DOM/"""
        
    def dummyPP(a,b=sys.stdout):
        print """PrettyPrinting not supported right now. Please install 4DOM from """\
              """http://www.fourthought.com/opentech/projects/4DOM/"""
        b.write(a.toxml())
    PP = dummyPP
    localCreateDocument  = createDocument





def convert(growDict):
    """
    given a BMR file object, convet it to a DOM
    """
    import sys


    doc = localCreateDocument()
    GD = doc.createElement('GD')
    doc.appendChild(GD)
    _traverse(GD,growDict,doc)
    return doc



def dumpAsXML(aDOM,f=sys.stdout):
    """
    given a document object model, do some fancy prettyprinting into XML
    """
    PP(aDOM,f)
    

def _traverse(node,growDict,factory):
    """
    traverse the starNode hierarchy and insert the appropriate DOM elements

    """
    isDict = 0
    try:
	if type(growDict) == type({}):
	    isDict = 1

	if isinstance(growDict,GrowDict):
	    isDict = 1
    except:
	pass
    
    if  isDict:
	for key, value in growDict.items():
	    newNode  = factory.createElement("entry")
	    newNode.setAttribute("key", _encode(key))
	
	    _traverse(newNode,value,factory)
	    node.appendChild(newNode)

    else:
	newNode =  factory.createTextNode(_encode(growDict))
	node.appendChild(newNode)

	

import DateTime
class TimeStamp:
    """
    simple class to implement timestamps consisting of a date and a 
    comment
    """
    def __init__(self,comment=''):
	self.date = DateTime.now()
	self.comment = comment

    def __str__(self):
	return '<TimeStamp: ' + str(self.date) + ' ' + self.comment + '>'

    __repr__ = __str__

    def __cmp__(self,other):
	return cmp(self.date,other.date)

    def toFn(self):
	from string import zfill
	d  = self.date
	year  = str(d.year)
	month = zfill(str(d.month),2)
	day   = zfill(str(d.day),2)
	hour  = zfill(str(d.hour),2)
	minute= zfill(str(d.minute),2)
	
	return '%s_%s_%s_%s:%s' %(year,month,day,hour,minute)


class TimeStampGrowDict(GrowDict):
    """
    simple extension to the GrowDict to support timestamps. 
    An instance is earlier/less than another if its last timestamp 
    is ealier/less.
    """
    
    def __init__(self,comment=''):
	self.timeStampL = [TimeStamp(comment)]
	GrowDict.__init__(self)
	
    def stamp(self,comment):
	self.timeStampL.insert(0,TimeStamp(comment))

    def __cmp__(self,other):
	return cmp(self.timeStampL[0],other.timeStampL[0])
    
    def pretty(self,f=sys.stdout,indent=0):
	if indent==0:
	    for ts in self.timeStampL:
		f.write(str(ts) + '\n')
	GrowDict.pretty(self,f,indent)

-- 
______________________________________________________________________

Lutz Ehrlich		 	http://www.embl-heidelberg.de/~ehrlich
				mailto:lutz.ehrlich at embl-heidelberg.de

European Molecular Biology Laboratory		phone: +49-6221-387-140
Meyerhofstr. 1					fax  : +49-6221-387-517
D-69012 Heidelberg, Germany				


From jstok at bluedog.apana.org.au  Sun Jan 30 22:35:08 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Mon, 31 Jan 2000 14:35:08 +1100
Subject: Choosing a language
References: <01bf6b45$ff345b40$412b69ce@default> <slrn8994sf.ndn.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <6f7l4.14957$3b6.61050@ozemail.com.au>

Remco Gerlich wrote in message ...

>Patrick Cooper wrote in comp.lang.python:
>> I want to start learning computer programming and from what I have found
so
>> far I have chosen to start with Python.  I am going to be focusing on CGI
>> first but eventually want to work with desktop applicates.  Python seems
to
>> be the most versitle and easiest to learn.  Before I go in any more depth
>> does anyone think I should start with a different language?  I want to
>> eventually learn C, but think Python is the best place to start.
>
>It's what I usually recommend to people. It's easy to learn, and
>it can be used for a *lot* of things, including CGI. Other languages
>are usually more specialistic.
>
>There are two complaints about Python as a first language that I've heard:
>1) Basically all the documentation is in English. Not a problem for you.

Python isn't the only language that has a documentation gap like that.  The
whole programming world is incredibly English-centric, and most European
programmers seem to be able to speak English.





From pinard at progiciels-bpi.ca  Mon Jan 17 17:07:23 2000
From: pinard at progiciels-bpi.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: Mon, 17 Jan 2000 17:07:23 -0500 (EST)
Subject: Prerelease: Free po-utils 0.4
Message-ID: <20000117220723.213E617C@titan.progiciels-bpi.ca>

A new pretest for the PO utilities is now available as:

   http://www.iro.umontreal.ca/contrib/po/po-utils/po-utils-0.4.tar.gz

See `NEWS'.  For `xpot', I've corrected a few Python problems, and added
more deferred translatable strings: ''"""text""" and ""'''text'''.

For `xpot', I removed the only right context in the scanner, allowing it
to run faster.  Some `#:' reference problems remain: PO file input do not
have them saved, and they could be refilled better on output.

.------.
| NEWS |
`------'

.* Package renamed from `xpot' to `po-utils'.
.* Include and install PO mode.
.* Process C, C++ and Emacs LISP (if not Mule).

.-----------.
| ChangeLog |
`-----------'

2000-01-17  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* configure.in: Rename the package from xpot to po-utils.

	* configure.in: Check where to install Emacs LISP.

2000-01-16  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* configure.in: Version 0.4.

.---------------.
| src/ChangeLog |
`---------------'

2000-01-17  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* Makefile.am: Take care of installing po-mode.el.
	* elisp-comp: New file, from elsewhere.
	* po-mode.el: To avoid compilation diagnostics, move customization
	before portability matters and add some variable declarations.

	* po-mode.el (po-subedit-cycle-auxiliary): Adapt to multiple edits.

	* scanner.l (STATE_PYTHON): Include opening parenthesis in
	identifier, instead of using slower right context.
	* parser.y (is_translatable): New, to skip it back before lookup.
	Adjust syntax, and use is_translatable.

	* scanner.l (grow_comment): Accept a delimiter argument,
	then simplify many calls.

	* scanner.l (STATE_PYTHON): Add heptuple and octuple strings.

	* parser.y, scanner.l: Implement C and C++.

2000-01-16  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* scanner.l (unexpected_character): Take a container argument.
	Callers adjusted.

	* scanner.l (initialize_scan): Skip ./ in file name.

	* xpot.c (initialize_tables): Recognise defconst and defcustom
	as introducing symbol definitions.
	* scanner.l (STATE_ELISP_STRING): Just return to STATE_ELISP,
	do not attempt returning ELISP_OTHER_ATOM.
	* parser.y (elisp_expr): Accept STRING.

	* scanner.l (prepare_for_return): Avoid calling prepare_for_rescan
	if at end of file.  When a string immediately preceded end of
	file, a spurious NUL was diagnosed.
	<<EOF>>: Do not set starter_sent.  Otherwise, a STRING token, not
	introduced by a starter token, was causing a parse error.
	Reported by Bernhard Herzog.

2000-01-11  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* parser.y, scanner.l, xpot.c, xpot.h: Implement Python.
	Rename -d to -p for --domain and -r to -v for --release.
	Implement -d for --doc-strings.  Miscellaneous cleanup.
	Also try guessing language from shell name or -*- marks.

2000-01-04  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-extract-part-unquoted): New function.
	(po-extract-unquoted, po-skip-over-python-string): Use it.

	* po-mode.el (po-find-python-string): Ignore when used to index.

2000-01-03  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-check-file-header): Add a new line after default
	header only for non-empty PO files.

	* po-mode.el (po-preset-string-functions): New function.  Add dummy
	functions for unknown mode.  Simplify other functions consequently.
	(po-tags-loop-scan): Use po-preset-string-functions.

	* po-mode.el: Have marking functions to leave cursor after marked
	string.  Do not return updated end anymore.

2000-01-02  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-tags-search): Better use the Emacs tags interface,
	and simplified.  Nested tags tables should be handled now.
	(po-tags-loop-scan, po-tags-loop-operate): New functions.
	(po-string-contents, po-string-buffer): New buffer local variables.
	(po-next-file-list): Deleted.  All usages adjusted.

1999-12-10  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-send-mail): Clarify some prompts.
	Reported by Laurent Bourbeau.

1999-12-06  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-skip-over-python-string): Missing parentheses.
	Reported by Andreas Schwab.

1999-09-17  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-find-python-string, po-mark-python-string,
	po-skip-over-python-string): New functions.
	(po-tags-search, po-mark-found-string): Use them in Python mode.

1999-07-28  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-set-msgstr): Typo in function name.
	Reported by Karl Eichwalder and Rodrigo Stulzer Lopes.

1999-07-27  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-team-name-to-code): Add Galician.

1999-07-25  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-team-name-to-code): New variable.
	(po-guess-archive-name): Use it.

	* po-mode.el (po-match-string).  New function.  Use it.

1999-06-01  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-guess-archive-name, po-guess-team-address): Accept
	more addresses, do not restrict to necessarily @li.org.

1998-12-06  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-guess-archive-name): Allow Free as well as GNU
	for domain name prefix.

1998-10-23  Fran?ois Pinard  <pinard at iro.umontreal.ca>

	* po-mode.el (po-validate): Manage so validation is done in a
	buffer named *PO validation* instead of *compilation*.
	Reported by Kalle Niemitalo.

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




From aahz at netcom.com  Mon Jan 31 13:01:11 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jan 2000 18:01:11 GMT
Subject: Eight suggestions for Python books (long)
References: <87268i$786$1@nnrp1.deja.com> <873l7f$be4$1@mach.vub.ac.be>
Message-ID: <874il7$jcf$1@nntp9.atl.mindspring.net>

In article <873l7f$be4$1 at mach.vub.ac.be>,
Thomas Hamelryck  <thamelry at vub.ac.be> wrote:
>
>BTW, I don't think a python cookbook would make much sense. 
>Python is so simple that in most cases you don't need to consult a
>book to find the solution. In most other cases, the currently available 
>books suffice.

I kind of agree with you; OTOH, if we want Python to be equal to Perl as
a sysadmin tool, there needs to be a cookbook so that someone who does
*not* know Python can use it.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From tim.hochberg at ieee.org  Wed Jan  5 22:39:13 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Thu, 06 Jan 2000 03:39:13 GMT
Subject: How to with **kw
References: <850vre$v3h$1@nnrp1.deja.com>
Message-ID: <B7Uc4.2024$R51.20554@news1.rdc1.az.home.com>

<wiktor1325 at my-deja.com> wrote in message
news:850vre$v3h$1 at nnrp1.deja.com...
> Is there a way to pass **kw args from
> one function to another,like:
>
> def func1(**kw):
>   for key in kw.keys():
>      print key
>
> def func2(**kw):
>   a_way_to_call_func1_with(kw)
>
> thanks in advance

You probably want to use apply:

def func2(**kw):
    apply(func1, (), kw) # () is the nonkw args.

-tim




From cdh at CompleteIS.com  Thu Jan 13 16:54:00 2000
From: cdh at CompleteIS.com (Chris D. Halverson)
Date: Thu, 13 Jan 2000 21:54:00 GMT
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com> <002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com> <jtu2kh52y6.fsf@wazor.biostat.wisc.edu>
Message-ID: <wpiu0xmydk.fsf@CompleteIS.com>

William Annis <annis at biostat.wisc.edu> writes:

> "Fredrik Lundh" <fredrik at pythonware.com> writes:
> 
> > #  -*- Mode: Python; tab-width: 4 -*-
> 
>         Those all work, but as I recall those work only on the first
> line of the file.  A quick test seems to confirm this.

You can also put them at the end as in:

#
# Local variables:
# mode: python
# tab-width: 4
# end:
#

cdh

-- 
Chris D. Halverson                         Complete Internet Solutions
PGP mail accepted, finger for public key   http://www.CompleteIS.com/~cdh/


From skip at mojam.com  Wed Jan 26 09:59:06 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 26 Jan 2000 08:59:06 -0600 (CST)
Subject: Interesting article on programming languages
In-Reply-To: <388E7D5B.7840983F@erols.com>
References: <388E7D5B.7840983F@erols.com>
Message-ID: <14479.3002.103586.77207@beluga.mojam.com>

re: "A Critical Look at Programming Languages" at
http://www.gamespy.com/articles/devweek_b.shtm

Thanks for recommending the article.  I particularly liked this quote in the
section on assembly language:

    Assembly programmers didn't ever think about objects or inheritance, and
    gave limited thought to functions and data structures. The value of such
    constructs is hidden when the language has no native way of expression
    them.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From fdrake at acm.org  Mon Jan 17 11:40:00 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 17 Jan 2000 11:40:00 -0500 (EST)
Subject: SUGGEST: os.getuser()
In-Reply-To: <20000116174004.A5995@stopcontact.palga.uucp>
References: <20000116174004.A5995@stopcontact.palga.uucp>
Message-ID: <14467.17888.278265.327200@weyr.cnri.reston.va.us>

Gerrit Holl writes:
 > may I suggest this function for in the posix module:

  If you're using the CVS version, there is os.getlogin(), which is
similar.  The differences are pretty minor, and have to do with
multiple user names being assigned to the same user id (which is
legal, haowever strange).  I'm sure Win32 offers some way to get the
identity of the console user, but I don't know it off-hand.  A Windows 
expert could tell us whether there's a correlation which makes sense
to use as a direct implementation for that family of platforms.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives


Return-Path: <gerrit.holl at pobox.com>
Delivered-To: python-list at dinsdale.python.org
Received: from python.org (parrot.python.org [132.151.1.90])
	by dinsdale.python.org (Postfix) with ESMTP id DA0B61CDF4
	for <python-list at dinsdale.python.org>; Mon, 17 Jan 2000 11:19:06 -0500 (EST)
Received: from scribe.pobox.com (scribe.pobox.com [208.210.124.35])
	by python.org (8.9.1a/8.9.1) with ESMTP id LAA07284
	for <python-list at python.org>; Mon, 17 Jan 2000 11:16:38 -0500 (EST)
Received: from pobox.com (i340.asd.euronet.nl [194.134.12.233])
	by scribe.pobox.com (Postfix) with ESMTP id 10901325F4
	for <python-list at python.org>; Mon, 17 Jan 2000 11:16:33 -0500 (EST)
Received: (qmail 13543 invoked from network); 17 Jan 2000 15:54:35 -0000
Received: from unknown (HELO stopcontact.palga.uucp) (qmailr at 10.0.0.5)
  by 10.0.0.1 with SMTP; 17 Jan 2000 15:54:35 -0000
Received: (qmail 3795 invoked by uid 102); 17 Jan 2000 15:55:10 -0000
Date: Mon, 17 Jan 2000 16:55:10 +0100
From: Gerrit Holl <gerrit.holl at pobox.com>
To: python-list at python.org
Subject: Re: Version history of python
Message-ID: <20000117165510.A3786 at stopcontact.palga.uucp>
Reply-To: gerrit at nl.linux.org
Mail-Followup-To: python-list at python.org
References: <85v0ih$409$1 at nnrp1.deja.com> <20000117165029.A3761 at stopcontact.palga.uucp>
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="J/dobhs11T7y2rNN"
X-Mailer: Mutt 1.0i
In-Reply-To: <20000117165029.A3761 at stopcontact.palga.uucp>; from gerrit on Mon, Jan 17, 2000 at 04:50:29PM +0100
X-homepage: www.nl.linux.org/~gerrit
X-OS: Linux 2.2.13
X-Uptime: 4:50pm  up  8:15,  3 users,  load average: 0.00, 0.02, 0.02
Sender: python-list-admin at python.org
Errors-To: python-list-admin at python.org
X-BeenThere: python-list at python.org
X-Mailman-Version: 1.2 (experimental)
Precedence: bulk
List-Id: General discussion list for the Python programming language <python-list.python.org>

--J/dobhs11T7y2rNN
Content-Type: text/plain; charset=us-ascii

Hi again,

gerrit wrote on 948124229:
> File attached.

*really* attached this time.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----

--J/dobhs11T7y2rNN
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename=HISTORY

Python history
--------------

This file contains the release messages for previous Python releases.
As you read on you go back to the dark ages of Python's history.


======================================================================


>From 1.5 to 1.5.1
=================

General
-------

- The documentation is now unbundled.  It has also been extensively
modified (mostly to implement a new and more uniform formatting
style).  We figure that most people will prefer to download one of the
preformatted documentation sets (HTML, PostScript or PDF) and that
only a minority have a need for the LaTeX or FrameMaker sources.  Of
course, the unbundled documentation sources still released -- just not
in the same archive file, and perhaps not on the same date.

- All bugs noted on the errors page (and many unnoted) are fixed.  All
new bugs take their places.

- No longer a core dump when attempting to print (or repr(), or str())
a list or dictionary that contains an instance of itself; instead, the
recursive entry is printed as [...] or {...}.  See Py_ReprEnter() and
Py_ReprLeave() below.  Comparisons of such objects still go beserk,
since this requires a different kind of fix; fortunately, this is a
less common scenario in practice.

Syntax change
-------------

- The raise statement can now be used without arguments, to re-raise 
a previously set exception.  This should be used after catching an
exception with an except clause only, either in the except clause or
later in the same function.

Import and module handling
--------------------------

- The implementation of import has changed to use a mutex (when
threading is supported).  This means that when two threads
simultaneously import the same module, the import statements are
serialized.  Recursive imports are not affected.

- Rewrote the finalization code almost completely, to be much more
careful with the order in which modules are destroyed.  Destructors
will now generally be able to reference built-in names such as None
without trouble.

- Case-insensitive platforms such as Mac and Windows require the case
of a module's filename to match the case of the module name as
specified in the import statement (see below).

- The code for figuring out the default path now distinguishes between
files, modules, executable files, and directories.  When expecting a
module, we also look for the .pyc or .pyo file.

Parser/tokenizer changes
------------------------

- The tokenizer can now warn you when your source code mixes tabs and
spaces for indentation in a manner that depends on how much a tab is
worth in spaces.  Use "python -t" or "python -v" to enable this
option.  Use "python -tt" to turn the warnings into errors.  (See also
tabnanny.py and tabpolice.py below.)

- Return unsigned characters from tok_nextc(), so '\377' isn't
mistaken for an EOF character.

- Fixed two pernicious bugs in the tokenizer that only affected AIX.
One was actually a general bug that was triggered by AIX's smaller I/O
buffer size.  The other was a bug in the AIX optimizer's loop
unrolling code; swapping two statements made the problem go away.

Tools, demos and miscellaneous files
------------------------------------

- There's a new version of Misc/python-mode.el (the Emacs mode for
Python) which is much smarter about guessing the indentation style
used in a particular file.  Lots of other cool features too!

- There are two new tools in Tools/scripts: tabnanny.py and
tabpolice.py, implementing two different ways of checking whether a
file uses indentation in a way that is sensitive to the interpretation
of a tab.  The preferred module is tabnanny.py (by Tim Peters).

- Some new demo programs:

	Demo/tkinter/guido/paint.py -- Dave Mitchell
	Demo/sockets/unixserver.py -- Piet van Oostrum
	

- Much better freeze support.  The freeze script can now freeze
hierarchical module names (with a corresponding change to import.c),
and has a few extra options (e.g. to suppress freezing specific
modules).  It also does much more on Windows NT.

- Version 1.0 of the faq wizard is included (only very small changes
since version 0.9.0).

- New feature for the ftpmirror script: when removing local files
(i.e., only when -r is used), do a recursive delete.

Configuring and building Python
-------------------------------

- Get rid of the check for -linet -- recent Sequent Dynix systems don't
need this any more and apparently it screws up their configuration.

- Some changes because gcc on SGI doesn't support '-all'.

- Changed the build rules to use $(LIBRARY) instead of
  -L.. -lpython$(VERSION)
since the latter trips up the SunOS 4.1.x linker (sigh).

- Fix the bug where the '# dgux is broken' comment in the Makefile
tripped over Make on some platforms.

- Changes for AIX: install the python.exp file; properly use
$(srcdir); the makexp_aix script now removes C++ entries of the form
Class::method.

- Deleted some Makefile targets only used by the (long obsolete)
gMakefile hacks.

Extension modules
-----------------

- Performance and threading improvements to the socket and bsddb
modules, by Christopher Lindblad of Infoseek.

- Added operator.__not__ and operator.not_.

- In the thread module, when a thread exits due to an unhandled
exception, don't store the exception information in sys.last_*; it
prevents proper calling of destructors of local variables.

- Fixed a number of small bugs in the cPickle module.

- Changed find() and rfind() in the strop module so that
find("x","",2) returns -1, matching the implementation in string.py.

- In the time module, be more careful with the result of ctime(), and
test for HAVE_MKTIME before usinmg mktime().

- Doc strings contributed by Mitch Chapman to the termios, pwd, gdbm
modules.

- Added the LOG_SYSLOG constant to the syslog module, if defined.

Standard library modules
------------------------

- All standard library modules have been converted to an indentation
style using either only tabs or only spaces -- never a mixture -- if
they weren't already consistent according to tabnanny.  This means
that the new -t option (see above) won't complain about standard
library modules.

- New standard library modules:

	threading -- GvR and the thread-sig
		Java style thread objects -- USE THIS!!!

	getpass -- Piers Lauder
		simple utilities to prompt for a password and to
		retrieve the current username

	imaplib -- Piers Lauder
		interface for the IMAP4 protocol

	poplib -- David Ascher, Piers Lauder
		interface for the POP3 protocol

	smtplib -- Dragon De Monsyne
		interface for the SMTP protocol

- Some obsolete modules moved to a separate directory (Lib/lib-old)
which is *not* in the default module search path:

	Para
	addpack
	codehack
	fmt
	lockfile
	newdir
	ni
	rand
	tb

- New version of the PCRE code (Perl Compatible Regular Expressions --
the re module and the supporting pcre extension) by Andrew Kuchling.
Incompatible new feature in re.sub(): the handling of escapes in the
replacement string has changed.

- Interface change in the copy module: a __deepcopy__ method is now
called with the memo dictionary as an argument.

- Feature change in the tokenize module: differentiate between NEWLINE
token (an official newline) and NL token (a newline that the grammar
ignores).

- Several bugfixes to the urllib module.  It is now truly thread-safe,
and several bugs and a portability problem have been fixed.  New
features, all due to Sjoerd Mullender: When creating a temporary file,
it gives it an appropriate suffix.  Support the "data:" URL scheme.
The open() method uses the tempcache.

- New version of the xmllib module (this time with a test suite!) by
Sjoerd Mullender.

- Added debugging code to the telnetlib module, to be able to trace
the actual traffic.

- In the rfc822 module, added support for deleting a header (still no
support for adding headers, though).  Also fixed a bug where an
illegal address would cause a crash in getrouteaddr(), fixed a
sign reversal in mktime_tz(), and use the local timezone by default
(the latter two due to Bill van Melle).

- The normpath() function in the dospath and ntpath modules no longer
does case normalization -- for that, use the separate function
normcase() (which always existed); normcase() has been sped up and
fixed (it was the cause of a crash in Mark Hammond's installer in
certain locales).

- New command supported by the ftplib module: rmd(); also fixed some
minor bugs.

- The profile module now uses a different timer function by default -- 
time.clock() is generally better than os.times().  This makes it work
better on Windows NT, too.

- The tempfile module now recovers when os.getcwd() raises an
exception.

- Fixed some bugs in the random module; gauss() was subtly wrong, and
vonmisesvariate() should return a full circle.  Courtesy Mike Miller,
Lambert Meertens (gauss()), and Magnus Kessler (vonmisesvariate()).

- Better default seed in the whrandom module, courtesy Andrew Kuchling.

- Fix slow close() in shelve module.

- The Unix mailbox class in the mailbox module is now more robust when
a line begins with the string "From " but is definitely not the start
of a new message.  The pattern used can be changed by overriding a
method or class variable.

- Added a rmtree() function to the copy module.

- Fixed several typos in the pickle module.  Also fixed problems when
unpickling in restricted execution environments.

- Added docstrings and fixed a typo in the py_compile and compileall
modules.  At Mark Hammond's repeated request, py_compile now append a
newline to the source if it needs one.  Both modules support an extra
parameter to specify the purported source filename (to be used in
error messages).

- Some performance tweaks by Jeremy Hylton to the gzip module.

- Fixed a bug in the merge order of dictionaries in the ConfigParser
module.  Courtesy Barry Warsaw.

- In the multifile module, support the optional second parameter to
seek() when possible.

- Several fixes to the gopherlib module by Lars Marius Garshol.  Also, 
urlparse now correctly handles Gopher URLs with query strings.

- Fixed a tiny bug in format_exception() in the traceback module.
Also rewrite tb_lineno() to be compatible with JPython (and not
disturb the current exception!); by Jim Hugunin.

- The httplib module is more robust when servers send a short response 
-- courtesy Tim O'Malley.

Tkinter and friends
-------------------

- Various typos and bugs fixed.

- New module Tkdnd implements a drag-and-drop protocol (within one
application only).

- The event_*() widget methods have been restructured slightly -- they
no longer use the default root.

- The interfaces for the bind*() and unbind() widget methods have been
redesigned; the bind*() methods now return the name of the Tcl command 
created for the callback, and this can be passed as a optional
argument to unbind() in order to delete the command (normally, such
commands are automatically unbound when the widget is destroyed, but
for some applications this isn't enough).

- Variable objects now have trace methods to interface to Tcl's
variable tracing facilities.

- Image objects now have an optional keyword argument, 'master', to
specify a widget (tree) to which they belong.  The image_names() and
image_types() calls are now also widget methods.

- There's a new global call, Tkinter.NoDefaultRoot(), which disables
all use of the default root by the Tkinter library.  This is useful to
debug applications that are in the process of being converted from
relying on the default root to explicit specification of the root
widget.

- The 'exit' command is deleted from the Tcl interpreter, since it
provided a loophole by which one could (accidentally) exit the Python
interpreter without invoking any cleanup code.

- Tcl_Finalize() is now registered as a Python low-level exit handle,
so Tcl will be finalized when Python exits.

The Python/C API
----------------

- New function PyThreadState_GetDict() returns a per-thread dictionary
intended for storing thread-local global variables.

- New functions Py_ReprEnter() and Py_ReprLeave() use the per-thread
dictionary to allow recursive container types to detect recursion in
their repr(), str() and print implementations.

- New function PyObject_Not(x) calculates (not x) according to Python's 
standard rules (basically, it negates the outcome PyObject_IsTrue(x).

- New function _PyModule_Clear(), which clears a module's dictionary
carefully without removing the __builtins__ entry.  This is implied
when a module object is deallocated (this used to clear the dictionary
completely).

- New function PyImport_ExecCodeModuleEx(), which extends
PyImport_ExecCodeModule() by adding an extra parameter to pass it the
true file.

- New functions Py_GetPythonHome() and Py_SetPythonHome(), intended to
allow embedded applications to force a different value for PYTHONHOME.

- New global flag Py_FrozenFlag is set when this is a "frozen" Python
binary; it suppresses warnings about not being able to find the
standard library directories.

- New global flag Py_TabcheckFlag is incremented by the -t option and
causes the tokenizer to issue warnings or errors about inconsistent
mixing of tabs and spaces for indentation.

Miscellaneous minor changes and bug fixes
-----------------------------------------

- Improved the error message when an attribute of an attribute-less
object is requested -- include the name of the attribute and the type
of the object in the message.

- Sped up int(), long(), float() a bit.

- Fixed a bug in list.sort() that would occasionally dump core.

- Fixed a bug in PyNumber_Power() that caused numeric arrays to fail
when taken tothe real power.

- Fixed a number of bugs in the file reading code, at least one of
which could cause a core dump on NT, and one of which would
occasionally cause file.read() to return less than the full contents
of the file.

- Performance hack by Vladimir Marangozov for stack frame creation.

- Make sure setvbuf() isn't used unless HAVE_SETVBUF is defined.

Windows 95/NT
-------------

- The .lib files are now part of the distribution; they are collected
in the subdirectory "libs" of the installation directory.

- The extension modules (.pyd files) are now collected in a separate
subdirectory of the installation directory named "DLLs".

- The case of a module's filename must now match the case of the
module name as specified in the import statement.  This is an
experimental feature -- if it turns out to break in too many
situations, it will be removed (or disabled by default) in the future.
It can be disabled on a per-case basis by setting the environment
variable PYTHONCASEOK (to any value).


======================================================================


>From 1.5b2 to 1.5
=================

- Newly documentated module: BaseHTTPServer.py, thanks to Greg Stein.

- Added doc strings to string.py, stropmodule.c, structmodule.c,
thanks to Charles Waldman.

- Many nits fixed in the manuals, thanks to Fred Drake and many others
(especially Rob Hooft and Andrew Kuchling).  The HTML version now uses
HTML markup instead of inline GIF images for tables; only two images
are left (for obsure bits of math).  The index of the HTML version has
also been much improved.  Finally, it is once again possible to
generate an Emacs info file from the library manual (but I don't
commit to supporting this in future versions).

- New module: telnetlib.py (a simple telnet client library).

- New tool: Tools/versioncheck/, by Jack Jansen.

- Ported zlibmodule.c and bsddbmodule.c to NT; The project file for MS
DevStudio 5.0 now includes new subprojects to build the zlib and bsddb
extension modules.

- Many small changes again to Tkinter.py -- mostly bugfixes and adding
missing routines.  Thanks to Greg McFarlane for reporting a bunch of
problems and proofreading my fixes.

- The re module and its documentation are up to date with the latest
version released to the string-sig (Dec. 22).

- Stop test_grp.py from failing when the /etc/group file is empty
(yes, this happens!).

- Fix bug in integer conversion (mystrtoul.c) that caused
4294967296==0 to be true!

- The VC++ 4.2 project file should be complete again.

- In tempfile.py, use a better template on NT, and add a new optional
argument "suffix" with default "" to specify a specific extension for
the temporary filename (needed sometimes on NT but perhaps also handy
elsewhere).

- Fixed some bugs in the FAQ wizard, and converted it to use re
instead of regex.

- Fixed a mysteriously undetected error in dlmodule.c (it was using a
totally bogus routine name to raise an exception).

- Fixed bug in import.c which wasn't using the new "dos-8x3" name yet.

- Hopefully harmless changes to the build process to support shared
libraries on DG/UX.  This adds a target to create
libpython$(VERSION).so; however this target is *only* for DG/UX.

- Fixed a bug in the new format string error checking in getargs.c.

- A simple fix for infinite recursion when printing __builtins__:
reset '_' to None before printing and set it to the printed variable
*after* printing (and only when printing is successful).

- Fixed lib-tk/SimpleDialog.py to keep the dialog visible even if the
parent window is not (Skip Montanaro).

- Fixed the two most annoying problems with ftp URLs in
urllib.urlopen(); an empty file now correctly raises an error, and it
is no longer required to explicitly close the returned "file" object
before opening another ftp URL to the same host and directory.


======================================================================


>From 1.5b1 to 1.5b2
===================

- Fixed a bug in cPickle.c that caused it to crash right away because
the version string had a different format.

- Changes in pickle.py and cPickle.c: when unpickling an instance of a
class that doesn't define the __getinitargs__() method, the __init__()
constructor is no longer called.  This makes a much larger group of
classes picklable by default, but may occasionally change semantics.
To force calling __init__() on unpickling, define a __getinitargs__()
method.  Other changes too, in particular cPickle now handles classes
defined in packages correctly.  The same change applies to copying
instances with copy.py.  The cPickle.c changes and some pickle.py
changes are courtesy Jim Fulton.

- Locale support in he "re" (Perl regular expressions) module.  Use 
the flag re.L (or re.LOCALE) to enable locale-specific matching
rules for \w and \b.  The in-line syntax for this flag is (?L).

- The built-in function isinstance(x, y) now also succeeds when y is
a type object and type(x) is y.

- repr() and str() of class and instance objects now reflect the
package/module in which the class is defined.

- Module "ni" has been removed.  (If you really need it, it's been
renamed to "ni1".  Let me know if this causes any problems for you.
Package authors are encouraged to write __init__.py files that
support both ni and 1.5 package support, so the same version can be
used with Python 1.4 as well as 1.5.)

- The thread module is now automatically included when threads are
configured.  (You must remove it from your existing Setup file,
since it is now in its own Setup.thread file.)

- New command line option "-x" to skip the first line of the script;
handy to make executable scripts on non-Unix platforms.

- In importdl.c, add the RTLD_GLOBAL to the dlopen() flags.  I
haven't checked how this affects things, but it should make symbols
in one shared library available to the next one.

- The Windows installer now installs in the "Program Files" folder on
the proper volume by default.

- The Windows configuration adds a new main program, "pythonw", and
registers a new extension, ".pyw" that invokes this.  This is a
pstandard Python interpreter that does not pop up a console window;
handy for pure Tkinter applications.  All output to the original
stdout and stderr is lost; reading from the original stdin yields
EOF.  Also, both python.exe and pythonw.exe now have a pretty icon
(a green snake in a box, courtesy Mark Hammond).

- Lots of improvements to emacs-mode.el again.  See Barry's web page:
http://www.python.org/ftp/emacs/pmdetails.html.

- Lots of improvements and additions to the library reference manual;
many by Fred Drake.

- Doc strings for the following modules: rfc822.py, posixpath.py,
ntpath.py, httplib.py.  Thanks to Mitch Chapman and Charles Waldman.

- Some more regression testing.

- An optional 4th (maxsplit) argument to strop.replace().

- Fixed handling of maxsplit in string.splitfields().

- Tweaked os.environ so it can be pickled and copied.

- The portability problems caused by indented preprocessor commands
and C++ style comments should be gone now.

- In random.py, added Pareto and Weibull distributions.

- The crypt module is now disabled in Modules/Setup.in by default; it
is rarely needed and causes errors on some systems where users often
don't know how to deal with those.

- Some improvements to the _tkinter build line suggested by Case Roole.

- A full suite of platform specific files for NetBSD 1.x, submitted by 
Anders Andersen.

- New Solaris specific header STROPTS.py.

- Moved a confusing occurrence of *shared* from the comments in
Modules/Setup.in (people would enable this one instead of the real
one, and get disappointing results).

- Changed the default mode for directories to be group-writable when
the installation process creates them.

- Check for pthread support in "-l_r" for FreeBSD/NetBSD, and support
shared libraries for both.

- Support FreeBSD and NetBSD in posixfile.py.

- Support for the "event" command, new in Tk 4.2.  By Case Roole.

- Add Tix_SafeInit() support to tkappinit.c.

- Various bugs fixed in "re.py" and "pcre.c".

- Fixed a bug (broken use of the syntax table) in the old "regexpr.c".

- In frozenmain.c, stdin is made unbuffered too when PYTHONUNBUFFERED
is set.

- Provide default blocksize for retrbinary in ftplib.py (Skip
Montanaro).

- In NT, pick the username up from different places in user.py (Jeff
Bauer).

- Patch to urlparse.urljoin() for ".." and "..#1", Marc Lemburg.

- Many small improvements to Jeff Rush' OS/2 support.

- ospath.py is gone; it's been obsolete for so many years now...

- The reference manual is now set up to prepare better HTML (still
using webmaker, alas).

- Add special handling to /Tools/freeze for Python modules that are
imported implicitly by the Python runtime: 'site' and 'exceptions'.

- Tools/faqwiz 0.8.3 -- add an option to suppress URL processing
inside <PRE>, by "Scott".

- Added ConfigParser.py, a generic parser for sectioned configuration
files.

- In _localemodule.c, LC_MESSAGES is not always defined; put it
between #ifdefs.

- Typo in resource.c: RUSAGE_CHILDERN -> RUSAGE_CHILDREN.

- Demo/scripts/newslist.py: Fix the way the version number is gotten
out of the RCS revision.

- PyArg_Parse[Tuple] now explicitly check for bad characters at the
end of the format string.

- Revamped PC/example_nt to support VC++ 5.x.

- <listobject>.sort() now uses a modified quicksort by Raymund Galvin,
after studying the GNU libg++ quicksort.  This should be much faster
if there are lots of duplicates, and otherwise at least as good.

- Added "uue" as an alias for "uuencode" to mimetools.py.  (Hm, the
uudecode bug where it complaints about trailing garbage is still there 
:-( ).

- pickle.py requires integers in text mode to be in decimal notation
(it used to accept octal and hex, even though it would only generate
decimal numbers).

- In string.atof(), don't fail when the "re" module is unavailable.
Plug the ensueing security leak by supplying an empty __builtins__
directory to eval().

- A bunch of small fixes and improvements to Tkinter.py.

- Fixed a buffer overrun in PC/getpathp.c.


======================================================================


>From 1.5a4 to 1.5b1
===================

- The Windows NT/95 installer now includes full HTML of all manuals.
It also has a checkbox that lets you decide whether to install the
interpreter and library.  The WISE installer script for the installer
is included in the source tree as PC/python15.wse, and so are the
icons used for Python files.  The config.c file for the Windows build
is now complete with the pcre module.

- sys.ps1 and sys.ps2 can now arbitrary objects; their str() is
evaluated for the prompt.

- The reference manual is brought up to date (more or less -- it still
needs work, e.g. in the area of package import).

- The icons used by latex2html are now included in the Doc
subdirectory (mostly so that tarring up the HTML files can be fully
automated).  A simple index.html is also added to Doc (it only works
after you have successfully run latex2html).

- For all you would-be proselytizers out there: a new version of
Misc/BLURB describes Python more concisely, and Misc/comparisons
compares Python to several other languages.  Misc/BLURB.WINDOWS
contains a blurb specifically aimed at Windows programmers (by Mark
Hammond).

- A new version of the Python mode for Emacs is included as
Misc/python-mode.el.  There are too many new features to list here.
See http://www.python.org/ftp/emacs/pmdetails.html for more info.

- New module fileinput makes iterating over the lines of a list of
files easier.  (This still needs some more thinking to make it more
extensible.)

- There's full OS/2 support, courtesy Jeff Rush.  To build the OS/2
version, see PC/readme.txt and PC/os2vacpp.  This is for IBM's Visual
Age C++ compiler.  I expect that Jeff will also provide a binary
release for this platform.

- On Linux, the configure script now uses '-Xlinker -export-dynamic'
instead of '-rdynamic' to link the main program so that it exports its
symbols to shared libraries it loads dynamically.  I hope this doesn't
break on older Linux versions; it is needed for mklinux and appears to
work on Linux 2.0.30.

- Some Tkinter resstructuring: the geometry methods that apply to a
master are now properly usable on toplevel master widgets.  There's a
new (internal) widget class, BaseWidget.  New, longer "official" names
for the geometry manager methods have been added,
e.g. "grid_columnconfigure()" instead of "columnconfigure()".  The old
shorter names still work, and where there's ambiguity, pack wins over
place wins over grid.  Also, the bind_class method now returns its
value.

- New, RFC-822 conformant parsing of email addresses and address lists
in the rfc822 module, courtesy Ben Escoto.

- New, revamped tkappinit.c with support for popular packages (PIL,
TIX, BLT, TOGL).  For the last three, you need to execute the Tcl
command "load {} Tix" (or Blt, or Togl) to gain access to them.
The Modules/Setup line for the _tkinter module has been rewritten
using the cool line-breaking feature of most Bourne shells.

- New socket method connect_ex() returns the error code from connect()
instead of raising an exception on errors; this makes the logic
required for asynchronous connects simpler and more efficient.

- New "locale" module with (still experimental) interface to the
standard C library locale interface, courtesy Martin von Loewis.  This
does not repeat my mistake in 1.5a4 of always calling
setlocale(LC_ALL, "").  In fact, we've pretty much decided that
Python's standard numerical formatting operations should always use
the conventions for the C locale; the locale module contains utility
functions to format numbers according to the user specified locale.
(All this is accomplished by an explicit call to setlocale(LC_NUMERIC,
"C") after locale-changing calls.)  See the library manual. (Alas, the
promised changes to the "re" module for locale support have not been
materialized yet.  If you care, volunteer!)

- Memory leak plugged in Py_BuildValue when building a dictionary.

- Shared modules can now live inside packages (hierarchical module
namespaces).  No changes to the shared module itself are needed.

- Improved policy for __builtins__: this is a module in __main__ and a
dictionary everywhere else.

- Python no longer catches SIGHUP and SIGTERM by default.  This was
impossible to get right in the light of thread contexts.  If you want
your program to clean up when a signal happens, use the signal module
to set up your own signal handler.

- New Python/C API PyNumber_CoerceEx() does not return an exception
when no coercion is possible.  This is used to fix a problem where
comparing incompatible numbers for equality would raise an exception
rather than return false as in Python 1.4 -- it once again will return
false.

- The errno module is changed again -- the table of error messages
(errorstr) is removed.  Instead, you can use os.strerror().  This
removes redundance and a potential locale dependency.

- New module xmllib, to parse XML files.  By Sjoerd Mullender.

- New C API PyOS_AfterFork() is called after fork() in posixmodule.c.
It resets the signal module's notion of what the current process ID
and thread are, so that signal handlers will work after (and across)
calls to os.fork().

- Fixed most occurrences of fatal errors due to missing thread state.

- For vgrind (a flexible source pretty printer) fans, there's a simple
Python definition in Misc/vgrindefs, courtesy Neale Pickett.

- Fixed memory leak in exec statement.

- The test.pystone module has a new function, pystones(loops=LOOPS),
which returns a (benchtime, stones) tuple.  The main() function now
calls this and prints the report.

- Package directories now *require* the presence of an __init__.py (or
__init__.pyc) file before they are considered as packages.  This is
done to prevent accidental subdirectories with common names from
overriding modules with the same name.

- Fixed some strange exceptions in __del__ methods in library modules
(e.g. urllib).  This happens because the builtin names are already
deleted by the time __del__ is called.  The solution (a hack, but it
works) is to set some instance variables to 0 instead of None.

- The table of built-in module initializers is replaced by a pointer
variable.  This makes it possible to switch to a different table at
run time, e.g. when a collection of modules is loaded from a shared
library.  (No example code of how to do this is given, but it is
possible.)  The table is still there of course, its name prefixed with
an underscore and used to initialize the pointer.

- The warning about a thread still having a frame now only happens in
verbose mode.

- Change the signal finialization so that it also resets the signal
handlers.  After this has been called, our signal handlers are no
longer active!

- New version of tokenize.py (by Ka-Ping Yee) recognizes raw string
literals.  There's now also a test fort this module.

- The copy module now also uses __dict__.update(state) instead of
going through individual attribute assignments, for class instances
without a __setstate__ method.

- New module reconvert translates old-style (regex module) regular
expressions to new-style (re module, Perl-style) regular expressions.

- Most modules that used to use the regex module now use the re
module.  The grep module has a new pgrep() function which uses
Perl-style regular expressions.

- The (very old, backwards compatibility) regexp.py module has been
deleted.

- Restricted execution (rexec): added the pcre module (support for the
re module) to the list of trusted extension modules.

- New version of Jim Fulton's CObject object type, adds
PyCObject_FromVoidPtrAndDesc() and PyCObject_GetDesc() APIs.

- Some patches to Lee Busby's fpectl mods that accidentally didn't
make it into 1.5a4.

- In the string module, add an optional 4th argument to count(),
matching find() etc.

- Patch for the nntplib module by Charles Waldman to add optional user
and password arguments to NNTP.__init__(), for nntp servers that need
them.

- The str() function for class objects now returns
"modulename.classname" instead of returning the same as repr().

- The parsing of \xXX escapes no longer relies on sscanf().

- The "sharedmodules" subdirectory of the installation is renamed to
"lib-dynload".  (You may have to edit your Modules/Setup file to fix
this in an existing installation!)

- Fixed Don Beaudry's mess-up with the OPT test in the configure
script.  Certain SGI platforms will still issue a warning for each
compile; there's not much I can do about this since the compiler's
exit status doesn't indicate that I was using an obsolete option.

- Fixed Barry's mess-up with {}.get(), and added test cases for it.

- Shared libraries didn't quite work under AIX because of the change
in status of the GNU readline interface.  Fix due to by Vladimir
Marangozov.


======================================================================


>From 1.5a3 to 1.5a4
===================

- faqwiz.py: version 0.8; Recognize https:// as URL; <html>...</html>
feature; better install instructions; removed faqmain.py (which was an
older version).

- nntplib.py: Fixed some bugs reported by Lars Wirzenius (to Debian)
about the treatment of lines starting with '.'.  Added a minimal test
function.

- struct module: ignore most whitespace in format strings.

- urllib.py: close the socket and temp file in URLopener.retrieve() so
that multiple retrievals using the same connection work.

- All standard exceptions are now classes by default; use -X to make
them strings (for backward compatibility only).

- There's a new standard exception hierarchy, defined in the standard
library module exceptions.py (which you never need to import
explicitly).  See
http://grail.cnri.reston.va.us/python/essays/stdexceptions.html for
more info.

- Three new C API functions:

  - int PyErr_GivenExceptionMatches(obj1, obj2)

    Returns 1 if obj1 and obj2 are the same object, or if obj1 is an
    instance of type obj2, or of a class derived from obj2

  - int PyErr_ExceptionMatches(obj)

    Higher level wrapper around PyErr_GivenExceptionMatches() which uses
    PyErr_Occurred() as obj1.  This will be the more commonly called
    function.

  - void PyErr_NormalizeException(typeptr, valptr, tbptr)

    Normalizes exceptions, and places the normalized values in the
    arguments.  If type is not a class, this does nothing.  If type is a
    class, then it makes sure that value is an instance of the class by:

    1. if instance is of the type, or a class derived from type, it does
       nothing.

    2. otherwise it instantiates the class, using the value as an
       argument.  If value is None, it uses an empty arg tuple, and if
       the value is a tuple, it uses just that.

- Another new C API function: PyErr_NewException() creates a new
exception class derived from Exception; when -X is given, it creates a
new string exception.

- core interpreter: remove the distinction between tuple and list
unpacking; allow an arbitrary sequence on the right hand side of any
unpack instruction.  (UNPACK_LIST and UNPACK_TUPLE now do the same
thing, which should really be called UNPACK_SEQUENCE.)

- classes: Allow assignments to an instance's __dict__ or __class__,
so you can change ivars (including shared ivars -- shock horror) and
change classes dynamically.  Also make the check on read-only
attributes of classes less draconic -- only the specials names
__dict__, __bases__, __name__ and __{get,set,del}attr__ can't be
assigned.

- Two new built-in functions: issubclass() and isinstance().  Both
take classes as their second arguments.  The former takes a class as
the first argument and returns true iff first is second, or is a
subclass of second.  The latter takes any object as the first argument
and returns true iff first is an instance of the second, or any
subclass of second.

- configure: Added configuration tests for presence of alarm(),
pause(), and getpwent().

- Doc/Makefile: changed latex2html targets.

- classes: Reverse the search order for the Don Beaudry hook so that
the first class with an applicable hook wins.  Makes more sense.

- Changed the checks made in Py_Initialize() and Py_Finalize().  It is
now legal to call these more than once.  The first call to
Py_Initialize() initializes, the first call to Py_Finalize()
finalizes.  There's also a new API, Py_IsInitalized() which checks
whether we are already initialized (in case you want to leave things
as they were).

- Completely disable the declarations for malloc(), realloc() and
free().  Any 90's C compiler has these in header files, and the tests
to decide whether to suppress the declarations kept failing on some
platforms.

- *Before* (instead of after) signalmodule.o is added, remove both
intrcheck.o and sigcheck.o.  This should get rid of warnings in ar or
ld on various systems.

- Added reop to PC/config.c

- configure: Decided to use -Aa -D_HPUX_SOURCE on HP-UX platforms.
Removed outdated HP-UX comments from README.  Added Cray T3E comments.

- Various renames of statically defined functions that had name
conflicts on some systems, e.g. strndup (GNU libc), join (Cray),
roundup (sys/types.h).

- urllib.py: Interpret three slashes in file: URL as local file (for
Netscape on Windows/Mac).

- copy.py: Make sure the objects returned by __getinitargs__() are
kept alive (in the memo) to avoid a certain kind of nasty crash.  (Not
easily reproducable because it requires a later call to
__getinitargs__() to return a tuple that happens to be allocated at
the same address.)

- Added definition of AR to toplevel Makefile.  Renamed @buildno temp
file to buildno1.

- Moved Include/assert.h to Parser/assert.h, which seems to be the
only place where it's needed.

- Tweaked the dictionary lookup code again for some more speed
(Vladimir Marangozov).

- NT build: Changed the way python15.lib is included in the other
projects.  Per Mark Hammond's suggestion, add it to the extra libs in
Settings instead of to the project's source files.

- regrtest.py: Change default verbosity so that there are only three
levels left: -q, default and -v.  In default mode, the name of each
test is now printed.  -v is the same as the old -vv.  -q is more quiet
than the old default mode.

- Removed the old FAQ from the distribution.  You now have to get it
from the web!

- Removed the PC/make_nt.in file from the distribution; it is no
longer needed.

- Changed the build sequence so that shared modules are built last.
This fixes things for AIX and doesn't hurt elsewhere.

- Improved test for GNU MP v1 in mpzmodule.c

- fileobject.c: ftell() on Linux discards all buffered data; changed
read() code to use lseek() instead to get the same effect

- configure.in, configure, importdl.c: NeXT sharedlib fixes

- tupleobject.c: PyTuple_SetItem asserts refcnt==1

- resource.c: Different strategy regarding whether to declare
getrusage() and getpagesize() -- #ifdef doesn't work, Linux has
conflicting decls in its headers.  Choice: only declare the return
type, not the argument prototype, and not on Linux.

- importdl.c, configure*: set sharedlib extensions properly for NeXT

- configure*, Makefile.in, Modules/Makefile.pre.in: AIX shared libraries
fixed; moved addition of PURIFY to LINKCC to configure

- reopmodule.c, regexmodule.c, regexpr.c, zlibmodule.c: needed casts
added to shup up various compilers.

- _tkinter.c: removed buggy mac #ifndef

- Doc: various Mac documentation changes, added docs for 'ic' module

- PC/make_nt.in: deleted

- test_time.py, test_strftime.py: tweaks to catch %Z (which may return
"")

- test_rotor.py: print b -> print `b`

- Tkinter.py: (tagOrId) -> (tagOrId,)

- Tkinter.py: the Tk class now also has a configure() method and
friends (they have been moved to the Misc class to accomplish this).

- dict.get(key[, default]) returns dict[key] if it exists, or default
if it doesn't.  The default defaults to None.  This is quicker for
some applications than using either has_key() or try:...except
KeyError:....

- Tools/webchecker/: some small changes to webchecker.py; added
websucker.py (a simple web site mirroring script).

- Dictionary objects now have a get() method (also in UserDict.py).
dict.get(key, default) returns dict[key] if it exists and default
otherwise; default defaults to None.

- Tools/scripts/logmerge.py: print the author, too.

- Changes to import: support for "import a.b.c" is now built in.  See
http://grail.cnri.reston.va.us/python/essays/packages.html
for more info.  Most important deviations from "ni.py": __init__.py is
executed in the package's namespace instead of as a submodule; and
there's no support for "__" or "__domain__".  Note that "ni.py" is not
changed to match this -- it is simply declared obsolete (while at the
same time, it is documented...:-( ).
Unfortunately, "ihooks.py" has not been upgraded (but see "knee.py"
for an example implementation of hierarchical module import written in
Python).

- More changes to import: the site.py module is now imported by
default when Python is initialized; use -S to disable it.  The site.py
module extends the path with several more directories: site-packages
inside the lib/python1.5/ directory, site-python in the lib/
directory, and pathnames mentioned in *.pth files found in either of
those directories.  See
http://grail.cnri.reston.va.us/python/essays/packages.html
for more info.

- Changes to standard library subdirectory names: those subdirectories
that are not packages have been renamed with a hypen in their name,
e.g. lib-tk, lib-stdwin, plat-win, plat-linux2, plat-sunos5, dos-8x3.
The test suite is now a package -- to run a test, you must now use
"import test.test_foo".

- A completely new re.py module is provided (thanks to Andrew
Kuchling, Tim Peters and Jeffrey Ollie) which uses Philip Hazel's
"pcre" re compiler and engine.  For a while, the "old" re.py (which
was new in 1.5a3!) will be kept around as re1.py.  The "old" regex
module and underlying parser and engine are still present -- while
regex is now officially obsolete, it will probably take several major
release cycles before it can be removed.

- The posix module now has a strerror() function which translates an
error code to a string.

- The emacs.py module (which was long obsolete) has been removed.

- The universal makefile Misc/Makefile.pre.in now features an
"install" target.  By default, installed shared libraries go into
$exec_prefix/lib/python$VERSION/site-packages/.

- The install-sh script is installed with the other configuration
specific files (in the config/ subdirectory).

- It turns out whatsound.py and sndhdr.py were identical modules.
Since there's also an imghdr.py file, I propose to make sndhdr.py the
official one.  For compatibility, whatsound.py imports * from
sndhdr.py.

- Class objects have a new attribute, __module__, giving the name of
the module in which they were declared.  This is useful for pickle and
for printing the full name of a class exception.

- Many extension modules no longer issue a fatal error when their
initialization fails; the importing code now checks whether an error
occurred during module initialization, and correctly propagates the
exception to the import statement.

- Most extension modules now raise class-based exceptions (except when
-X is used).

- Subtle changes to PyEval_{Save,Restore}Thread(): always swap the
thread state -- just don't manipulate the lock if it isn't there.

- Fixed a bug in Python/getopt.c that made it do the wrong thing when
an option was a single '-'.  Thanks to Andrew Kuchling.

- New module mimetypes.py will guess a MIME type from a filename's
extension.

- Windows: the DLL version is now settable via a resource rather than
being hardcoded.  This can be used for "branding" a binary Python
distribution.

- urllib.py is now threadsafe -- it now uses re instead of regex, and
sys.exc_info() instead of sys.exc_{type,value}.

- Many other library modules that used to use
sys.exc_{type,value,traceback} are now more thread-safe by virtue of
using sys.exc_info().

- The functions in popen2 have an optional buffer size parameter.
Also, the command argument can now be either a string (passed to the
shell) or a list of arguments (passed directly to execv).

- Alas, the thread support for _tkinter released with 1.5a3 didn't
work.  It's been rewritten.  The bad news is that it now requires a
modified version of a file in the standard Tcl distribution, which you
must compile with a -I option pointing to the standard Tcl source
tree.  For this reason, the thread support is disabled by default.

- The errno extension module adds two tables: errorcode maps errno
numbers to errno names (e.g. EINTR), and errorstr maps them to
message strings.  (The latter is redundant because the new call
posix.strerror() now does the same, but alla...)  (Marc-Andre Lemburg)

- The readline extension module now provides some interfaces to
internal readline routines that make it possible to write a completer
in Python.  An example completer, rlcompleter.py, is provided.

	When completing a simple identifier, it completes keywords,
	built-ins and globals in __main__; when completing
	NAME.NAME..., it evaluates (!) the expression up to the last
	dot and completes its attributes.

	It's very cool to do "import string" type "string.", hit the
	completion key (twice), and see the list of names defined by
	the string module!

	Tip: to use the tab key as the completion key, call

	    readline.parse_and_bind("tab: complete")

- The traceback.py module has a new function tb_lineno() by Marc-Andre
Lemburg which extracts the line number from the linenumber table in
the code object.  Apparently the traceback object doesn't contains the
right linenumber when -O is used.  Rather than guessing whether -O is
on or off, the module itself uses tb_lineno() unconditionally.

- Fixed Demo/tkinter/matt/canvas-moving-or-creating.py: change bind()
to tag_bind() so it works again.

- The pystone script is now a standard library module.  Example use:
"import test.pystone; test.pystone.main()".

- The import of the readline module in interactive mode is now also
attempted when -i is specified.  (Yes, I know, giving in to Marc-Andre
Lemburg, who asked for this. :-)

- rfc822.py: Entirely rewritten parseaddr() function by Sjoerd
Mullender, to be closer to the standard.  This fixes the getaddr()
method.  Unfortunately, getaddrlist() is as broken as ever, since it
splits on commas without regard for RFC 822 quoting conventions.

- pprint.py: correctly emit trailing "," in singleton tuples.

- _tkinter.c: export names for its type objects, TkappType and
TkttType.

- pickle.py: use __module__ when defined; fix a particularly hard to
reproduce bug that confuses the memo when temporary objects are
returned by custom pickling interfaces; and a semantic change: when
unpickling the instance variables of an instance, use
inst.__dict__.update(value) instead of a for loop with setattr() over
the value.keys().  This is more consistent (the pickling doesn't use
getattr() either but pickles inst.__dict__) and avoids problems with
instances that have a __setattr__ hook.  But it *is* a semantic change
(because the setattr hook is no longer used).  So beware!

- config.h is now installed (at last) in
$exec_prefix/include/python1.5/.  For most sites, this means that it
is actually in $prefix/include/python1.5/, with all the other Python
include files, since $prefix and $exec_prefix are the same by
default.

- The imp module now supports parts of the functionality to implement
import of hierarchical module names.  It now supports find_module()
and load_module() for all types of modules.  Docstrings have been
added for those functions in the built-in imp module that are still
relevant (some old interfaces are obsolete).  For a sample
implementation of hierarchical module import in Python, see the new
library module knee.py.

- The % operator on string objects now allows arbitrary nested parens
in a %(...)X style format.  (Brad Howes)

- Reverse the order in which Setup and Setup.local are passed to the
makesetup script.  This allows variable definitions in Setup.local to
override definitions in Setup.  (But you'll still have to edit Setup
if you want to disable modules that are enabled by default, or if such
modules need non-standard options.)

- Added PyImport_ImportModuleEx(name, globals, locals, fromlist); this
is like PyImport_ImporModule(name) but receives the globals and locals
dict and the fromlist arguments as well.  (The name is a char*; the
others are PyObject*s).

- The 'p' format in the struct extension module alloded to above is
new in 1.5a4.

- The types.py module now uses try-except in a few places to make it
more likely that it can be imported in restricted mode.  Some type
names are undefined in that case, e.g. CodeType (inaccessible),
FileType (not always accessible), and TracebackType and FrameType
(inaccessible).

- In urllib.py: added separate administration of temporary files
created y URLopener.retrieve() so cleanup() can properly remove them.
The old code removed everything in tempcache which was a bad idea if
the user had passed a non-temp file into it.  Also, in basejoin(),
interpret relative paths starting in "../".  This is necessary if the
server uses symbolic links.

- The Windows build procedure and project files are now based on
Microsoft Visual C++ 5.x.  The build now takes place in the PCbuild
directory.  It is much more robust, and properly builds separate Debug
and Release versions.  (The installer will be added shortly.)

- Added casts and changed some return types in regexpr.c to avoid
compiler warnings or errors on some platforms.

- The AIX build tools for shared libraries now supports VPATH.  (Donn
Cave)

- By default, disable the "portable" multimedia modules audioop,
imageop, and rgbimg, since they don't work on 64-bit platforms.

- Fixed a nasty bug in cStringIO.c when code was actually using the
close() method (the destructors would try to free certain fields a
second time).

- For those who think they need it, there's a "user.py" module.  This
is *not* imported by default, but can be imported to run user-specific
setup commands, ~/.pythonrc.py.

- Various speedups suggested by Fredrik Lundh, Marc-Andre Lemburg,
Vladimir Marangozov, and others.

- Added os.altsep; this is '/' on DOS/Windows, and None on systems
with a sane filename syntax.

- os.py: Write out the dynamic OS choice, to avoid exec statements.
Adding support for a new OS is now a bit more work, but I bet that
'dos' or 'nt' will cover most situations...

- The obsolete exception AccessError is now really gone.

- Tools/faqwiz/: New installation instructions show how to maintain
multiple FAQs.  Removed bootstrap script from end of faqwiz.py module.
Added instructions to bootstrap script, too.  Version bumped to 0.8.1.
Added <html>...</html> feature suggested by Skip Montanaro.  Added
leading text for Roulette, default to 'Hit Reload ...'.  Fix typo in
default SRCDIR.

- Documentation for the relatively new modules "keyword" and "symbol"
has been added (to the end of the section on the parser extension
module).

- In module bisect.py, but functions have two optional argument 'lo'
and 'hi' which allow you to specify a subsequence of the array to
operate on.

- In ftplib.py, changed most methods to return their status (even when
it is always "200 OK") rather than swallowing it.

- main() now calls setlocale(LC_ALL, ""), if setlocale() and
<locale.h> are defined.

- Changes to configure.in, the configure script, and both
Makefile.pre.in files, to support SGI's SGI_ABI platform selection
environment variable.


======================================================================


>From 1.4 to 1.5a3
=================

Security
--------

- If you are using the setuid script C wrapper (Misc/setuid-prog.c),
please use the new version.  The old version has a huge security leak.

Miscellaneous
-------------

- Because of various (small) incompatible changes in the Python
bytecode interpreter, the magic number for .pyc files has changed
again.

- The default module search path is now much saner.  Both on Unix and
Windows, it is essentially derived from the path to the executable
(which can be overridden by setting the environment variable
$PYTHONHOME).  The value of $PYTHONPATH on Windows is now inserted in
front of the default path, like in Unix (instead of overriding the
default path).  On Windows, the directory containing the executable is
added to the end of the path.

- A new version of python-mode.el for Emacs has been included.  Also,
a new file ccpy-style.el has been added to configure Emacs cc-mode for
the preferred style in Python C sources.

- On Unix, when using sys.argv[0] to insert the script directory in
front of sys.path, expand a symbolic link.  You can now install a
program in a private directory and have a symbolic link to it in a
public bin directory, and it will put the private directory in the
module search path.  Note that the symlink is expanded in sys.path[0]
but not in sys.argv[0], so you can still tell the name by which you
were invoked.

- It is now recommended to use ``#!/usr/bin/env python'' instead of
``#!/usr/local/bin/python'' at the start of executable scripts, except
for CGI scripts.  It has been determined that the use of /usr/bin/env
is more portable than that of /usr/local/bin/python -- scripts almost
never have to be edited when the Python interpreter lives in a
non-standard place.  Note that this doesn't work for CGI scripts since
the python executable often doesn't live in the HTTP server's default
search path.

- The silly -s command line option and the corresponding
PYTHONSUPPRESS environment variable (and the Py_SuppressPrint global
flag in the Python/C API) are gone.

- Most problems on 64-bit platforms should now be fixed.  Andrew
Kuchling helped.  Some uncommon extension modules are still not
clean (image and audio ops?).

- Fixed a bug where multiple anonymous tuple arguments would be mixed up
when using the debugger or profiler (reported by Just van Rossum).
The simplest example is ``def f((a,b),(c,d)): print a,b,c,d''; this
would print the wrong value when run under the debugger or profiler.

- The hacks that the dictionary implementation used to speed up
repeated lookups of the same C string were removed; these were a
source of subtle problems and don't seem to serve much of a purpose
any longer.

- All traces of support for the long dead access statement have been
removed from the sources.

- Plugged the two-byte memory leak in the tokenizer when reading an
interactive EOF.

- There's a -O option to the interpreter that removes SET_LINENO
instructions and assert statements (see below); it uses and produces
.pyo files instead of .pyc files.  The speedup is only a few percent
in most cases.  The line numbers are still available in the .pyo file,
as a separate table (which is also available in .pyc files).  However,
the removal of the SET_LINENO instructions means that the debugger
(pdb) can't set breakpoints on lines in -O mode.  The traceback module
contains a function to extract a line number from the code object
referenced in a traceback object.  In the future it should be possible
to write external bytecode optimizers that create better optimized
.pyo files, and there should be more control over optimization;
consider the -O option a "teaser".  Without -O, the assert statement
actually generates code that first checks __debug__; if this variable
is false, the assertion is not checked.  __debug__ is a built-in
variable whose value is initialized to track the -O flag (it's true
iff -O is not specified).  With -O, no code is generated for assert
statements, nor for code of the form ``if __debug__: <something>''.
Sorry, no further constant folding happens.


Performance
-----------

- It's much faster (almost twice for pystone.py -- see
Tools/scripts).  See the entry on string interning below.

- Some speedup by using separate free lists for method objects (both
the C and the Python variety) and for floating point numbers.

- Big speedup by allocating frame objects with a single malloc() call.
The Python/C API for frames is changed (you shouldn't be using this
anyway).

- Significant speedup by inlining some common opcodes for common operand 
types (e.g.  i+i, i-i, and list[i]).  Fredrik Lundh.

- Small speedup by reordering the method tables of some common
objects (e.g. list.append is now first).

- Big optimization to the read() method of file objects.  A read()
without arguments now attempts to use fstat to allocate a buffer of
the right size; for pipes and sockets, it will fall back to doubling
the buffer size.  While that the improvement is real on all systems,
it is most dramatic on Windows.


Documentation
-------------

- Many new pieces of library documentation were contributed, mostly by
Andrew Kuchling.  Even cmath is now documented!  There's also a
chapter of the library manual, "libundoc.tex", which provides a
listing of all undocumented modules, plus their status (e.g. internal,
obsolete, or in need of documentation).  Also contributions by Sue
Williams, Skip Montanaro, and some module authors who succumbed to
pressure to document their own contributed modules :-).  Note that
printing the documentation now kills fewer trees -- the margins have
been reduced.

- I have started documenting the Python/C API. Unfortunately this project 
hasn't been completed yet.  It will be complete before the final release of 
Python 1.5, though.  At the moment, it's better to read the LaTeX source 
than to attempt to run it through LaTeX and print the resulting dvi file.

- The posix module (and hence os.py) now has doc strings!  Thanks to Neil 
Schemenauer.  I received a few other contributions of doc strings.  In most 
other places, doc strings are still wishful thinking...


Language changes
----------------

- Private variables with leading double underscore are now a permanent 
feature of the language.  (These were experimental in release 1.4.  I have 
favorable experience using them; I can't label them "experimental" 
forever.)

- There's new string literal syntax for "raw strings".  Prefixing a string 
literal with the letter r (or R) disables all escape processing in the 
string; for example, r'\n' is a two-character string consisting of a 
backslash followed by the letter n.  This combines with all forms of string 
quotes; it is actually useful for triple quoted doc strings which might 
contain references to \n or \t.  An embedded quote prefixed with a 
backslash does not terminate the string, but the backslash is still 
included in the string; for example, r'\'' is a two-character string 
consisting of a backslash and a quote.  (Raw strings are also 
affectionately known as Robin strings, after their inventor, Robin 
Friedrich.)

- There's a simple assert statement, and a new exception
AssertionError.  For example, ``assert foo > 0'' is equivalent to ``if
not foo > 0: raise AssertionError''.  Sorry, the text of the asserted
condition is not available; it would be too complicated to generate
code for this (since the code is generated from a parse tree).
However, the text is displayed as part of the traceback!

- The raise statement has a new feature: when using "raise SomeClass,
somevalue" where somevalue is not an instance of SomeClass, it
instantiates SomeClass(somevalue).  In 1.5a4, if somevalue is an
instance of a *derived* class of SomeClass, the exception class raised
is set to somevalue.__class__, and SomeClass is ignored after that.

- Duplicate keyword arguments are now detected at compile time;
f(a=1,a=2) is now a syntax error.


Changes to builtin features
---------------------------

- There's a new exception FloatingPointError (used only by Lee Busby's
patches to catch floating point exceptions, at the moment).

- The obsolete exception ConflictError (presumably used by the long
obsolete access statement) has been deleted.

- There's a new function sys.exc_info() which returns the tuple 
(sys.exc_type, sys.exc_value, sys.exc_traceback) in a thread-safe way.

- There's a new variable sys.executable, pointing to the executable file 
for the Python interpreter.

- The sort() methods for lists no longer uses the C library qsort(); I
wrote my own quicksort implementation, with lots of help (in the form
of a kind of competition) from Tim Peters.  This solves a bug in
dictionary comparisons on some Solaris versions when Python is built
with threads, and makes sorting lists even faster.

- The semantics of comparing two dictionaries have changed, to make
comparison of unequal dictionaries faster.  A shorter dictionary is
always considered smaller than a larger dictionary.  For dictionaries
of the same size, the smallest differing element determines the
outcome (which yields the same results as before in this case, without
explicit sorting).  Thanks to Aaron Watters for suggesting something
like this.

- The semantics of try-except have changed subtly so that calling a
function in an exception handler that itself raises and catches an
exception no longer overwrites the sys.exc_* variables.  This also
alleviates the problem that objects referenced in a stack frame that
caught an exception are kept alive until another exception is caught
-- the sys.exc_* variables are restored to their previous value when
returning from a function that caught an exception.

- There's a new "buffer" interface.  Certain objects (e.g. strings and
arrays) now support the "buffer" protocol.  Buffer objects are acceptable 
whenever formerly a string was required for a write operation; mutable 
buffer objects can be the target of a read operation using the call
f.readinto(buffer).  A cool feature is that regular expression matching now 
also work on array objects.  Contribution by Jack Jansen.  (Needs 
documentation.)

- String interning: dictionary lookups are faster when the lookup
string object is the same object as the key in the dictionary, not
just a string with the same value.  This is done by having a pool of
"interned" strings.  Most names generated by the interpreter are now
automatically interned, and there's a new built-in function intern(s)
that returns the interned version of a string.  Interned strings are
not a different object type, and interning is totally optional, but by
interning most keys a speedup of about 15% was obtained for the
pystone benchmark.

- Dictionary objects have several new methods; clear() and copy() have
the obvious semantics, while update(d) merges the contents of another
dictionary d into this one, overriding existing keys.  The dictionary
implementation file is now called dictobject.c rather than the
confusing mappingobject.c.

- The intrinsic function dir() is much smarter; it looks in __dict__,
__members__ and __methods__.

- The intrinsic functions int(), long() and float() can now take a
string argument and then do the same thing as string.atoi(),
string.atol(), and string.atof().  No second 'base' argument is
allowed, and complex() does not take a string (nobody cared enough).

- When a module is deleted, its globals are now deleted in two phases.
In the first phase, all variables whose name begins with exactly one
underscore are replaced by None; in the second phase, all variables
are deleted.  This makes it possible to have global objects whose
destructors depend on other globals.  The deletion order within each
phase is still random.

- It is no longer an error for a function to be called without a
global variable __builtins__ -- an empty directory will be provided
by default.

- Guido's corollary to the "Don Beaudry hook": it is now possible to
do metaprogramming by using an instance as a base class.  Not for the
faint of heart; and undocumented as yet, but basically if a base class
is an instance, its class will be instantiated to create the new
class.  Jim Fulton will love it -- it also works with instances of his
"extension classes", since it is triggered by the presence of a
__class__ attribute on the purported base class.  See
Demo/metaclasses/index.html for an explanation and see that directory
for examples.

- Another change is that the Don Beaudry hook is now invoked when
*any* base class is special.  (Up to 1.5a3, the *last* special base
class is used; in 1.5a4, the more rational choice of the *first*
special base class is used.)

- New optional parameter to the readlines() method of file objects.
This indicates the number of bytes to read (the actual number of bytes
read will be somewhat larger due to buffering reading until the end of
the line).  Some optimizations have also been made to speed it up (but
not as much as read()).

- Complex numbers no longer have the ".conj" pseudo attribute; use
z.conjugate() instead, or complex(z.real, -z.imag).  Complex numbers
now *do* support the __members__ and __methods__ special attributes.

- The complex() function now looks for a __complex__() method on class
instances before giving up.

- Long integers now support arbitrary shift counts, so you can now
write 1L<<1000000, memory permitting.  (Python 1.4 reports "outrageous
shift count for this.)

- The hex() and oct() functions have been changed so that for regular
integers, they never emit a minus sign.  For example, on a 32-bit
machine, oct(-1) now returns '037777777777' and hex(-1) returns
'0xffffffff'.  While this may seem inconsistent, it is much more
useful.  (For long integers, a minus sign is used as before, to fit
the result in memory :-)

- The hash() function computes better hashes for several data types,
including strings, floating point numbers, and complex numbers.


New extension modules
---------------------

- New extension modules cStringIO.c and cPickle.c, written by Jim
Fulton and other folks at Digital Creations.  These are much more
efficient than their Python counterparts StringIO.py and pickle.py,
but don't support subclassing.  cPickle.c clocks up to 1000 times
faster than pickle.py; cStringIO.c's improvement is less dramatic but
still significant.

- New extension module zlibmodule.c, interfacing to the free zlib
library (gzip compatible compression).  There's also a module gzip.py
which provides a higher level interface.  Written by Andrew Kuchling
and Jeremy Hylton.

- New module readline; see the "miscellaneous" section above.

- New Unix extension module resource.c, by Jeremy Hylton, provides
access to getrlimit(), getrusage(), setrusage(), getpagesize(), and
related symbolic constants.

- New extension puremodule.c, by Barry Warsaw, which interfaces to the
Purify(TM) C API.  See also the file Misc/PURIFY.README.  It is also
possible to enable Purify by simply setting the PURIFY Makefile
variable in the Modules/Setup file.


Changes in extension modules
----------------------------

- The struct extension module has several new features to control byte
order and word size.  It supports reading and writing IEEE floats even
on platforms where this is not the native format.  It uses uppercase
format codes for unsigned integers of various sizes (always using
Python long ints for 'I' and 'L'), 's' with a size prefix for strings,
and 'p' for "Pascal strings" (with a leading length byte, included in
the size; blame Hannu Krosing; new in 1.5a4).  A prefix '>' forces
big-endian data and '<' forces little-endian data; these also select
standard data sizes and disable automatic alignment (use pad bytes as
needed).

- The array module supports uppercase format codes for unsigned data
formats (like the struct module).

- The fcntl extension module now exports the needed symbolic
constants.  (Formerly these were in FCNTL.py which was not available
or correct for all platforms.)

- The extension modules dbm, gdbm and bsddb now check that the
database is still open before making any new calls.

- The dbhash module is no more.  Use bsddb instead.  (There's a third
party interface for the BSD 2.x code somewhere on the web; support for
bsddb will be deprecated.)

- The gdbm module now supports a sync() method.

- The socket module now has some new functions: getprotobyname(), and
the set {ntoh,hton}{s,l}().

- Various modules now export their type object: socket.SocketType,
array.ArrayType.

- The socket module's accept() method now returns unknown addresses as
a tuple rather than raising an exception.  (This can happen in
promiscuous mode.)  Theres' also a new function getprotobyname().

- The pthread support for the thread module now works on most platforms.

- STDWIN is now officially obsolete.  Support for it will eventually
be removed from the distribution.

- The binascii extension module is now hopefully fully debugged.
(XXX Oops -- Fredrik Lundh promised me a uuencode fix that I never
received.)

- audioop.c: added a ratecv() function; better handling of overflow in
add().

- posixmodule.c: now exports the O_* flags (O_APPEND etc.).  On
Windows, also O_TEXT and O_BINARY.  The 'error' variable (the
exception is raises) is renamed -- its string value is now "os.error",
so newbies don't believe they have to import posix (or nt) to catch
it when they see os.error reported as posix.error.  The execve()
function now accepts any mapping object for the environment.

- A new version of the al (audio library) module for SGI was
contributed by Sjoerd Mullender.

- The regex module has a new function get_syntax() which retrieves the
syntax setting set by set_syntax().  The code was also sanitized,
removing worries about unclean error handling.  See also below for its
successor, re.py.

- The "new" module (which creates new objects of various types) once
again has a fully functioning new.function() method.  Dangerous as
ever!  Also, new.code() has several new arguments.

- A problem has been fixed in the rotor module: on systems with signed
characters, rotor-encoded data was not portable when the key contained
8-bit characters.  Also, setkey() now requires its argument rather
than having broken code to default it.

- The sys.builtin_module_names variable is now a tuple.  Another new
variables in sys is sys.executable (the full path to the Python
binary, if known).

- The specs for time.strftime() have undergone some revisions.  It
appears that not all format characters are supported in the same way
on all platforms.  Rather than reimplement it, we note these
differences in the documentation, and emphasize the shared set of
features.  There's also a thorough test set (that occasionally finds
problems in the C library implementation, e.g. on some Linuxes),
thanks to Skip Montanaro.

- The nis module seems broken when used with NIS+; unfortunately
nobody knows how to fix it.  It should still work with old NIS.


New library modules
-------------------

- New (still experimental) Perl-style regular expression module,
re.py, which uses a new interface for matching as well as a new
syntax; the new interface avoids the thread-unsafety of the regex
interface.  This comes with a helper extension reopmodule.c and vastly
rewritten regexpr.c.  Most work on this was done by Jeffrey Ollie, Tim
Peters, and Andrew Kuchling.  See the documentation libre.tex.  In
1.5, the old regex module is still fully supported; in the future, it
will become obsolete.

- New module gzip.py; see zlib above.

- New module keyword.py exports knowledge about Python's built-in
keywords.  (New version by Ka-Ping Yee.)

- New module pprint.py (with documentation) which supports
pretty-printing of lists, tuples, & dictionaries recursively.  By Fred
Drake.

- New module code.py.  The function code.compile_command() can
determine whether an interactively entered command is complete or not,
distinguishing incomplete from invalid input.  (XXX Unfortunately,
this seems broken at this moment, and I don't have the time to fix
it.  It's probably better to add an explicit interface to the parser
for this.)

- There is now a library module xdrlib.py which can read and write the
XDR data format as used by Sun RPC, for example.  It uses the struct
module.


Changes in library modules
--------------------------

- Module codehack.py is now completely obsolete.

- The pickle.py module has been updated to make it compatible with the
new binary format that cPickle.c produces.  By default it produces the
old all-ASCII format compatible with the old pickle.py, still much
faster than pickle.py; it will read both formats automatically.  A few
other updates have been made.

- A new helper module, copy_reg.py, is provided to register extensions
to the pickling code.

- Revamped module tokenize.py is much more accurate and has an
interface that makes it a breeze to write code to colorize Python
source code.  Contributed by Ka-Ping Yee.

- In ihooks.py, ModuleLoader.load_module() now closes the file under
all circumstances.

- The tempfile.py module has a new class, TemporaryFile, which creates
an open temporary file that will be deleted automatically when
closed.  This works on Windows and MacOS as well as on Unix.  (Jim
Fulton.)

- Changes to the cgi.py module: Most imports are now done at the
top of the module, which provides a speedup when using ni (Jim
Fulton).  The problem with file upload to a Windows platform is solved
by using the new tempfile.TemporaryFile class; temporary files are now
always opened in binary mode (Jim Fulton).  The cgi.escape() function
now takes an optional flag argument that quotes '"' to '&quot;'.  It
is now possible to invoke cgi.py from a command line script, to test
cgi scripts more easily outside an http server.  There's an optional
limit to the size of uploads to POST (Skip Montanaro).  Added a
'strict_parsing' option to all parsing functions (Jim Fulton).  The
function parse_qs() now uses urllib.unquote() on the name as well as
the value of fields (Clarence Gardner).  The FieldStorage class now
has a __len__() method.

- httplib.py: the socket object is no longer closed; all HTTP/1.*
responses are now accepted; and it is now thread-safe (by not using
the regex module).

- BaseHTTPModule.py: treat all HTTP/1.* versions the same.

- The popen2.py module is now rewritten using a class, which makes
access to the standard error stream and the process id of the
subprocess possible.

- Added timezone support to the rfc822.py module, in the form of a
getdate_tz() method and a parsedate_tz() function; also a mktime_tz().
Also added recognition of some non-standard date formats, by Lars
Wirzenius, and RFC 850 dates (Chris Lawrence).

- mhlib.py: various enhancements, including almost compatible parsing
of message sequence specifiers without invoking a subprocess.  Also
added a createmessage() method by Lars Wirzenius.

- The StringIO.StringIO class now supports readline(nbytes).  (Lars 
Wirzenius.)  (Of course, you should be using cStringIO for performance.)

- UserDict.py supports the new dictionary methods as well.

- Improvements for whrandom.py by Tim Peters: use 32-bit arithmetic to
speed it up, and replace 0 seed values by 1 to avoid degeneration.
A bug was fixed in the test for invalid arguments.

- Module ftplib.py: added support for parsing a .netrc file (Fred
Drake).  Also added an ntransfercmd() method to the FTP class, which
allows access to the expected size of a transfer when available, and a
parse150() function to the module which parses the corresponding 150
response.

- urllib.py: the ftp cache is now limited to 10 entries.  Added
quote_plus() and unquote_plus() functions which are like quote() and
unquote() but also replace spaces with '+' or vice versa, for
encoding/decoding CGI form arguments.  Catch all errors from the ftp
module.  HTTP requests now add the Host: header line.  The proxy
variable names are now mapped to lower case, for Windows.  The
spliturl() function no longer erroneously throws away all data past
the first newline.  The basejoin() function now intereprets "../"
correctly.  I *believe* that the problems with "exception raised in
__del__" under certain circumstances have been fixed (mostly by
changes elsewher in the interpreter).

- In urlparse.py, there is a cache for results in urlparse.urlparse();
its size limit is set to 20.  Also, new URL schemes shttp, https, and
snews are "supported".

- shelve.py: use cPickle and cStringIO when available.  Also added
a sync() method, which calls the database's sync() method if there is
one.

- The mimetools.py module now uses the available Python modules for
decoding quoted-printable, uuencode and base64 formats, rather than
creating a subprocess.

- The python debugger (pdb.py, and its base class bdb.py) now support
conditional breakpoints.  See the docs.

- The modules base64.py, uu.py and quopri.py can now be used as simple
command line utilities.

- Various small fixes to the nntplib.py module that I can't bother to
document in detail.

- Sjoerd Mullender's mimify.py module now supports base64 encoding and 
includes functions to handle the funny encoding you sometimes see in mail 
headers.  It is now documented.

- mailbox.py: Added BabylMailbox.  Improved the way the mailbox is
gotten from the environment.

- Many more modules now correctly open files in binary mode when this
is necessary on non-Unix platforms.

- The copying functions in the undocumented module shutil.py are
smarter.

- The Writer classes in the formatter.py module now have a flush()
method.

- The sgmllib.py module accepts hyphens and periods in the middle of
attribute names.  While this is against the SGML standard, there is
some HTML out there that uses this...

- The interface for the Python bytecode disassembler module, dis.py,
has been enhanced quite a bit.  There's now one main function,
dis.dis(), which takes almost any kind of object (function, module,
class, instance, method, code object) and disassembles it; without
arguments it disassembles the last frame of the last traceback.  The
other functions have changed slightly, too.

- The imghdr.py module recognizes new image types: BMP, PNG.

- The string.py module has a new function replace(str, old, new,
[maxsplit]) which does substring replacements.  It is actually
implemented in C in the strop module.  The functions [r]find() an
[r]index() have an optional 4th argument indicating the end of the
substring to search, alsoo implemented by their strop counterparts.
(Remember, never import strop -- import string uses strop when
available with zero overhead.)

- The string.join() function now accepts any sequence argument, not
just lists and tuples.

- The string.maketrans() requires its first two arguments to be
present.  The old version didn't require them, but there's not much
point without them, and the documentation suggests that they are
required, so we fixed the code to match the documentation.

- The regsub.py module has a function clear_cache(), which clears its
internal cache of compiled regular expressions.  Also, the cache now
takes the current syntax setting into account.  (However, this module
is now obsolete -- use the sub() or subn() functions or methods in the
re module.)

- The undocumented module Complex.py has been removed, now that Python
has built-in complex numbers.  A similar module remains as
Demo/classes/Complex.py, as an example.


Changes to the build process
----------------------------

- The way GNU readline is configured is totally different.  The
--with-readline configure option is gone.  It is now an extension
module, which may be loaded dynamically.  You must enable it (and
specify the correct linraries to link with) in the Modules/Setup file.
Importing the module installs some hooks which enable command line
editing.  When the interpreter shell is invoked interactively, it
attempts to import the readline module; when this fails, the default
input mechanism is used.  The hook variables are PyOS_InputHook and
PyOS_ReadlineFunctionPointer.  (Code contributed by Lee Busby, with
ideas from William Magro.)

- New build procedure: a single library, libpython1.5.a, is now built,
which contains absolutely everything except for a one-line main()
program (which calls Py_Main(argc, argv) to start the interpreter
shell).  This makes life much simpler for applications that need to
embed Python.  The serial number of the build is now included in the
version string (sys.version).

- As far as I can tell, neither gcc -Wall nor the Microsoft compiler
emits a single warning any more when compiling Python.

- A number of new Makefile variables have been added for special
situations, e.g. LDLAST is appended to the link command.  These are
used by editing the Makefile or passing them on the make command
line.

- A set of patches from Lee Busby has been integrated that make it
possible to catch floating point exceptions.  Use the configure option
--with-fpectl to enable the patches; the extension modules fpectl and
fpetest provide control to enable/disable and test the feature,
respectively.

- The support for shared libraries under AIX is now simpler and more
robust.  Thanks to Vladimir Marangozov for revamping his own patches!

- The Modules/makesetup script now reads a file Setup.local as well as
a file Setup.  Most changes to the Setup script can be done by editing
Setup.local instead, which makes it easier to carry a particular setup
over from one release to the next.

- The Modules/makesetup script now copies any "include" lines it
encounters verbatim into the output Makefile.  It also recognizes .cxx
and .cpp as C++ source files.

- The configure script is smarter about C compiler options; e.g. with
gcc it uses -O2 and -g when possible, and on some other platforms it
uses -Olimit 1500 to avoid a warning from the optimizer about the main
loop in ceval.c (which has more than 1000 basic blocks).

- The configure script now detects whether malloc(0) returns a NULL
pointer or a valid block (of length zero).  This avoids the nonsense
of always adding one byte to all malloc() arguments on most platforms.

- The configure script has a new option, --with-dec-threads, to enable
DEC threads on DEC Alpha platforms.  Also, --with-threads is now an
alias for --with-thread (this was the Most Common Typo in configure
arguments).

- Many changes in Doc/Makefile; amongst others, latex2html is now used
to generate HTML from all latex documents.


Change to the Python/C API
--------------------------

- Because some interfaces have changed, the PYTHON_API macro has been
bumped.  Most extensions built for the old API version will still run,
but I can't guarantee this.  Python prints a warning message on
version mismatches; it dumps core when the version mismatch causes a
serious problem :-)

- I've completed the Grand Renaming, with the help of Roger Masse and
Barry Warsaw.  This makes reading or debugging the code much easier.
Many other unrelated code reorganizations have also been carried out.
The allobjects.h header file is gone; instead, you would have to
include Python.h followed by rename2.h.  But you're better off running
Tools/scripts/fixcid.py -s Misc/RENAME on your source, so you can omit
the rename2.h; it will disappear in the next release.

- Various and sundry small bugs in the "abstract" interfaces have been
fixed.  Thanks to all the (involuntary) testers of the Python 1.4
version!  Some new functions have been added, e.g. PySequence_List(o),
equivalent to list(o) in Python.

- New API functions PyLong_FromUnsignedLong() and
PyLong_AsUnsignedLong().

- The API functions in the file cgensupport.c are no longer
supported.  This file has been moved to Modules and is only ever
compiled when the SGI specific 'gl' module is built.

- PyObject_Compare() can now raise an exception.  Check with
PyErr_Occurred().  The comparison function in an object type may also
raise an exception.

- The slice interface uses an upper bound of INT_MAX when no explicit
upper bound is given (e.x. for a[1:]).  It used to ask the object for
its length and do the calculations.

- Support for multiple independent interpreters.  See Doc/api.tex,
functions Py_NewInterpreter() and Py_EndInterpreter().  Since the
documentation is incomplete, also see the new Demo/pysvr example
(which shows how to use these in a threaded application) and the
source code.

- There is now a Py_Finalize() function which "de-initializes"
Python.  It is possible to completely restart the interpreter
repeatedly by calling Py_Finalize() followed by Py_Initialize().  A
change of functionality in Py_Initialize() means that it is now a
fatal error to call it while the interpreter is already initialized.
The old, half-hearted Py_Cleanup() routine is gone.  Use of Py_Exit()
is deprecated (it is nothing more than Py_Finalize() followed by
exit()).

- There are no known memory leaks left.  While Py_Finalize() doesn't
free *all* allocated memory (some of it is hard to track down),
repeated calls to Py_Finalize() and Py_Initialize() do not create
unaccessible heap blocks.

- There is now explicit per-thread state.  (Inspired by, but not the
same as, Greg Stein's free threading patches.)

- There is now better support for threading C applications.  There are
now explicit APIs to manipulate the interpreter lock.  Read the source
or the Demo/pysvr example; the new functions are
PyEval_{Acquire,Release}{Lock,Thread}().

- The test macro DEBUG has changed to Py_DEBUG, to avoid interference
with other libraries' DEBUG macros.  Likewise for any other test
macros that didn't yet start with Py_.

- New wrappers around malloc() and friends: Py_Malloc() etc. call
malloc() and call PyErr_NoMemory() when it fails; PyMem_Malloc() call
just malloc().  Use of these wrappers could be essential if multiple
memory allocators exist (e.g. when using certain DLL setups under
Windows).  (Idea by Jim Fulton.)

- New C API PyImport_Import() which uses whatever __import__() hook
that is installed for the current execution environment.  By Jim
Fulton.

- It is now possible for an extension module's init function to fail
non-fatally, by calling one of the PyErr_* functions and returning.

- The PyInt_AS_LONG() and PyFloat_AS_DOUBLE() macros now cast their
argument to the proper type, like the similar PyString macros already
did.  (Suggestion by Marc-Andre Lemburg.)  Similar for PyList_GET_SIZE
and PyList_GET_ITEM.

- Some of the Py_Get* function, like Py_GetVersion() (but not yet
Py_GetPath()) are now declared as returning a const char *.  (More
should follow.)

- Changed the run-time library to check for exceptions after object
comparisons.  PyObject_Compare() can now return an exception; use
PyErr_Occurred() to check (there is *no* special return value).

- PyFile_WriteString() and Py_Flushline() now return error indicators
instead of clearing exceptions.  This fixes an obscure bug where using
these would clear a pending exception, discovered by Just van Rossum.

- There's a new function, PyArg_ParseTupleAndKeywords(), which parses
an argument list including keyword arguments.  Contributed by Geoff
Philbrick.

- PyArg_GetInt() is gone.

- It's no longer necessary to include graminit.h when calling one of
the extended parser API functions.  The three public grammar start
symbols are now in Python.h as Py_single_input, Py_file_input, and
Py_eval_input.

- The CObject interface has a new function,
PyCObject_Import(module, name).  It calls PyCObject_AsVoidPtr()
on the object referenced by "module.name".


Tkinter
-------

- On popular demand, _tkinter once again installs a hook for readline
that processes certain Tk events while waiting for the user to type
(using PyOS_InputHook).

- A patch by Craig McPheeters plugs the most obnoxious memory leaks,
caused by command definitions referencing widget objects beyond their
lifetime.

- New standard dialog modules: tkColorChooser.py, tkCommonDialog.py,
tkMessageBox.py, tkFileDialog.py, tkSimpleDialog.py These interface
with the new Tk dialog scripts, and provide more "native platform"
style file selection dialog boxes on some platforms.  Contributed by
Fredrik Lundh.

- Tkinter.py: when the first Tk object is destroyed, it sets the
hiddel global _default_root to None, so that when another Tk object is
created it becomes the new default root.  Other miscellaneous
changes and fixes.

- The Image class now has a configure method.

- Added a bunch of new winfo options to Tkinter.py; we should now be
up to date with Tk 4.2.  The new winfo options supported are:
mananger, pointerx, pointerxy, pointery, server, viewable, visualid,
visualsavailable.

- The broken bind() method on Canvas objects defined in the Canvas.py
module has been fixed.  The CanvasItem and Group classes now also have
an unbind() method.

- The problem with Tkinter.py falling back to trying to import
"tkinter" when "_tkinter" is not found has been fixed -- it no longer
tries "tkinter", ever.  This makes diagnosing the problem "_tkinter
not configured" much easier and will hopefully reduce the newsgroup
traffic on this topic.

- The ScrolledText module once again supports the 'cnf' parameter, to
be compatible with the examples in Mark Lutz' book (I know, I know,
too late...)

- The _tkinter.c extension module has been revamped.  It now support
Tk versions 4.1 through 8.0; support for 4.0 has been dropped.  It
works well under Windows and Mac (with the latest Tk ports to those
platforms).  It also supports threading -- it is safe for one
(Python-created) thread to be blocked in _tkinter.mainloop() while
other threads modify widgets.  To make the changes visible, those
threads must use update_idletasks()method.  (The patch for threading
in 1.5a3 was broken; in 1.5a4, it is back in a different version,
which requires access to the Tcl sources to get it to work -- hence it
is disabled by default.)

- A bug in _tkinter.c has been fixed, where Split() with a string
containing an unmatched '"' could cause an exception or core dump.

- Unfortunately, on Windows and Mac, Tk 8.0 no longer supports
CreateFileHandler, so _tkinter.createfilehandler is not available on
those platforms when using Tk 8.0 or later.  I will have to rethink
how to interface with Tcl's lower-level event mechanism, or with its
channels (which are like Python's file-like objects).  Jack Jansen has
provided a fix for the Mac, so createfilehandler *is* actually
supported there; maybe I can adapt his fix for Windows.


Tools and Demos
---------------

- A new regression test suite is provided, which tests most of the
standard and built-in modules.  The regression test is run by invoking
the script Lib/test/regrtest.py.  Barry Warsaw wrote the test harnass;
he and Roger Masse contributed most of the new tests.

- New tool: faqwiz -- the CGI script that is used to maintain the
Python FAQ (http://grail.cnri.reston.va.us/cgi-bin/faqw.py).  In
Tools/faqwiz.

- New tool: webchecker -- a simple extensible web robot that, when
aimed at a web server, checks that server for dead links.  Available
are a command line utility as well as a Tkinter based GUI version.  In
Tools/webchecker.  A simplified version of this program is dissected
in my article in O'Reilly's WWW Journal, the issue on Scripting
Languages (Vol 2, No 2); Scripting the Web with Python (pp 97-120).
Includes a parser for robots.txt files by Skip Montanaro.

- New small tools: cvsfiles.py (prints a list of all files under CVS
n a particular directory tree), treesync.py (a rather Guido-specific
script to synchronize two source trees, one on Windows NT, the other
one on Unix under CVS but accessible from the NT box), and logmerge.py
(sort a collection of RCS or CVS logs by date).  In Tools/scripts.

- The freeze script now also works under Windows (NT).  Another
feature allows the -p option to be pointed at the Python source tree
instead of the installation prefix.  This was loosely based on part of
xfreeze by Sam Rushing and Bill Tutt.

- New examples (Demo/extend) that show how to use the generic
extension makefile (Misc/Makefile.pre.in).

- Tools/scripts/h2py.py now supports C++ comments.

- Tools/scripts/pystone.py script is upgraded to version 1.1; there
was a bug in version 1.0 (distributed with Python 1.4) that leaked
memory.  Also, in 1.1, the LOOPS variable is incremented to 10000.

- Demo/classes/Rat.py completely rewritten by Sjoerd Mullender.


Windows (NT and 95)
-------------------

- New project files for Developer Studio (Visual C++) 5.0 for Windows
NT (the old VC++ 4.2 Makefile is also still supported, but will
eventually be withdrawn due to its bulkiness).

- See the note on the new module search path in the "Miscellaneous" section 
above.

- Support for Win32s (the 32-bit Windows API under Windows 3.1) is
basically withdrawn.  If it still works for you, you're lucky.

- There's a new extension module, msvcrt.c, which provides various 
low-level operations defined in the Microsoft Visual C++ Runtime Library.  
These include locking(), setmode(), get_osfhandle(), set_osfhandle(), and 
console I/O functions like kbhit(), getch() and putch().

- The -u option not only sets the standard I/O streams to unbuffered
status, but also sets them in binary mode.  (This can also be done
using msvcrt.setmode(), by the way.)

- The, sys.prefix and sys.exec_prefix variables point to the directory 
where Python is installed, or to the top of the source tree, if it was run 
from there.

- The various os.path modules (posixpath, ntpath, macpath) now support
passing more than two arguments to the join() function, so
os.path.join(a, b, c) is the same as os.path.join(a, os.path.join(b,
c)).

- The ntpath module (normally used as os.path) supports ~ to $HOME 
expansion in expanduser().

- The freeze tool now works on Windows.

- See also the Tkinter category for a sad note on
_tkinter.createfilehandler().

- The truncate() method for file objects now works on Windows.

- Py_Initialize() is no longer called when the DLL is loaded.  You
must call it yourself.

- The time module's clock() function now has good precision through
the use of the Win32 API QueryPerformanceCounter().

- Mark Hammond will release Python 1.5 versions of PythonWin and his
other Windows specific code: the win32api extensions, COM/ActiveX
support, and the MFC interface.


Mac
---

- As always, the Macintosh port will be done by Jack Jansen.  He will
make a separate announcement for the Mac specific source code and the
binary distribution(s) when these are ready.


======================================================================


=====================================
==> Release 1.4 (October 25 1996) <==
=====================================

(Starting in reverse chronological order:)

- Changed disclaimer notice.

- Added SHELL=/bin/sh to Misc/Makefile.pre.in -- some Make versions
default to the user's login shell.

- In Lib/tkinter/Tkinter.py, removed bogus binding of <Delete> in Text
widget, and bogus bspace() function.

- In Lib/cgi.py, bumped __version__ to 2.0 and restored a truncated
paragraph.

- Fixed the NT Makefile (PC/vc40.mak) for VC 4.0 to set /MD for all
subprojects, and to remove the (broken) experimental NumPy
subprojects.

- In Lib/py_compile.py, cast mtime to long() so it will work on Mac
(where os.stat() returns mtimes as floats.)
- Set self.rfile unbuffered (like self.wfile) in SocketServer.py, to
fix POST in CGIHTTPServer.py.

- Version 2.83 of Misc/python-mode.el for Emacs is included.

- In Modules/regexmodule.c, fixed symcomp() to correctly handle a new
group starting immediately after a group tag.

- In Lib/SocketServer.py, changed the mode for rfile to unbuffered.

- In Objects/stringobject.c, fixed the compare function to do the
first char comparison in unsigned mode, for consistency with the way
other characters are compared by memcmp().

- In Lib/tkinter/Tkinter.py, fixed Scale.get() to support floats.

- In Lib/urllib.py, fix another case where openedurl wasn't set.

(XXX Sorry, the rest is in totally random order.  No time to fix it.)

- SyntaxError exceptions detected during code generation
(e.g. assignment to an expression) now include a line number.

- Don't leave trailing / or \ in script directory inserted in front of
sys.path.

- Added a note to Tools/scripts/classfix.py abouts its historical
importance.

- Added Misc/Makefile.pre.in, a universal Makefile for extensions
built outside the distribution.

- Rewritten Misc/faq2html.py, by Ka-Ping Yee.

- Install shared modules with mode 555 (needed for performance on some
platforms).

- Some changes to standard library modules to avoid calling append()
with more than one argument -- while supported, this should be
outlawed, and I don't want to set a bad example.

- bdb.py (and hence pdb.py) supports calling run() with a code object
instead of a code string.

- Fixed an embarrassing bug cgi.py which prevented correct uploading
of binary files from Netscape (which doesn't distinguish between
binary and text files).  Also added dormant logging support, which
makes it easier to debug the cgi module itself.

- Added default writer to constructor of NullFormatter class.

- Use binary mode for socket.makefile() calls in ftplib.py.

- The ihooks module no longer "installs" itself upon import -- this
was an experimental feature that helped ironing out some bugs but that
slowed down code that imported it without the need to install it
(e.g. the rexec module).  Also close the file in some cases and add
the __file__ attribute to loaded modules.

- The test program for mailbox.py is now more useful.

- Added getparamnames() to Message class in mimetools.py -- it returns
the names of parameters to the content-type header.

- Fixed a typo in ni that broke the loop stripping "__." from names.

- Fix sys.path[0] for scripts run via pdb.py's new main program.

- profile.py can now also run a script, like pdb.

- Fix a small bug in pyclbr -- don't add names starting with _ when
emulating from ... import *.

- Fixed a series of embarrassing typos in rexec's handling of standard
I/O redirection.  Added some more "safe" built-in modules: cmath,
errno, operator.

- Fixed embarrassing typo in shelve.py.

- Added SliceType and EllipsisType to types.py.

- In urllib.py, added handling for error 301 (same as 302); added
geturl() method to get the URL after redirection.

- Fixed embarrassing typo in xdrlib.py.  Also fixed typo in Setup.in
for _xdrmodule.c and removed redundant #include from _xdrmodule.c.

- Fixed bsddbmodule.c to add binary mode indicator on platforms that
have it.  This should make it working on Windows NT.

- Changed last uses of #ifdef NT to #ifdef MS_WINDOWS or MS_WIN32,
whatever applies.  Also rationalized some other tests for various MS
platforms.

- Added the sources for the NT installer script used for Python
1.4beta3.  Not tested with this release, but better than nothing.

- A compromise in pickle's defenses against Trojan horses: a
user-defined function is now okay where a class is expected.  A
built-in function is not okay, to prevent pickling something that
will execute os.system("rm -f *") when unpickling.

- dis.py will print the name of local variables referenced by local
load/store/delete instructions.

- Improved portability of SimpleHTTPServer module to non-Unix
platform.

- The thread.h interface adds an extra argument to down_sema().  This
only affects other C code that uses thread.c; the Python thread module
doesn't use semaphores (which aren't provided on all platforms where
Python threads are supported).  Note: on NT, this change is not
implemented.

- Fixed some typos in abstract.h; corrected signature of
PyNumber_Coerce, added PyMapping_DelItem.  Also fixed a bug in
abstract.c's PyObject_CallMethod().

- apply(classname, (), {}) now works even if the class has no
__init__() method.

- Implemented complex remainder and divmod() (these would dump core!).
Conversion of complex numbers to int, long int or float now raises an
exception, since there is no meaningful way to do it without losing
information.

- Fixed bug in built-in complex() function which gave the wrong result
for two real arguments.

- Change the hash algorithm for strings -- the multiplier is now
1000003 instead of 3, which gives better spread for short strings.

- New default path for Windows NT, the registry structure now supports
default paths for different install packages.  (Mark Hammond -- the
next PythonWin release will use this.)

- Added more symbols to the python_nt.def file.

- When using GNU readline, set rl_readline_name to "python".

- The Ellipses built-in name has been renamed to Ellipsis -- this is
the correct singular form.  Thanks to Ka-Ping Yee, who saved us from
eternal embarrassment.

- Bumped the PYTHON_API_VERSION to 1006, due to the Ellipses ->
Ellipsis name change.

- Updated the library reference manual.  Added documentation of
restricted mode (rexec, Bastion) and the formatter module (for use
with the htmllib module).  Fixed the documentation of htmllib
(finally).

- The reference manual is now maintained in FrameMaker.

- Upgraded scripts Doc/partparse.py and Doc/texi2html.py.

- Slight improvements to Doc/Makefile.

- Added fcntl.lockf(). This should be used for Unix file locking
instead of the posixfile module; lockf() is more portable.

- The getopt module now supports long option names, thanks to Lars
Wizenius.

- Plenty of changes to Tkinter and Canvas, mostly due to Fred Drake
and Nils Fischbeck.

- Use more bits of time.time() in whrandom's default seed().

- Performance hack for regex module's regs attribute.

- Don't close already closed socket in socket module.

- Correctly handle separators containing embedded nulls in
strop.split, strop.find and strop.rfind.  Also added more detail to
error message for strop.atoi and friends.

- Moved fallback definition for hypot() to Python/hypot.c.

- Added fallback definition for strdup, in Python/strdup.c.

- Fixed some bugs where a function would return 0 to indicate an error
where it should return -1.

- Test for error returned by time.localtime(), and rationalized its MS
tests.

- Added Modules/Setup.local file, which is processed after Setup.

- Corrected bug in toplevel Makefile.in -- execution of regen script
would not use the right PATH and PYTHONPATH.

- Various and sundry NeXT configuration changes (sigh).

- Support systems where libreadline needs neither termcap nor curses.

- Improved ld_so_aix script and python.exp file (for AIX).

- More stringent test for working <stdarg.h> in configure script.

- Removed Demo/www subdirectory -- it was totally out of date.

- Improved demos and docs for Fred Drake's parser module; fixed one
typo in the module itself.


=========================================
==> Release 1.4beta3 (August 26 1996) <==
=========================================


(XXX This is less readable that it should.  I promise to restructure
it for the final 1.4 release.)


What's new in 1.4beta3 (since beta2)?
-------------------------------------

- Name mangling to implement a simple form of class-private variables.
A name of the form "__spam" can't easily be used outside the class.
(This was added in 1.4beta3, but left out of the 1.4beta3 release
message.)

- In urllib.urlopen(): HTTP URLs containing user:passwd at host are now
handled correctly when using a proxy server.

- In ntpath.normpath(): don't truncate to 8+3 format.

- In mimetools.choose_boundary(): don't die when getuid() or getpid()
aren't defined.

- Module urllib: some optimizations to (un)quoting.

- New module MimeWriter for writing MIME documents.

- More changes to formatter module.

- The freeze script works once again and is much more robust (using
sys.prefix etc.).  It also supports a -o option to specify an
output directory.

- New module whichdb recognizes dbm, gdbm and bsddb/dbhash files.

- The Doc/Makefile targets have been reorganized somewhat to remove the 
insistence on always generating PostScript.

- The texinfo to html filter (Doc/texi2html.py) has been improved somewhat.

- "errors.h" has been renamed to "pyerrors.h" to resolve a long-standing 
name conflict on the Mac.

- Linking a module compiled with a different setting for Py_TRACE_REFS now 
generates a linker error rather than a core dump.

- The cgi module has a new convenience function print_exception(), which 
formats a python exception using HTML.  It also fixes a bug in the 
compatibility code and adds a dubious feature which makes it possible to 
have two query strings, one in the URL and one in the POST data.

- A subtle change in the unpickling of class instances makes it possible 
to unpickle in restricted execution mode, where the __dict__ attribute is 
not available (but setattr() is).

- Documentation for os.path.splitext() (== posixpath.splitext()) has been 
cleared up.  It splits at the *last* dot.

- posixfile locking is now also correctly supported on AIX.

- The tempfile module once again honors an initial setting of tmpdir.  It 
now works on Windows, too.

- The traceback module has some new functions to extract, format and print 
the active stack.

- Some translation functions in the urllib module have been made a little 
less sluggish.

- The addtag_* methods for Canvas widgets in Tkinter as well as in the 
separate Canvas class have been fixed so they actually do something 
meaningful.

- A tiny _test() function has been added to Tkinter.py.

- A generic Makefile for dynamically loaded modules is provided in the Misc 
subdirectory (Misc/gMakefile).

- A new version of python-mode.el for Emacs is provided.  See
http://www.python.org/ftp/emacs/pmdetails.html for details.  The
separate file pyimenu.el is no longer needed, imenu support is folded
into python-mode.el.

- The configure script can finally correctly find the readline library in a 
non-standard location.  The LDFLAGS variable is passed on the the Makefiles 
from the configure script.

- Shared libraries are now installed as programs (i.e. with executable 
permission).  This is required on HP-UX and won't hurt on other systems.

- The objc.c module is no longer part of the distribution.  Objective-C 
support may become available as contributed software on the ftp site.

- The sybase module is no longer part of the distribution.  A much
improved sybase module is available as contributed software from the
ftp site.

- _tkinter is now compatible with Tcl 7.5 / Tk 4.1 patch1 on Windows and 
Mac (don't use unpatched Tcl/Tk!).  The default line in the Setup.in file 
now links with Tcl 7.5 / Tk 4.1 rather than 7.4/4.0.

- In Setup, you can now write "*shared*" instead of "*noconfig*", and you 
can use *.so and *.sl as shared libraries.

- Some more fidgeting for AIX shared libraries.

- The mpz module is now compatible with GMP 2.x.  (Not tested by me.)
(Note -- a complete replacement by Niels Mo"ller, called gpmodule, is
available from the contrib directory on the ftp site.)

- A warning is written to sys.stderr when a __del__ method raises an 
exception (formerly, such exceptions were completely ignored).

- The configure script now defines HAVE_OLD_CPP if the C preprocessor is 
incapable of ANSI style token concatenation and stringification.

- All source files (except a few platform specific modules) are once again 
compatible with K&R C compilers as well as ANSI compilers.  In particular,
ANSI-isms have been removed or made conditional in complexobject.c, 
getargs.c and operator.c.

- The abstract object API has three new functions, PyObject_DelItem, 
PySequence_DelItem, and PySequence_DelSlice.

- The operator module has new functions delitem and delslice, and the 
functions "or" and "and" are renamed to "or_" and "and_" (since "or" and 
"and" are reserved words).  ("__or__" and "__and__" are unchanged.)

- The environment module is no longer supported; putenv() is now a function 
in posixmodule (also under NT).

- Error in filter(<function>, "") has been fixed.

- Unrecognized keyword arguments raise TypeError, not KeyError.

- Better portability, fewer bugs and memory leaks, fewer compiler warnings, 
some more documentation.

- Bug in float power boundary case (0.0 to the negative integer power) 
fixed.

- The test of negative number to the float power has been moved from the 
built-in pow() functin to floatobject.c (so complex numbers can yield the 
correct result).

- The bug introduced in beta2 where shared libraries loaded (using 
dlopen()) from the current directory would fail, has been fixed.

- Modules imported as shared libraries now also have a __file__ attribute, 
giving the filename from which they were loaded.  The only modules without 
a __file__ attribute now are built-in modules.

- On the Mac, dynamically loaded modules can end in either ".slb" or 
".<platform>.slb" where <platform> is either "CFM68K" or "ppc".  The ".slb" 
extension should only be used for "fat" binaries.

- C API addition: marshal.c now supports 
PyMarshal_WriteObjectToString(object).

- C API addition: getargs.c now supports
PyArg_ParseTupleAndKeywords(args, kwdict, format, kwnames, ...)
to parse keyword arguments.

- The PC versioning scheme (sys.winver) has changed once again.  the 
version number is now "<digit>.<digit>.<digit>.<apiversion>", where the 
first three <digit>s are the Python version (e.g. "1.4.0" for Python 1.4, 
"1.4.1" for Python 1.4.1 -- the beta level is not included) and 
<apiversion> is the four-digit PYTHON_API_VERSION (currently 1005).

- h2py.py accepts whitespace before the # in CPP directives

- On Solaris 2.5, it should now be possible to use either Posix threads or 
Solaris threads (XXX: how do you select which is used???).  (Note: the 
Python pthreads interface doesn't fully support semaphores yet -- anyone 
care to fix this?)

- Thread support should now work on AIX, using either DCE threads or 
pthreads.

- New file Demo/sockets/unicast.py

- Working Mac port, with CFM68K support, with Tk 4.1 support (though not 
both) (XXX)

- New project setup for PC port, now compatible with PythonWin, with 
_tkinter and NumPy support (XXX)

- New module site.py (XXX)

- New module xdrlib.py and optional support module _xdrmodule.c (XXX)

- parser module adapted to new grammar, complete w/ Doc & Demo (XXX)

- regen script fixed (XXX)

- new machdep subdirectories Lib/{aix3,aix4,next3_3,freebsd2,linux2} (XXX)

- testall now also tests math module (XXX)

- string.atoi c.s. now raise an exception for an empty input string.

- At last, it is no longer necessary to define HAVE_CONFIG_H in order to 
have config.h included at various places.

- Unrecognized keyword arguments now raise TypeError rather than KeyError.

- The makesetup script recognizes files with extension .so or .sl as
(shared) libraries.

- 'access' is no longer a reserved word, and all code related to its 
implementation is gone (or at least #ifdef'ed out).  This should make 
Python a little speedier too!

- Performance enhancements suggested by Sjoerd Mullender.  This includes 
the introduction of two new optional function pointers in type object, 
getattro and setattro, which are like getattr and setattr but take a 
string object instead of a C string pointer.

- New operations in string module: lstrip(s) and rstrip(s) strip whitespace 
only on the left or only on the right, A new optional third argument to 
split() specifies the maximum number of separators honored (so 
splitfields(s, sep, n) returns a list of at most n+1 elements).  (Since 
1.3, splitfields(s, None) is totally equivalent to split(s).)
string.capwords() has an optional second argument specifying the 
separator (which is passed to split()).

- regsub.split() has the same addition as string.split().  regsub.splitx(s, 
sep, maxsep) implements the functionality that was regsub.split(s, 1) in 
1.4beta2 (return a list containing the delimiters as well as the words).

- Final touch for AIX loading, rewritten Misc/AIX-NOTES.

- In Modules/_tkinter.c, when using Tk 4.1 or higher, use className
argument to _tkinter.create() to set Tcl's argv0 variable, so X
resources use the right resource class again.

- Add #undef fabs to Modules/mathmodule.c for macintosh.

- Added some macro renames for AIX in Modules/operator.c.

- Removed spurious 'E' from Doc/liberrno.tex.

- Got rid of some cruft in Misc/ (dlMakefile, pyimenu.el); added new
Misc/gMakefile and new version of Misc/python-mode.el.

- Fixed typo in Lib/ntpath.py (islink has "return false" which gives a
NameError).

- Added missing "from types import *" to Lib/tkinter/Canvas.py.

- Added hint about using default args for __init__ to pickle docs.

- Corrected typo in Inclide/abstract.h: PySequence_Lenth ->
PySequence_Length.

- Some improvements to Doc/texi2html.py.

- In Python/import.c, Cast unsigned char * in struct _frozen to char *
in calls to rds_object().

- In doc/ref4.tex, added note about scope of lambda bodies.

What's new in 1.4beta2 (since beta1)?
-------------------------------------

- Portability bug in the md5.h header solved.

- The PC build procedure now really works, and sets sys.platform to a
meaningful value (a few things were botched in beta 1).  Lib/dos_8x3
is now a standard part of the distribution (alas).

- More improvements to the installation procedure.  Typing "make install" 
now inserts the version number in the pathnames of almost everything 
installed, and creates the machine dependent modules (FCNTL.py etc.) if not 
supplied by the distribution.  (XXX There's still a problem with the latter 
because the "regen" script requires that Python is installed.  Some manual 
intervention may still be required.) (This has been fixed in 1.4beta3.)

- New modules: errno, operator (XXX).

- Changes for use with Numerical Python: builtin function slice() and
Ellipses object, and corresponding syntax:

	x[lo:hi:stride]		==	x[slice(lo, hi, stride)]
	x[a, ..., z]		==	x[(a, Ellipses, z)]

- New documentation for errno and cgi mdoules.

- The directory containing the script passed to the interpreter is
inserted in from of sys.path; "." is no longer a default path
component.

- Optional third string argument to string.translate() specifies
characters to delete.  New function string.maketrans() creates a
translation table for translate() or for regex.compile().

- Module posix (and hence module os under Unix) now supports putenv().
Moreover, module os is enhanced so that if putenv() is supported,
assignments to os.environ entries make the appropriate putenv() call.
(XXX the putenv() implementation can leak a small amount of memory per
call.)

- pdb.py can now be invoked from the command line to debug a script:
python pdb.py <script> <arg> ...

- Much improved parseaddr() in rfc822.

- In cgi.py, you can now pass an alternative value for environ to
nearly all functions.

- You can now assign to instance variables whose name begins and ends
with '__'.

- New version of Fred Drake's parser module and associates (token,
symbol, AST).

- New PYTHON_API_VERSION value and .pyc file magic number (again!).

- The "complex" internal structure type is now called "Py_complex" to
avoid name conflicts.

- Numerous small bugs fixed.

- Slight pickle speedups.

- Some slight speedups suggested by Sjoerd (more coming in 1.4 final).

- NeXT portability mods by Bill Bumgarner integrated.

- Modules regexmodule.c, bsddbmodule.c and xxmodule.c have been
converted to new naming style.


What's new in 1.4beta1 (since 1.3)?
-----------------------------------

- Added sys.platform and sys.exec_platform for Bill Janssen.

- Installation has been completely overhauled.  "make install" now installs 
everything, not just the python binary.  Installation uses the install-sh 
script (borrowed from X11) to install each file.

- New functions in the posix module: mkfifo, plock, remove (== unlink),
and ftruncate.  More functions are also available under NT.

- New function in the fcntl module: flock.

- Shared library support for FreeBSD.

- The --with-readline option can now be used without a DIRECTORY argument, 
for systems where libreadline.* is in one of the standard places.  It is 
also possible for it to be a shared library.

- The extension tkinter has been renamed to _tkinter, to avoid confusion 
with Tkinter.py oncase insensitive file systems.  It now supports Tk 4.1 as 
well as 4.0.

- Author's change of address from CWI in Amsterdam, The Netherlands, to 
CNRI in Reston, VA, USA.

- The math.hypot() function is now always available (if it isn't found in 
the C math library, Python provides its own implementation).

- The latex documentation is now compatible with latex2e, thanks to David 
Ascher.

- The expression x**y is now equivalent to pow(x, y).

- The indexing expression x[a, b, c] is now equivalent to x[(a, b, c)].

- Complex numbers are now supported.  Imaginary constants are written with 
a 'j' or 'J' prefix, general complex numbers can be formed by adding a real 
part to an imaginary part, like 3+4j.  Complex numbers are always stored in 
floating point form, so this is equivalent to 3.0+4.0j.  It is also 
possible to create complex numbers with the new built-in function 
complex(re, [im]).  For the footprint-conscious, complex number support can 
be disabled by defining the symbol WITHOUT_COMPLEX.

- New built-in function list() is the long-awaited counterpart of tuple().

- There's a new "cmath" module which provides the same functions as the 
"math" library but with complex arguments and results.  (There are very 
good reasons why math.sqrt(-1) still raises an exception -- you have to use 
cmath.sqrt(-1) to get 1j for an answer.)

- The Python.h header file (which is really the same as allobjects.h except 
it disables support for old style names) now includes several more files, 
so you have to have fewer #include statements in the average extension.

- The NDEBUG symbol is no longer used.  Code that used to be dependent on 
the presence of NDEBUG is now present on the absence of DEBUG.  TRACE_REFS 
and REF_DEBUG have been renamed to Py_TRACE_REFS and Py_REF_DEBUG, 
respectively.  At long last, the source actually compiles and links without 
errors when this symbol is defined.

- Several symbols that didn't follow the new naming scheme have been 
renamed (usually by adding to rename2.h) to use a Py or _Py prefix.  There 
are no external symbols left without a Py or _Py prefix, not even those 
defined by sources that were incorporated from elsewhere (regexpr.c, 
md5c.c).  (Macros are a different story...)

- There are now typedefs for the structures defined in config.c and 
frozen.c.

- New PYTHON_API_VERSION value and .pyc file magic number.

- New module Bastion.  (XXX)

- Improved performance of StringIO module.

- UserList module now supports + and * operators.

- The binhex and binascii modules now actually work.

- The cgi module has been almost totally rewritten and documented.
It now supports file upload and a new data type to handle forms more 
flexibly.

- The formatter module (for use with htmllib) has been overhauled (again).

- The ftplib module now supports passive mode and has doc strings.

- In (ideally) all places where binary files are read or written, the file 
is now correctly opened in binary mode ('rb' or 'wb') so the code will work 
on Mac or PC.

- Dummy versions of os.path.expandvars() and expanduser() are now provided 
on non-Unix platforms.

- Module urllib now has two new functions url2pathname and pathname2url 
which turn local filenames into "file:..." URLs using the same rules as 
Netscape (why be different).  it also supports urlretrieve() with a 
pathname parameter, and honors the proxy environment variables (http_proxy 
etc.).  The URL parsing has been improved somewhat, too.

- Micro improvements to urlparse.  Added urlparse.urldefrag() which 
removes a trailing ``#fragment'' if any.

- The mailbox module now supports MH style message delimiters as well.

- The mhlib module contains some new functionality: setcontext() to set the 
current folder and parsesequence() to parse a sequence as commonly passed 
to MH commands (e.g. 1-10 or last:5).

- New module mimify for conversion to and from MIME format of email 
messages.

- Module ni now automatically installs itself when first imported -- this 
is against the normal rule that modules should define classes and functions 
but not invoke them, but appears more useful in the case that two 
different, independent modules want to use ni's features.

- Some small performance enhancements in module pickle.

- Small interface change to the profile.run*() family of functions -- more 
sensible handling of return values.

- The officially registered Mac creator for Python files is 'Pyth'.  This 
replaces 'PYTH' which was used before but never registered.

- Added regsub.capwords().  (XXX)

- Added string.capwords(), string.capitalize() and string.translate().  
(XXX)

- Fixed an interface bug in the rexec module: it was impossible to pass a 
hooks instance to the RExec class.  rexec now also supports the dynamic 
loading of modules from shared libraries.  Some other interfaces have been 
added too.

- Module rfc822 now caches the headers in a dictionary for more efficient 
lookup.

- The sgmllib module now understands a limited number of SGML "shorthands" 
like <A/.../ for <A>...</A>.  (It's not clear that this was a good idea...)

- The tempfile module actually tries a number of different places to find a 
usable temporary directory.  (This was prompted by certain Linux 
installations that appear to be missing a /usr/tmp directory.) [A bug in 
the implementation that would ignore a pre-existing tmpdir global has been 
fixed in beta3.]

- Much improved and enhanved FileDialog module for Tkinter.

- Many small changes to Tkinter, to bring it more in line with Tk 4.0 (as 
well as Tk 4.1).

- New socket interfaces include ntohs(), ntohl(), htons(), htonl(), and 
s.dup().  Sockets now work correctly on Windows.  On Windows, the built-in 
extension is called _socket and a wrapper module win/socket.py provides 
"makefile()" and "dup()" functionality.  On Windows, the select module 
works only with socket objects.

- Bugs in bsddb module fixed (e.g. missing default argument values).

- The curses extension now includes <ncurses.h> when available.

- The gdbm module now supports opening databases in "fast" mode by 
specifying 'f' as the second character or the mode string.

- new variables sys.prefix and sys.exec_prefix pass corresponding 
configuration options / Makefile variables to the Python programmer.

- The ``new'' module now supports creating new user-defined classes as well 
as instances thereof.

- The soundex module now sports get_soundex() to get the soundex value for an 
arbitrary string (formerly it would only do soundex-based string 
comparison) as well as doc strings.

- New object type "cobject" to safely wrap void pointers for passing them 
between various extension modules.

- More efficient computation of float**smallint.

- The mysterious bug whereby "x.x" (two occurrences of the same 
one-character name) typed from the commandline would sometimes fail 
mysteriously.

- The initialization of the readline function can now be invoked by a C 
extension through PyOS_ReadlineInit().

- There's now an externally visible pointer PyImport_FrozenModules which 
can be changed by an embedding application.

- The argument parsing functions now support a new format character 'D' to 
specify complex numbers.

- Various memory leaks plugged and bugs fixed.

- Improved support for posix threads (now that real implementations are 
beginning to apepar).  Still no fully functioning semaphores.

- Some various and sundry improvements and new entries in the Tools 
directory.


=====================================
==> Release 1.3 (13 October 1995) <==
=====================================

Major change
============

Two words: Keyword Arguments.  See the first section of Chapter 12 of
the Tutorial.

(The rest of this file is textually the same as the remaining sections
of that chapter.)


Changes to the WWW and Internet tools
=====================================

The "htmllib" module has been rewritten in an incompatible fashion.
The new version is considerably more complete (HTML 2.0 except forms,
but including all ISO-8859-1 entity definitions), and easy to use.
Small changes to "sgmllib" have also been made, to better match the
tokenization of HTML as recognized by other web tools.

A new module "formatter" has been added, for use with the new
"htmllib" module.

The "urllib"and "httplib" modules have been changed somewhat to allow
overriding unknown URL types and to support authentication.  They now
use "mimetools.Message" instead of "rfc822.Message" to parse headers.
The "endrequest()" method has been removed from the HTTP class since
it breaks the interaction with some servers.

The "rfc822.Message" class has been changed to allow a flag to be
passed in that says that the file is unseekable.

The "ftplib" module has been fixed to be (hopefully) more robust on
Linux.

Several new operations that are optionally supported by servers have
been added to "nntplib": "xover", "xgtitle", "xpath" and "date".

Other Language Changes
======================

The "raise" statement now takes an optional argument which specifies
the traceback to be used when printing the exception's stack trace.
This must be a traceback object, such as found in "sys.exc_traceback".
When omitted or given as "None", the old behavior (to generate a stack
trace entry for the current stack frame) is used.

The tokenizer is now more tolerant of alien whitespace.  Control-L in
the leading whitespace of a line resets the column number to zero,
while Control-R just before the end of the line is ignored.

Changes to Built-in Operations
==============================

For file objects, "f.read(0)" and "f.readline(0)" now return an empty
string rather than reading an unlimited number of bytes.  For the
latter, omit the argument altogether or pass a negative value.

A new system variable, "sys.platform", has been added.  It specifies
the current platform, e.g. "sunos5" or "linux1".

The built-in functions "input()" and "raw_input()" now use the GNU
readline library when it has been configured (formerly, only
interactive input to the interpreter itself was read using GNU
readline).  The GNU readline library provides elaborate line editing
and history.  The Python debugger ("pdb") is the first beneficiary of
this change.

Two new built-in functions, "globals()" and "locals()", provide access
to dictionaries containming current global and local variables,
respectively.  (These augment rather than replace "vars()", which
returns the current local variables when called without an argument,
and a module's global variables when called with an argument of type
module.)

The built-in function "compile()" now takes a third possible value for
the kind of code to be compiled: specifying "'single'" generates code
for a single interactive statement, which prints the output of
expression statements that evaluate to something else than "None".

Library Changes
===============

There are new module "ni" and "ihooks" that support importing modules
with hierarchical names such as "A.B.C".  This is enabled by writing
"import ni; ni.ni()" at the very top of the main program.  These
modules are amply documented in the Python source.

The module "rexec" has been rewritten (incompatibly) to define a class
and to use "ihooks".

The "string.split()" and "string.splitfields()" functions are now the
same function (the presence or absence of the second argument
determines which operation is invoked); similar for "string.join()"
and "string.joinfields()".

The "Tkinter" module and its helper "Dialog" have been revamped to use
keyword arguments.  Tk 4.0 is now the standard.  A new module
"FileDialog" has been added which implements standard file selection
dialogs.

The optional built-in modules "dbm" and "gdbm" are more coordinated
--- their "open()" functions now take the same values for their "flag"
argument, and the "flag" and "mode" argument have default values (to
open the database for reading only, and to create the database with
mode "0666" minuse the umask, respectively).  The memory leaks have
finally been fixed.

A new dbm-like module, "bsddb", has been added, which uses the BSD DB
package's hash method.

A portable (though slow) dbm-clone, implemented in Python, has been
added for systems where none of the above is provided.  It is aptly
dubbed "dumbdbm".

The module "anydbm" provides a unified interface to "bsddb", "gdbm",
"dbm", and "dumbdbm", choosing the first one available.

A new extension module, "binascii", provides a variety of operations
for conversion of text-encoded binary data.

There are three new or rewritten companion modules implemented in
Python that can encode and decode the most common such formats: "uu"
(uuencode), "base64" and "binhex".

A module to handle the MIME encoding quoted-printable has also been
added: "quopri".

The parser module (which provides an interface to the Python parser's
abstract syntax trees) has been rewritten (incompatibly) by Fred
Drake.  It now lets you change the parse tree and compile the result!

The \code{syslog} module has been upgraded and documented.

Other Changes
=============

The dynamic module loader recognizes the fact that different filenames
point to the same shared library and loads the library only once, so
you can have a single shared library that defines multiple modules.
(SunOS / SVR4 style shared libraries only.)

Jim Fulton's ``abstract object interface'' has been incorporated into
the run-time API.  For more detailes, read the files
"Include/abstract.h" and "Objects/abstract.c".

The Macintosh version is much more robust now.

Numerous things I have forgotten or that are so obscure no-one will
notice them anyway :-)


===================================
==> Release 1.2 (13 April 1995) <==
===================================

- Changes to Misc/python-mode.el:
  - Wrapping and indentation within triple quote strings should work
    properly now.
  - `Standard' bug reporting mechanism (use C-c C-b)
  - py-mark-block was moved to C-c C-m
  - C-c C-v shows you the python-mode version
  - a basic python-font-lock-keywords has been added for Emacs 19
    font-lock colorizations.
  - proper interaction with pending-del and del-sel modes.
  - New py-electric-colon (:) command for improved outdenting.  Also
    py-indent-line (TAB) should handle outdented lines better.
  - New commands py-outdent-left (C-c C-l) and py-indent-right (C-c C-r)

- The Library Reference has been restructured, and many new and
existing modules are now documented, in particular the debugger and
the profiler, as well as the persistency and the WWW/Internet support
modules.

- All known bugs have been fixed.  For example the pow(2,2,3L) bug on
Linux has been fixed.  Also the re-entrancy problems with __del__ have
been fixed.

- All known memory leaks have been fixed.

- Phase 2 of the Great Renaming has been executed.  The header files
now use the new names (PyObject instead of object, etc.).  The linker
also sees the new names.  Most source files still use the old names,
by virtue of the rename2.h header file.  If you include Python.h, you
only see the new names.  Dynamically linked modules have to be
recompiled.  (Phase 3, fixing the rest of the sources, will be
executed gradually with the release later versions.)

- The hooks for implementing "safe-python" (better called "restricted
execution") are in place.  Specifically, the import statement is
implemented by calling the built-in function __import__, and the
built-in names used in a particular scope are taken from the
dictionary __builtins__ in that scope's global dictionary.  See also
the new (unsupported, undocumented) module rexec.py.

- The import statement now supports the syntax "import a.b.c" and
"from a.b.c import name".  No officially supported implementation
exists, but one can be prototyped by replacing the built-in __import__
function.  A proposal by Ken Manheimer is provided as newimp.py.

- All machinery used by the import statement (or the built-in
__import__ function) is now exposed through the new built-in module
"imp" (see the library reference manual).  All dynamic loading
machinery is moved to the new file importdl.c.

- Persistent storage is supported through the use of the modules
"pickle" and "shelve" (implemented in Python).  There's also a "copy"
module implementing deepcopy and normal (shallow) copy operations.
See the library reference manual.

- Documentation strings for many objects types are accessible through
the __doc__ attribute.  Modules, classes and functions support special
syntax to initialize the __doc__ attribute: if the first statement
consists of just a string literal, that string literal becomes the
value of the __doc__ attribute.  The default __doc__ attribute is
None.  Documentation strings are also supported for built-in
functions, types and modules; however this feature hasn't been widely
used yet.  See the 'new' module for an example.  (Basically, the type
object's tp_doc field contains the doc string for the type, and the
4th member of the methodlist structure contains the doc string for the
method.)

- The __coerce__ and __cmp__ methods for user-defined classes once
again work as expected.  As an example, there's a new standard class
Complex in the library.

- The functions posix.popen() and posix.fdopen() now have an optional
third argument to specify the buffer size, and default their second
(mode) argument to 'r' -- in analogy to the builtin open() function.
The same applies to posixfile.open() and the socket method makefile().

- The thread.exit_thread() function now raises SystemExit so that
'finally' clauses are honored and a memory leak is plugged.

- Improved X11 and Motif support, by Sjoerd Mullender.  This extension
is being maintained and distributed separately.

- Improved support for the Apple Macintosh, in part by Jack Jansen,
e.g. interfaces to (a few) resource mananger functions, get/set file
type and creator, gestalt, sound manager, speech manager, MacTCP, comm
toolbox, and the think C console library.  This is being maintained
and distributed separately.

- Improved version for Windows NT, by Mark Hammond.  This is being
maintained and distributed separately.

- Used autoconf 2.0 to generate the configure script.  Adapted
configure.in to use the new features in autoconf 2.0.

- It now builds on the NeXT without intervention, even on the 3.3
Sparc pre-release.

- Characters passed to isspace() and friends are masked to nonnegative
values.

- Correctly compute pow(-3.0, 3).

- Fix portability problems with getopt (configure now checks for a
non-GNU getopt).

- Don't add frozenmain.o to libPython.a.

- Exceptions can now be classes.  ALl built-in exceptions are still
string objects, but this will change in the future.

- The socket module exports a long list of socket related symbols.
(More built-in modules will export their symbolic constants instead of
relying on a separately generated Python module.)

- When a module object is deleted, it clears out its own dictionary.
This fixes a circularity in the references between functions and
their global dictionary.

- Changed the error handling by [new]getargs() e.g. for "O&".

- Dynamic loading of modules using shared libraries is supported for
several new platforms.

- Support "O&", "[...]" and "{...}" in mkvalue().

- Extension to findmethod(): findmethodinchain() (where a chain is a
linked list of methodlist arrays).  The calling interface for
findmethod() has changed: it now gets a pointer to the (static!)
methodlist structure rather than just to the function name -- this
saves copying flags etc. into the (short-lived) method object.

- The callable() function is now public.

- Object types can define a few new operations by setting function
pointers in the type object structure: tp_call defines how an object
is called, and tp_str defines how an object's str() is computed.


===================================
==> Release 1.1.1 (10 Nov 1994) <==
===================================

This is a pure bugfix release again.  See the ChangeLog file for details.

One exception: a few new features were added to tkinter.


=================================
==> Release 1.1 (11 Oct 1994) <==
=================================

This release adds several new features, improved configuration and
portability, and fixes more bugs than I can list here (including some
memory leaks).

The source compiles and runs out of the box on more platforms than
ever -- including Windows NT.  Makefiles or projects for a variety of
non-UNIX platforms are provided.

APOLOGY: some new features are badly documented or not at all.  I had
the choice -- postpone the new release indefinitely, or release it
now, with working code but some undocumented areas.  The problem with
postponing the release is that people continue to suffer from existing
bugs, and send me patches based on the previous release -- which I
can't apply directly because my own source has changed.  Also, some
new modules (like signal) have been ready for release for quite some
time, and people are anxiously waiting for them.  In the case of
signal, the interface is simple enough to figure out without
documentation (if you're anxious enough :-).  In this case it was not
simple to release the module on its own, since it relies on many small
patches elsewhere in the source.

For most new Python modules, the source code contains comments that
explain how to use them.  Documentation for the Tk interface, written
by Matt Conway, is available as tkinter-doc.tar.gz from the Python
home and mirror ftp sites (see Misc/FAQ for ftp addresses).  For the
new operator overloading facilities, have a look at Demo/classes:
Complex.py and Rat.py show how to implement a numeric type without and
with __coerce__ method.  Also have a look at the end of the Tutorial
document (Doc/tut.tex).  If you're still confused: use the newsgroup
or mailing list.


New language features:

    - More flexible operator overloading for user-defined classes
    (INCOMPATIBLE WITH PREVIOUS VERSIONS!)  See end of tutorial.

    - Classes can define methods named __getattr__, __setattr__ and
    __delattr__ to trap attribute accesses.  See end of tutorial.

    - Classes can define method __call__ so instances can be called
    directly.  See end of tutorial.


New support facilities:

    - The Makefiles (for the base interpreter as well as for extensions)
    now support creating dynamically loadable modules if the platform
    supports shared libraries.

    - Passing the interpreter a .pyc file as script argument will execute
    the code in that file.  (On the Mac such files can be double-clicked!)

    - New Freeze script, to create independently distributable "binaries"
    of Python programs -- look in Demo/freeze

    - Improved h2py script (in Demo/scripts) follows #includes and
    supports macros with one argument

    - New module compileall generates .pyc files for all modules in a
    directory (tree) without also executing them

    - Threads should work on more platforms


New built-in modules:

    - tkinter (support for Tcl's Tk widget set) is now part of the base
    distribution

    - signal allows catching or ignoring UNIX signals (unfortunately still
    undocumented -- any taker?)

    - termios provides portable access to POSIX tty settings

    - curses provides an interface to the System V curses library

    - syslog provides an interface to the (BSD?) syslog daemon

    - 'new' provides interfaces to create new built-in object types
    (e.g. modules and functions)

    - sybase provides an interface to SYBASE database


New/obsolete built-in methods:

    - callable(x) tests whether x can be called

    - sockets now have a setblocking() method

    - sockets no longer have an allowbroadcast() method

    - socket methods send() and sendto() return byte count


New standard library modules:

    - types.py defines standard names for built-in types, e.g. StringType

    - urlparse.py parses URLs according to the latest Internet draft

    - uu.py does uuencode/uudecode (not the fastest in the world, but
    quicker than installing uuencode on a non-UNIX machine :-)

    - New, faster and more powerful profile module.py

    - mhlib.py provides interface to MH folders and messages


New facilities for extension writers (unfortunately still
undocumented):

    - newgetargs() supports optional arguments and improved error messages

    - O!, O& O? formats for getargs allow more versatile type checking of
    non-standard types

    - can register pending asynchronous callback, to be called the next
    time the Python VM begins a new instruction (Py_AddPendingCall)

    - can register cleanup routines to be called when Python exits
    (Py_AtExit)

    - makesetup script understands C++ files in Setup file (use file.C
    or file.cc)

    - Make variable OPT is passed on to sub-Makefiles

    - An init<module>() routine may signal an error by not entering
    the module in the module table and raising an exception instead

    - For long module names, instead of foobarbletchmodule.c you can
    use foobarbletch.c

    - getintvalue() and getfloatvalue() try to convert any object
    instead of requiring an "intobject" or "floatobject"

    - All the [new]getargs() formats that retrieve an integer value
    will now also work if a float is passed

    - C function listtuple() converts list to tuple, fast

    - You should now call sigcheck() instead of intrcheck();
    sigcheck() also sets an exception when it returns nonzero


====================================
==> Release 1.0.3 (14 July 1994) <==
====================================

This release consists entirely of bug fixes to the C sources; see the
head of ../ChangeLog for a complete list.  Most important bugs fixed:

- Sometimes the format operator (string%expr) would drop the last
character of the format string

- Tokenizer looped when last line did not end in \n

- Bug when triple-quoted string ended in quote plus newline

- Typo in socketmodule (listen) (== instead of =)

- typing vars() at the >>> prompt would cause recursive output


==================================
==> Release 1.0.2 (4 May 1994) <==
==================================

Overview of the most visible changes.  Bug fixes are not listed.  See
also ChangeLog.

Tokens
------

* String literals follow Standard C rules: they may be continued on
the next line using a backslash; adjacent literals are concatenated
at compile time.

* A new kind of string literals, surrounded by triple quotes (""" or
'''), can be continued on the next line without a backslash.

Syntax
------

* Function arguments may have a default value, e.g. def f(a, b=1);
defaults are evaluated at function definition time.  This also applies
to lambda.

* The try-except statement has an optional else clause, which is
executed when no exception occurs in the try clause.

Interpreter
-----------

* The result of a statement-level expression is no longer printed,
except_ for expressions entered interactively.  Consequently, the -k
command line option is gone.

* The result of the last printed interactive expression is assigned to
the variable '_'.

* Access to implicit global variables has been speeded up by removing
an always-failing dictionary lookup in the dictionary of local
variables (mod suggested by Steve Makewski and Tim Peters).

* There is a new command line option, -u, to force stdout and stderr
to be unbuffered.

* Incorporated Steve Majewski's mods to import.c for dynamic loading
under AIX.

* Fewer chances of dumping core when trying to reload or re-import
static built-in, dynamically loaded built-in, or frozen modules.

* Loops over sequences now don't ask for the sequence's length when
they start, but try to access items 0, 1, 2, and so on until they hit
an IndexError.  This makes it possible to create classes that generate
infinite or indefinite sequences a la Steve Majewski.  This affects
for loops, the (not) in operator, and the built-in functions filter(),
map(), max(), min(), reduce().

Changed Built-in operations
---------------------------

* The '%' operator on strings (printf-style formatting) supports a new
feature (adapted from a patch by Donald Beaudry) to allow
'%(<key>)<format>' % {...} to take values from a dictionary by name
instead of from a tuple by position (see also the new function
vars()).

* The '%s' formatting operator is changed to accept any type and
convert it to a string using str().

* Dictionaries with more than 20,000 entries can now be created
(thanks to Steve Kirsch).

New Built-in Functions
----------------------

* vars() returns a dictionary containing the local variables; vars(m)
returns a dictionary containing the variables of module m.  Note:
dir(x) is now equivalent to vars(x).keys().

Changed Built-in Functions
--------------------------

* open() has an optional third argument to specify the buffer size: 0
for unbuffered, 1 for line buffered, >1 for explicit buffer size, <0
for default.

* open()'s second argument is now optional; it defaults to "r".

* apply() now checks that its second argument is indeed a tuple.

New Built-in Modules
--------------------

Changed Built-in Modules
------------------------

The thread module no longer supports exit_prog().

New Python Modules
------------------

* Module addpack contains a standard interface to modify sys.path to
find optional packages (groups of related modules).

* Module urllib contains a number of functions to access
World-Wide-Web files specified by their URL.

* Module httplib implements the client side of the HTTP protocol used
by World-Wide-Web servers.

* Module gopherlib implements the client side of the Gopher protocol.

* Module mailbox (by Jack Jansen) contains a parser for UNIX and MMDF
style mailbox files.

* Module random contains various random distributions, e.g. gauss().

* Module lockfile locks and unlocks open files using fcntl (inspired
by a similar module by Andy Bensky).

* Module ntpath (by Jaap Vermeulen) implements path operations for
Windows/NT.

* Module test_thread (in Lib/test) contains a small test set for the
thread module.

Changed Python Modules
----------------------

* The string module's expandvars() function is now documented and is
implemented in Python (using regular expressions) instead of forking
off a shell process.

* Module rfc822 now supports accessing the header fields using the
mapping/dictionary interface, e.g. h['subject'].

* Module pdb now makes it possible to set a break on a function
(syntax: break <expression>, where <expression> yields a function
object).

Changed Demos
-------------

* The Demo/scripts/freeze.py script is working again (thanks to Jaap
Vermeulen).

New Demos
---------

* Demo/threads/Generator.py is a proposed interface for restartable
functions a la Tim Peters.

* Demo/scripts/newslist.py, by Quentin Stafford-Fraser, generates a
directory full of HTML pages which between them contain links to all
the newsgroups available on your server.

* Demo/dns contains a DNS (Domain Name Server) client.

* Demo/lutz contains miscellaneous demos by Mark Lutz (e.g. psh.py, a
nice enhanced Python shell!!!).

* Demo/turing contains a Turing machine by Amrit Prem.

Documentation
-------------

* Documented new language features mentioned above (but not all new
modules).

* Added a chapter to the Tutorial describing recent additions to
Python.

* Clarified some sentences in the reference manual,
e.g. break/continue, local/global scope, slice assignment.

Source Structure
----------------

* Moved Include/tokenizer.h to Parser/tokenizer.h.

* Added Python/getopt.c for systems that don't have it.

Emacs mode
----------

* Indentation of continuated lines is done more intelligently;
consequently the variable py-continuation-offset is gone.

========================================
==> Release 1.0.1 (15 February 1994) <==
========================================

* Many portability fixes should make it painless to build Python on
several new platforms, e.g. NeXT, SEQUENT, WATCOM, DOS, and Windows.

* Fixed test for <stdarg.h> -- this broke on some platforms.

* Fixed test for shared library dynalic loading -- this broke on SunOS
4.x using the GNU loader.

* Changed order and number of SVR4 networking libraries (it is now
-lsocket -linet -lnsl, if these libraries exist).

* Installing the build intermediate stages with "make libainstall" now
also installs config.c.in, Setup and makesetup, which are used by the
new Extensions mechanism.

* Improved README file contains more hints and new troubleshooting
section.

* The built-in module strop now defines fast versions of three more
functions of the standard string module: atoi(), atol() and atof().
The strop versions of atoi() and atol() support an optional second
argument to specify the base (default 10).  NOTE: you don't have to
explicitly import strop to use the faster versions -- the string
module contains code to let versions from stop override the default
versions.

* There is now a working Lib/dospath.py for those who use Python under
DOS (or Windows).  Thanks, Jaap!

* There is now a working Modules/dosmodule.c for DOS (or Windows)
system calls.

* Lib.os.py has been reorganized (making it ready for more operating
systems).

* Lib/ospath.py is now obsolete (use os.path instead).

* Many fixes to the tutorial to make it match Python 1.0.  Thanks,
Tim!

* Fixed Doc/Makefile, Doc/README and various scripts there.

* Added missing description of fdopen to Doc/libposix.tex.

* Made cleanup() global, for the benefit of embedded applications.

* Added parsing of addresses and dates to Lib/rfc822.py.

* Small fixes to Lib/aifc.py, Lib/sunau.py, Lib/tzparse.py to make
them usable at all.

* New module Lib/wave.py reads RIFF (*.wav) audio files.

* Module Lib/filewin.py moved to Lib/stdwin/filewin.py where it
belongs.

* New options and comments for Modules/makesetup (used by new
Extension mechanism).

* Misc/HYPE contains text of announcement of 1.0.0 in comp.lang.misc
and elsewhere.

* Fixed coredump in filter(None, 'abcdefg').


=======================================
==> Release 1.0.0 (26 January 1994) <==
=======================================

As is traditional, so many things have changed that I can't pretend to
be complete in these release notes, but I'll try anyway :-)

Note that the very last section is labeled "remaining bugs".


Source organization and build process
-------------------------------------

* The sources have finally been split: instead of a single src
subdirectory there are now separate directories Include, Parser,
Grammar, Objects, Python and Modules.  Other directories also start
with a capital letter: Misc, Doc, Lib, Demo.

* A few extensions (notably Amoeba and X support) have been moved to a
separate subtree Extensions, which is no longer in the core
distribution, but separately ftp'able as extensions.tar.Z.  (The
distribution contains a placeholder Ext-dummy with a description of
the Extensions subtree as well as the most recent versions of the
scripts used there.)

* A few large specialized demos (SGI video and www) have been
moved to a separate subdirectory Demo2, which is no longer in the core
distribution, but separately ftp'able as demo2.tar.Z.

* Parts of the standard library have been moved to subdirectories:
there are now standard subdirectories stdwin, test, sgi and sun4.

* The configuration process has radically changed: I now use GNU
autoconf.  This makes it much easier to build on new Unix flavors, as
well as fully supporting VPATH (if your Make has it).  The scripts
Configure.py and Addmodule.sh are no longer needed.  Many source files
have been adapted in order to work with the symbols that the configure
script generated by autoconf defines (or not); the resulting source is
much more portable to different C compilers and operating systems,
even non Unix systems (a Mac port was done in an afternoon).  See the
toplevel README file for a description of the new build process.

* GNU readline (a slightly newer version) is now a subdirectory of the
Python toplevel.  It is still not automatically configured (being
totally autoconf-unaware :-).  One problem has been solved: typing
Control-C to a readline prompt will now work.  The distribution no
longer contains a "super-level" directory (above the python toplevel
directory), and dl, dl-dld and GNU dld are no longer part of the
Python distribution (you can still ftp them from
ftp.cwi.nl:/pub/dynload).

* The DOS functions have been taken out of posixmodule.c and moved
into a separate file dosmodule.c.

* There's now a separate file version.c which contains nothing but
the version number.

* The actual main program is now contained in config.c (unless NO_MAIN
is defined); pythonmain.c now contains a function realmain() which is
called from config.c's main().

* All files needed to use the built-in module md5 are now contained in
the distribution.  The module has been cleaned up considerably.


Documentation
-------------

* The library manual has been split into many more small latex files,
so it is easier to edit Doc/lib.tex file to create a custom library
manual, describing only those modules supported on your system.  (This
is not automated though.)

* A fourth manual has been added, titled "Extending and Embedding the
Python Interpreter" (Doc/ext.tex), which collects information about
the interpreter which was previously spread over several files in the
misc subdirectory.

* The entire documentation is now also available on-line for those who
have a WWW browser (e.g. NCSA Mosaic).  Point your browser to the URL
"http://www.cwi.nl/~guido/Python.html".


Syntax
------

* Strings may now be enclosed in double quotes as well as in single
quotes.  There is no difference in interpretation.  The repr() of
string objects will use double quotes if the string contains a single
quote and no double quotes.  Thanks to Amrit Prem for these changes!

* There is a new keyword 'exec'.  This replaces the exec() built-in
function.  If a function contains an exec statement, local variable
optimization is not performed for that particular function, thus
making assignment to local variables in exec statements less
confusing.  (As a consequence, os.exec and python.exec have been
renamed to execv.)

* There is a new keyword 'lambda'.  An expression of the form

	lambda <parameters> : <expression>

yields an anonymous function.  This is really only syntactic sugar;
you can just as well define a local function using

	def some_temporary_name(<parameters>): return <expression>

Lambda expressions are particularly useful in combination with map(),
filter() and reduce(), described below.  Thanks to Amrit Prem for
submitting this code (as well as map(), filter(), reduce() and
xrange())!


Built-in functions
------------------

* The built-in module containing the built-in functions is called
__builtin__ instead of builtin.

* New built-in functions map(), filter() and reduce() perform standard
functional programming operations (though not lazily):

- map(f, seq) returns a new sequence whose items are the items from
seq with f() applied to them.

- filter(f, seq) returns a subsequence of seq consisting of those
items for which f() is true.

- reduce(f, seq, initial) returns a value computed as follows:
	acc = initial
	for item in seq: acc = f(acc, item)
	return acc

* New function xrange() creates a "range object".  Its arguments are
the same as those of range(), and when used in a for loop a range
objects also behaves identical.  The advantage of xrange() over
range() is that its representation (if the range contains many
elements) is much more compact than that of range().  The disadvantage
is that the result cannot be used to initialize a list object or for
the "Python idiom" [RED, GREEN, BLUE] = range(3).  On some modern
architectures, benchmarks have shown that "for i in range(...): ..."
actually executes *faster* than "for i in xrange(...): ...", but on
memory starved machines like PCs running DOS range(100000) may be just
too big to be represented at all...

* Built-in function exec() has been replaced by the exec statement --
see above.


The interpreter
---------------

* Syntax errors are now not printed to stderr by the parser, but
rather the offending line and other relevant information are packed up
in the SyntaxError exception argument.  When the main loop catches a
SyntaxError exception it will print the error in the same format as
previously, but at the proper position in the stack traceback.

* You can now set a maximum to the number of traceback entries
printed by assigning to sys.tracebacklimit.  The default is 1000.

* The version number in .pyc files has changed yet again.

* It is now possible to have a .pyc file without a corresponding .py
file.  (Warning: this may break existing installations if you have an
old .pyc file lingering around somewhere on your module search path
without a corresponding .py file, when there is a .py file for a
module of the same name further down the path -- the new interpreter
will find the first .pyc file and complain about it, while the old
interpreter would ignore it and use the .py file further down.)

* The list sys.builtin_module_names is now sorted and also contains
the names of a few hardwired built-in modules (sys, __main__ and
__builtin__).

* A module can now find its own name by accessing the global variable
__name__.  Assigning to this variable essentially renames the module
(it should also be stored under a different key in sys.modules).
A neat hack follows from this: a module that wants to execute a main
program when called as a script no longer needs to compare
sys.argv[0]; it can simply do "if __name__ == '__main__': main()".

* When an object is printed by the print statement, its implementation
of str() is used.  This means that classes can define __str__(self) to
direct how their instances are printed.  This is different from
__repr__(self), which should define an unambigous string
representation of the instance.  (If __str__() is not defined, it
defaults to __repr__().)

* Functions and code objects can now be compared meaningfully.

* On systems supporting SunOS or SVR4 style shared libraries, dynamic
loading of modules using shared libraries is automatically configured.
Thanks to Bill Jansen and Denis Severson for contributing this change!


Built-in objects
----------------

* File objects have acquired a new method writelines() which is the
reverse of readlines().  (It does not actually write lines, just a
list of strings, but the symmetry makes the choice of name OK.)


Built-in modules
----------------

* Socket objects no longer support the avail() method.  Use the select
module instead, or use this function to replace it:

	def avail(f):
		import select
		return f in select.select([f], [], [], 0)[0]

* Initialization of stdwin is done differently.  It actually modifies
sys.argv (taking out the options the X version of stdwin recognizes)
the first time it is imported.

* A new built-in module parser provides a rudimentary interface to the
python parser.  Corresponding standard library modules token and symbol
defines the numeric values of tokens and non-terminal symbols.

* The posix module has aquired new functions setuid(), setgid(),
execve(), and exec() has been renamed to execv().

* The array module is extended with 8-byte object swaps, the 'i'
format character, and a reverse() method.  The read() and write()
methods are renamed to fromfile() and tofile().

* The rotor module has freed of portability bugs.  This introduces a
backward compatibility problem: strings encoded with the old rotor
module can't be decoded by the new version.

* For select.select(), a timeout (4th) argument of None means the same
as leaving the timeout argument out.

* Module strop (and hence standard library module string) has aquired
a new function: rindex().  Thanks to Amrit Prem!

* Module regex defines a new function symcomp() which uses an extended
regular expression syntax: parenthesized subexpressions may be labeled
using the form "\(<labelname>...\)", and the group() method can return
sub-expressions by name.  Thanks to Tracy Tims for these changes!

* Multiple threads are now supported on Solaris 2.  Thanks to Sjoerd
Mullender!


Standard library modules
------------------------

* The library is now split in several subdirectories: all stuff using
stdwin is in Lib/stdwin, all SGI specific (or SGI Indigo or GL) stuff
is in Lib/sgi, all Sun Sparc specific stuff is in Lib/sun4, and all
test modules are in Lib/test.  The default module search path will
include all relevant subdirectories by default.

* Module os now knows about trying to import dos.  It defines
functions execl(), execle(), execlp() and execvp().

* New module dospath (should be attacked by a DOS hacker though).

* All modules defining classes now define __init__() constructors
instead of init() methods.  THIS IS AN INCOMPATIBLE CHANGE!

* Some minor changes and bugfixes module ftplib (mostly Steve
Majewski's suggestions); the debug() method is renamed to
set_debuglevel().

* Some new test modules (not run automatically by testall though):
test_audioop, test_md5, test_rgbimg, test_select.

* Module string now defines rindex() and rfind() in analogy of index()
and find().  It also defines atof() and atol() (and corresponding
exceptions) in analogy to atoi().

* Added help() functions to modules profile and pdb.

* The wdb debugger (now in Lib/stdwin) now shows class or instance
variables on a double click.  Thanks to Sjoerd Mullender!

* The (undocumented) module lambda has gone -- you couldn't import it
any more, and it was basically more a demo than a library module...


Multimedia extensions
---------------------

* The optional built-in modules audioop and imageop are now standard
parts of the interpreter.  Thanks to Sjoerd Mullender and Jack Jansen
for contributing this code!

* There's a new operation in audioop: minmax().

* There's a new built-in module called rgbimg which supports portable
efficient reading of SGI RCG image files.  Thanks also to Paul
Haeberli for the original code!  (Who will contribute a GIF reader?)

* The module aifc is gone -- you should now always use aifc, which has
received a facelift.

* There's a new module sunau., for reading Sun (and NeXT) audio files.

* There's a new module audiodev which provides a uniform interface to
(SGI Indigo and Sun Sparc) audio hardware.

* There's a new module sndhdr which recognizes various sound files by
looking in their header and checking for various magic words.


Optimizations
-------------

* Most optimizations below can be configured by compile-time flags.
Thanks to Sjoerd Mullender for submitting these optimizations!

* Small integers (default -1..99) are shared -- i.e. if two different
functions compute the same value it is possible (but not
guaranteed!!!) that they return the same *object*.  Python programs
can detect this but should *never* rely on it.

* Empty tuples (which all compare equal) are shared in the same
manner.

* Tuples of size up to 20 (default) are put in separate free lists
when deallocated.

* There is a compile-time option to cache a string's hash function,
but this appeared to have a negligeable effect, and as it costs 4
bytes per string it is disabled by default.


Embedding Python
----------------

* The initialization interface has been simplified somewhat.  You now
only call "initall()" to initialize the interpreter.

* The previously announced renaming of externally visible identifiers
has not been carried out.  It will happen in a later release.  Sorry.


Miscellaneous bugs that have been fixed
---------------------------------------

* All known portability bugs.

* Version 0.9.9 dumped core in <listobject>.sort() which has been
fixed.  Thanks to Jaap Vermeulen for fixing this and posting the fix
on the mailing list while I was away!

* Core dump on a format string ending in '%', e.g. in the expression
'%' % None.

* The array module yielded a bogus result for concatenation (a+b would
yield a+a).

* Some serious memory leaks in strop.split() and strop.splitfields().

* Several problems with the nis module.

* Subtle problem when copying a class method from another class
through assignment (the method could not be called).


Remaining bugs
--------------

* One problem with 64-bit machines remains -- since .pyc files are
portable and use only 4 bytes to represent an integer object, 64-bit
integer literals are silently truncated when written into a .pyc file.
Work-around: use eval('123456789101112').

* The freeze script doesn't work any more.  A new and more portable
one can probably be cooked up using tricks from Extensions/mkext.py.

* The dos support hasn't been tested yet.  (Really Soon Now we should
have a PC with a working C compiler!)


===================================
==> Release 0.9.9 (29 Jul 1993) <==
===================================

I *believe* these are the main user-visible changes in this release,
but there may be others.  SGI users may scan the {src,lib}/ChangeLog
files for improvements of some SGI specific modules, e.g. aifc and
cl.  Developers of extension modules should also read src/ChangeLog.


Naming of C symbols used by the Python interpreter
--------------------------------------------------

* This is the last release using the current naming conventions.  New
naming conventions are explained in the file misc/NAMING.
Summarizing, all externally visible symbols get (at least) a "Py"
prefix, and most functions are renamed to the standard form
PyModule_FunctionName.

* Writers of extensions are urged to start using the new naming
conventions.  The next release will use the new naming conventions
throughout (it will also have a different source directory
structure).

* As a result of the preliminary work for the great renaming, many
functions that were accidentally global have been made static.


BETA X11 support
----------------

* There are now modules interfacing to the X11 Toolkit Intrinsics, the
Athena widgets, and the Motif 1.1 widget set.  These are not yet
documented except through the examples and README file in the demo/x11
directory.  It is expected that this interface will be replaced by a
more powerful and correct one in the future, which may or may not be
backward compatible.  In other words, this part of the code is at most
BETA level software!  (Note: the rest of Python is rock solid as ever!)

* I understand that the above may be a bit of a disappointment,
however my current schedule does not allow me to change this situation
before putting the release out of the door.  By releasing it
undocumented and buggy, at least some of the (working!) demo programs,
like itr (my Internet Talk Radio browser) become available to a larger
audience.

* There are also modules interfacing to SGI's "Glx" widget (a GL
window wrapped in a widget) and to NCSA's "HTML" widget (which can
format HyperText Markup Language, the document format used by the
World Wide Web).

* I've experienced some problems when building the X11 support.  In
particular, the Xm and Xaw widget sets don't go together, and it
appears that using X11R5 is better than using X11R4.  Also the threads
module and its link time options may spoil things.  My own strategy is
to build two Python binaries: one for use with X11 and one without
it, which can contain a richer set of built-in modules.  Don't even
*think* of loading the X11 modules dynamically...


Environmental changes
---------------------

* Compiled files (*.pyc files) created by this Python version are
incompatible with those created by the previous version.  Both
versions detect this and silently create a correct version, but it
means that it is not a good idea to use the same library directory for
an old and a new interpreter, since they will start to "fight" over
the *.pyc files...

* When a stack trace is printed, the exception is printed last instead
of first.  This means that if the beginning of the stack trace
scrolled out of your window you can still see what exception caused
it.

* Sometimes interrupting a Python operation does not work because it
hangs in a blocking system call.  You can now kill the interpreter by
interrupting it three times.  The second time you interrupt it, a
message will be printed telling you that the third interrupt will kill
the interpreter.  The "sys.exitfunc" feature still makes limited
clean-up possible in this case.


Changes to the command line interface
-------------------------------------

* The python usage message is now much more informative.

* New option -i enters interactive mode after executing a script --
useful for debugging.

* New option -k raises an exception when an expression statement
yields a value other than None.

* For each option there is now also a corresponding environment
variable.


Using Python as an embedded language
------------------------------------

* The distribution now contains (some) documentation on the use of
Python as an "embedded language" in other applications, as well as a
simple example.  See the file misc/EMBEDDING and the directory embed/.


Speed improvements
------------------

* Function local variables are now generally stored in an array and
accessed using an integer indexing operation, instead of through a
dictionary lookup.  (This compensates the somewhat slower dictionary
lookup caused by the generalization of the dictionary module.)


Changes to the syntax
---------------------

* Continuation lines can now *sometimes* be written without a
backslash: if the continuation is contained within nesting (), [] or
{} brackets the \ may be omitted.  There's a much improved
python-mode.el in the misc directory which knows about this as well.

* You can no longer use an empty set of parentheses to define a class
without base classes.  That is, you no longer write this:

	class Foo(): # syntax error
		...

You must write this instead:

	class Foo:
		...

This was already the preferred syntax in release 0.9.8 but many
people seemed not to have picked it up.  There's a Python script that
fixes old code: demo/scripts/classfix.py.

* There's a new reserved word: "access".  The syntax and semantics are
still subject of of research and debate (as well as undocumented), but
the parser knows about the keyword so you must not use it as a
variable, function, or attribute name.


Changes to the semantics of the language proper
-----------------------------------------------

* The following compatibility hack is removed: if a function was
defined with two or more arguments, and called with a single argument
that was a tuple with just as many arguments, the items of this tuple
would be used as the arguments.  This is no longer supported.


Changes to the semantics of classes and instances
-------------------------------------------------

* Class variables are now also accessible as instance variables for
reading (assignment creates an instance variable which overrides the
class variable of the same name though).

* If a class attribute is a user-defined function, a new kind of
object is returned: an "unbound method".  This contains a pointer to
the class and can only be called with a first argument which is a
member of that class (or a derived class).

* If a class defines a method __init__(self, arg1, ...) then this
method is called when a class instance is created by the classname()
construct.  Arguments passed to classname() are passed to the
__init__() method.  The __init__() methods of base classes are not
automatically called; the derived __init__() method must call these if
necessary (this was done so the derived __init__() method can choose
the call order and arguments for the base __init__() methods).

* If a class defines a method __del__(self) then this method is called
when an instance of the class is about to be destroyed.  This makes it
possible to implement clean-up of external resources attached to the
instance.  As with __init__(), the __del__() methods of base classes
are not automatically called.  If __del__ manages to store a reference
to the object somewhere, its destruction is postponed; when the object
is again about to be destroyed its __del__() method will be called
again.

* Classes may define a method __hash__(self) to allow their instances
to be used as dictionary keys.  This must return a 32-bit integer.


Minor improvements
------------------

* Function and class objects now know their name (the name given in
the 'def' or 'class' statement that created them).

* Class instances now know their class name.


Additions to built-in operations
--------------------------------

* The % operator with a string left argument implements formatting
similar to sprintf() in C.  The right argument is either a single
value or a tuple of values.  All features of Standard C sprintf() are
supported except %p.

* Dictionaries now support almost any key type, instead of just
strings.  (The key type must be an immutable type or must be a class
instance where the class defines a method __hash__(), in order to
avoid losing track of keys whose value may change.)

* Built-in methods are now compared properly: when comparing x.meth1
and y.meth2, if x is equal to y and the methods are defined by the
same function, x.meth1 compares equal to y.meth2.


Additions to built-in functions
-------------------------------

* str(x) returns a string version of its argument.  If the argument is
a string it is returned unchanged, otherwise it returns `x`.

* repr(x) returns the same as `x`.  (Some users found it easier to
have this as a function.)

* round(x) returns the floating point number x rounded to an whole
number, represented as a floating point number.  round(x, n) returns x
rounded to n digits.

* hasattr(x, name) returns true when x has an attribute with the given
name.

* hash(x) returns a hash code (32-bit integer) of an arbitrary
immutable object's value.

* id(x) returns a unique identifier (32-bit integer) of an arbitrary
object.

* compile() compiles a string to a Python code object.

* exec() and eval() now support execution of code objects.


Changes to the documented part of the library (standard modules)
----------------------------------------------------------------

* os.path.normpath() (a.k.a. posixpath.normpath()) has been fixed so
the border case '/foo/..' returns '/' instead of ''.

* A new function string.find() is added with similar semantics to
string.index(); however when it does not find the given substring it
returns -1 instead of raising string.index_error.


Changes to built-in modules
---------------------------

* New optional module 'array' implements operations on sequences of
integers or floating point numbers of a particular size.  This is
useful to manipulate large numerical arrays or to read and write
binary files consisting of numerical data.

* Regular expression objects created by module regex now support a new
method named group(), which returns one or more \(...\) groups by number.
The number of groups is increased from 10 to 100.

* Function compile() in module regex now supports an optional mapping
argument; a variable casefold is added to the module which can be used
as a standard uppercase to lowercase mapping.

* Module time now supports many routines that are defined in the
Standard C time interface (<time.h>): gmtime(), localtime(),
asctime(), ctime(), mktime(), as well as these variables (taken from
System V): timezone, altzone, daylight and tzname.  (The corresponding
functions in the undocumented module calendar have been removed; the
undocumented and unfinished module tzparse is now obsolete and will
disappear in a future release.)

* Module strop (the fast built-in version of standard module string)
now uses C's definition of whitespace instead of fixing it to space,
tab and newline; in practice this usually means that vertical tab,
form feed and return are now also considered whitespace.  It exports
the string of characters that are considered whitespace as well as the
characters that are considered lowercase or uppercase.

* Module sys now defines the variable builtin_module_names, a list of
names of modules built into the current interpreter (including not
yet imported, but excluding two special modules that always have to be
defined -- sys and builtin).

* Objects created by module sunaudiodev now also support flush() and
close() methods.

* Socket objects created by module socket now support an optional
flags argument for their methods sendto() and recvfrom().

* Module marshal now supports dumping to and loading from strings,
through the functions dumps() and loads().

* Module stdwin now supports some new functionality.  You may have to
ftp the latest version: ftp.cwi.nl:/pub/stdwin/stdwinforviews.tar.Z.)


Bugs fixed
----------

* Fixed comparison of negative long integers.

* The tokenizer no longer botches input lines longer than BUFSIZ.

* Fixed several severe memory leaks in module select.

* Fixed memory leaks in modules socket and sv.

* Fixed memory leak in divmod() for long integers.

* Problems with definition of floatsleep() on Suns fixed.

* Many portability bugs fixed (and undoubtedly new ones added :-).


Changes to the build procedure
------------------------------

* The Makefile supports some new targets: "make default" and "make
all".  Both are by normally equivalent to "make python".

* The Makefile no longer uses $> since it's not supported by all
versions of Make.

* The header files now all contain #ifdef constructs designed to make
it safe to include the same header file twice, as well as support for
inclusion from C++ programs (automatic extern "C" { ... } added).


Freezing Python scripts
-----------------------

* There is now some support for "freezing" a Python script as a
stand-alone executable binary file.  See the script
demo/scripts/freeze.py.  It will require some site-specific tailoring
of the script to get this working, but is quite worthwhile if you write
Python code for other who may not have built and installed Python.


MS-DOS
------

* A new MS-DOS port has been done, using MSC 6.0 (I believe).  Thanks,
Marcel van der Peijl!  This requires fewer compatibility hacks in
posixmodule.c.  The executable is not yet available but will be soon
(check the mailing list).

* The default PYTHONPATH has changed.


Changes for developers of extension modules
-------------------------------------------

* Read src/ChangeLog for full details.


SGI specific changes
--------------------

* Read src/ChangeLog for full details.


==================================
==> Release 0.9.8 (9 Jan 1993) <==
==================================

I claim no completeness here, but I've tried my best to scan the log
files throughout my source tree for interesting bits of news.  A more
complete account of the changes is to be found in the various
ChangeLog files. See also "News for release 0.9.7beta" below if you're
still using release 0.9.6, and the file HISTORY if you have an even
older release.

	--Guido


Changes to the language proper
------------------------------

There's only one big change: the conformance checking for function
argument lists (of user-defined functions only) is stricter.  Earlier,
you could get away with the following:

	(a) define a function of one argument and call it with any
	    number of arguments; if the actual argument count wasn't
	    one, the function would receive a tuple containing the
	    arguments arguments (an empty tuple if there were none).

	(b) define a function of two arguments, and call it with more
	    than two arguments; if there were more than two arguments,
	    the second argument would be passed as a tuple containing
	    the second and further actual arguments.

(Note that an argument (formal or actual) that is a tuple is counted as
one; these rules don't apply inside such tuples, only at the top level
of the argument list.)

Case (a) was needed to accommodate variable-length argument lists;
there is now an explicit "varargs" feature (precede the last argument
with a '*').  Case (b) was needed for compatibility with old class
definitions: up to release 0.9.4 a method with more than one argument
had to be declared as "def meth(self, (arg1, arg2, ...)): ...".
Version 0.9.6 provide better ways to handle both casees, bot provided
backward compatibility; version 0.9.8 retracts the compatibility hacks
since they also cause confusing behavior if a function is called with
the wrong number of arguments.

There's a script that helps converting classes that still rely on (b),
provided their methods' first argument is called "self":
demo/scripts/methfix.py.

If this change breaks lots of code you have developed locally, try
#defining COMPAT_HACKS in ceval.c.

(There's a third compatibility hack, which is the reverse of (a): if a
function is defined with two or more arguments, and called with a
single argument that is a tuple with just as many arguments, the items
of this tuple will be used as the arguments.  Although this can (and
should!) be done using the built-in function apply() instead, it isn't
withdrawn yet.)


One minor change: comparing instance methods works like expected, so
that if x is an instance of a user-defined class and has a method m,
then (x.m==x.m) yields 1.


The following was already present in 0.9.7beta, but not explicitly
mentioned in the NEWS file: user-defined classes can now define types
that behave in almost allrespects like numbers.  See
demo/classes/Rat.py for a simple example.


Changes to the build process
----------------------------

The Configure.py script and the Makefile has been made somewhat more
bullet-proof, after reports of (minor) trouble on certain platforms.

There is now a script to patch Makefile and config.c to add a new
optional built-in module: Addmodule.sh.  Read the script before using!

Useing Addmodule.sh, all optional modules can now be configured at
compile time using Configure.py, so there are no modules left that
require dynamic loading.

The Makefile has been fixed to make it easier to use with the VPATH
feature of some Make versions (e.g. SunOS).


Changes affecting portability
-----------------------------

Several minor portability problems have been solved, e.g. "malloc.h"
has been renamed to "mymalloc.h", "strdup.c" is no longer used, and
the system now tolerates malloc(0) returning 0.

For dynamic loading on the SGI, Jack Jansen's dl 1.6 is now
distributed with Python.  This solves several minor problems, in
particular scripts invoked using #! can now use dynamic loading.


Changes to the interpreter interface
------------------------------------

On popular demand, there's finally a "profile" feature for interactive
use of the interpreter.  If the environment variable $PYTHONSTARTUP is
set to the name of an existing file, Python statements in this file
are executed when the interpreter is started in interactive mode.

There is a new clean-up mechanism, complementing try...finally: if you
assign a function object to sys.exitfunc, it will be called when
Python exits or receives a SIGTERM or SIGHUP signal.

The interpreter is now generally assumed to live in
/usr/local/bin/python (as opposed to /usr/local/python).  The script
demo/scripts/fixps.py will update old scripts in place (you can easily
modify it to do other similar changes).

Most I/O that uses sys.stdin/stdout/stderr will now use any object
assigned to those names as long as the object supports readline() or
write() methods.

The parser stack has been increased to 500 to accommodate more
complicated expressions (7 levels used to be the practical maximum,
it's now about 38).

The limit on the size of the *run-time* stack has completely been
removed -- this means that tuple or list displays can contain any
number of elements (formerly more than 50 would crash the
interpreter). 


Changes to existing built-in functions and methods
--------------------------------------------------

The built-in functions int(), long(), float(), oct() and hex() now
also apply to class instalces that define corresponding methods
(__int__ etc.).


New built-in functions
----------------------

The new functions str() and repr() convert any object to a string.
The function repr(x) is in all respects equivalent to `x` -- some
people prefer a function for this.  The function str(x) does the same
except if x is already a string -- then it returns x unchanged
(repr(x) adds quotes and escapes "funny" characters as octal escapes).

The new function cmp(x, y) returns -1 if x<y, 0 if x==y, 1 if x>y.


Changes to general built-in modules
-----------------------------------

The time module's functions are more general: time() returns a
floating point number and sleep() accepts one.  Their accuracies
depends on the precision of the system clock.  Millisleep is no longer
needed (although it still exists for now), but millitimer is still
needed since on some systems wall clock time is only available with
seconds precision, while a source of more precise time exists that
isn't synchronized with the wall clock.  (On UNIX systems that support
the BSD gettimeofday() function, time.time() is as time.millitimer().)

The string representation of a file object now includes an address:
'<file 'filename', mode 'r' at #######>' where ###### is a hex number
(the object's address) to make it unique.

New functions added to posix: nice(), setpgrp(), and if your system
supports them: setsid(), setpgid(), tcgetpgrp(), tcsetpgrp().

Improvements to the socket module: socket objects have new methods
getpeername() and getsockname(), and the {get,set}sockopt methods can
now get/set any kind of option using strings built with the new struct
module.  And there's a new function fromfd() which creates a socket
object given a file descriptor (useful for servers started by inetd,
which have a socket connected to stdin and stdout).


Changes to SGI-specific built-in modules
----------------------------------------

The FORMS library interface (fl) now requires FORMS 2.1a.  Some new
functions have been added and some bugs have been fixed.

Additions to al (audio library interface): added getname(),
getdefault() and getminmax().

The gl modules doesn't call "foreground()" when initialized (this
caused some problems) like it dit in 0.9.7beta (but not before).
There's a new gl function 'gversion() which returns a version string.

The interface to sv (Indigo video interface) has totally changed.
(Sorry, still no documentation, but see the examples in
demo/sgi/{sv,video}.)


Changes to standard library modules
-----------------------------------

Most functions in module string are now much faster: they're actually
implemented in C.  The module containing the C versions is called
"strop" but you should still import "string" since strop doesn't
provide all the interfaces defined in string (and strop may be renamed
to string when it is complete in a future release).

string.index() now accepts an optional third argument giving an index
where to start searching in the first argument, so you can find second
and further occurrences (this is similar to the regular expression
functions in regex).

The definition of what string.splitfields(anything, '') should return
is changed for the last time: it returns a singleton list containing
its whole first argument unchanged.  This is compatible with
regsub.split() which also ignores empty delimiter matches.

posixpath, macpath: added dirname() and normpath() (and basename() to
macpath).

The mainloop module (for use with stdwin) can now demultiplex input
from other sources, as long as they can be polled with select().


New built-in modules
--------------------

Module struct defines functions to pack/unpack values to/from strings
representing binary values in native byte order.

Module strop implements C versions of many functions from string (see
above).

Optional module fcntl defines interfaces to fcntl() and ioctl() --
UNIX only.  (Not yet properly documented -- see however src/fcntl.doc.)

Optional module mpz defines an interface to an altaernative long
integer implementation, the GNU MPZ library.

Optional module md5 uses the GNU MPZ library to calculate MD5
signatures of strings.

There are also optional new modules specific to SGI machines: imageop
defines some simple operations to images represented as strings; sv
interfaces to the Indigo video board; cl interfaces to the (yet
unreleased) compression library.


New standard library modules
----------------------------

(Unfortunately the following modules are not all documented; read the
sources to find out more about them!)

autotest: run testall without showing any output unless it differs
from the expected output

bisect: use bisection to insert or find an item in a sorted list

colorsys: defines conversions between various color systems (e.g. RGB
<-> YUV)

nntplib: a client interface to NNTP servers

pipes: utility to construct pipeline from templates, e.g. for
conversion from one file format to another using several utilities.

regsub: contains three functions that are more or less compatible with
awk functions of the same name: sub() and gsub() do string
substitution, split() splits a string using a regular expression to
define how separators are define.

test_types: test operations on the built-in types of Python

toaiff: convert various audio file formats to AIFF format

tzparse: parse the TZ environment parameter (this may be less general
than it could be, let me know if you fix it).

(Note that the obsolete module "path" no longer exists.)


New SGI-specific library modules
--------------------------------

CL: constants for use with the built-in compression library interface (cl)

Queue: a multi-producer, multi-consumer queue class implemented for
use with the built-in thread module

SOCKET: constants for use with built-in module socket, e.g. to set/get
socket options.  This is SGI-specific because the constants to be
passed are system-dependent.  You can generate a version for your own
system by running the script demo/scripts/h2py.py with
/usr/include/sys/socket.h as input.

cddb: interface to the database used the the CD player

torgb: convert various image file types to rgb format (requires pbmplus)


New demos
---------

There's an experimental interface to define Sun RPC clients and
servers in demo/rpc.

There's a collection of interfaces to WWW, WAIS and Gopher (both
Python classes and program providing a user interface) in demo/www.
This includes a program texi2html.py which converts texinfo files to
HTML files (the format used hy WWW).

The ibrowse demo has moved from demo/stdwin/ibrowse to demo/ibrowse.

For SGI systems, there's a whole collection of programs and classes
that make use of the Indigo video board in demo/sgi/{sv,video}.  This
represents a significant amount of work that we're giving away!

There are demos "rsa" and "md5test" that exercise the mpz and md5
modules, respectively.  The rsa demo is a complete implementation of
the RSA public-key cryptosystem!

A bunch of games and examples submitted by Stoffel Erasmus have been
included in demo/stoffel.

There are miscellaneous new files in some existing demo
subdirectories: classes/bitvec.py, scripts/{fixps,methfix}.py,
sgi/al/cmpaf.py, sockets/{mcast,gopher}.py.

There are also many minor changes to existing files, but I'm too lazy
to run a diff and note the differences -- you can do this yourself if
you save the old distribution's demos.  One highlight: the
stdwin/python.py demo is much improved!


Changes to the documentation
----------------------------

The LaTeX source for the library uses different macros to enable it to
be converted to texinfo, and from there to INFO or HTML format so it
can be browsed as a hypertext.  The net result is that you can now
read the Python library documentation in Emacs info mode!


Changes to the source code that affect C extension writers
----------------------------------------------------------

The function strdup() no longer exists (it was used only in one places
and is somewhat of a a portability problem sice some systems have the
same function in their C library.

The functions NEW() and RENEW() allocate one spare byte to guard
against a NULL return from malloc(0) being taken for an error, but
this should not be relied upon.


=========================
==> Release 0.9.7beta <==
=========================


Changes to the language proper
------------------------------

User-defined classes can now implement operations invoked through
special syntax, such as x[i] or `x` by defining methods named
__getitem__(self, i) or __repr__(self), etc.


Changes to the build process
----------------------------

Instead of extensive manual editing of the Makefile to select
compile-time options, you can now run a Configure.py script.
The Makefile as distributed builds a minimal interpreter sufficient to
run Configure.py.  See also misc/BUILD

The Makefile now includes more "utility" targets, e.g. install and
tags/TAGS

Using the provided strtod.c and strtol.c are now separate options, as
on the Sun the provided strtod.c dumps core :-(

The regex module is now an option chosen by the Makefile, since some
(old) C compilers choke on regexpr.c


Changes affecting portability
-----------------------------

You need STDWIN version 0.9.7 (released 30 June 1992) for the stdwin
interface

Dynamic loading is now supported for Sun (and other non-COFF systems)
throug dld-3.2.3, as well as for SGI (a new version of Jack Jansen's
DL is out, 1.4)

The system-dependent code for the use of the select() system call is
moved to one file: myselect.h

Thanks to Jaap Vermeulen, the code should now port cleanly to the
SEQUENT


Changes to the interpreter interface
------------------------------------

The interpretation of $PYTHONPATH in the environment is different: it
is inserted in front of the default path instead of overriding it


Changes to existing built-in functions and methods
--------------------------------------------------

List objects now support an optional argument to their sort() method,
which is a comparison function similar to qsort(3) in C

File objects now have a method fileno(), used by the new select module
(see below)


New built-in function
---------------------

coerce(x, y): take two numbers and return a tuple containing them
both converted to a common type


Changes to built-in modules
---------------------------

sys: fixed core dumps in settrace() and setprofile()

socket: added socket methods setsockopt() and getsockopt(); and
fileno(), used by the new select module (see below)

stdwin: added fileno() == connectionnumber(), in support of new module
select (see below)

posix: added get{eg,eu,g,u}id(); waitpid() is now a separate function.

gl: added qgetfd()

fl: added several new functions, fixed several obscure bugs, adapted
to FORMS 2.1


Changes to standard modules
---------------------------

posixpath: changed implementation of ismount()

string: atoi() no longer mistakes leading zero for octal number

...


New built-in modules
--------------------

Modules marked "dynamic only" are not configured at compile time but
can be loaded dynamically.  You need to turn on the DL or DLD option in
the Makefile for support dynamic loading of modules (this requires
external code).

select: interfaces to the BSD select() system call

dbm: interfaces to the (new) dbm library (dynamic only)

nis: interfaces to some NIS functions (aka yellow pages)

thread: limited form of multiple threads (sgi only)

audioop: operations useful for audio programs, e.g. u-LAW and ADPCM
coding (dynamic only)

cd: interface to Indigo SCSI CDROM player audio library (sgi only)

jpeg: read files in JPEG format (dynamic only, sgi only; needs
external code)

imgfile: read SGI image files (dynamic only, sgi only)

sunaudiodev: interface to sun's /dev/audio (dynamic only, sun only)

sv: interface to Indigo video library (sgi only)

pc: a minimal set of MS-DOS interfaces (MS-DOS only)

rotor: encryption, by Lance Ellinghouse (dynamic only)


New standard modules
--------------------

Not all these modules are documented.  Read the source:
lib/<modulename>.py.  Sometimes a file lib/<modulename>.doc contains
additional documentation.

imghdr: recognizes image file headers

sndhdr: recognizes sound file headers

profile: print run-time statistics of Python code

readcd, cdplayer: companion modules for built-in module cd (sgi only)

emacs: interface to Emacs using py-connect.el (see below).

SOCKET: symbolic constant definitions for socket options

SUNAUDIODEV: symbolic constant definitions for sunaudiodef (sun only)

SV: symbolic constat definitions for sv (sgi only)

CD: symbolic constat definitions for cd (sgi only)


New demos
---------

scripts/pp.py: execute Python as a filter with a Perl-like command
line interface

classes/: examples using the new class features

threads/: examples using the new thread module

sgi/cd/: examples using the new cd module


Changes to the documentation
----------------------------

The last-minute syntax changes of release 0.9.6 are now reflected
everywhere in the manuals

The reference manual has a new section (3.2) on implementing new kinds
of numbers, sequences or mappings with user classes

Classes are now treated extensively in the tutorial (chapter 9)

Slightly restructured the system-dependent chapters of the library
manual

The file misc/EXTENDING incorporates documentation for mkvalue() and
a new section on error handling

The files misc/CLASSES and misc/ERRORS are no longer necessary

The doc/Makefile now creates PostScript files automatically


Miscellaneous changes
---------------------

Incorporated Tim Peters' changes to python-mode.el, it's now version
1.06

A python/Emacs bridge (provided by Terrence M. Brannon) lets a Python
program running in an Emacs buffer execute Emacs lisp code.  The
necessary Python code is in lib/emacs.py.  The Emacs code is
misc/py-connect.el (it needs some external Emacs lisp code)


Changes to the source code that affect C extension writers
----------------------------------------------------------

New service function mkvalue() to construct a Python object from C
values according to a "format" string a la getargs()

Most functions from pythonmain.c moved to new pythonrun.c which is
in libpython.a.  This should make embedded versions of Python easier

ceval.h is split in eval.h (which needs compile.h and only declares
eval_code) and ceval.h (which doesn't need compile.hand declares the
rest)

ceval.h defines macros BGN_SAVE / END_SAVE for use with threads (to
improve the parallellism of multi-threaded programs by letting other
Python code run when a blocking system call or something similar is
made)

In structmember.[ch], new member types BYTE, CHAR and unsigned
variants have been added

New file xxmodule.c is a template for new extension modules.


==================================
==> RELEASE 0.9.6 (6 Apr 1992) <==
==================================

Misc news in 0.9.6:
- Restructured the misc subdirectory
- Reference manual completed, library manual much extended (with indexes!)
- the GNU Readline library is now distributed standard with Python
- the script "../demo/scripts/classfix.py" fixes Python modules using old
  class syntax
- Emacs python-mode.el (was python.el) vastly improved (thanks, Tim!)
- Because of the GNU copyleft business I am not using the GNU regular
  expression implementation but a free re-implementation by Tatu Ylonen
  that recently appeared in comp.sources.misc (Bravo, Tatu!)

New features in 0.9.6:
- stricter try stmt syntax: cannot mix except and finally clauses on 1 try
- New module 'os' supplants modules 'mac' and 'posix' for most cases;
  module 'path' is replaced by 'os.path'
- os.path.split() return value differs from that of old path.split()
- sys.exc_type, sys.exc_value, sys.exc_traceback are set to the exception
  currently being handled
- sys.last_type, sys.last_value, sys.last_traceback remember last unhandled
  exception
- New function string.expandtabs() expands tabs in a string
- Added times() interface to posix (user & sys time of process & children)
- Added uname() interface to posix (returns OS type, hostname, etc.)
- New built-in function execfile() is like exec() but from a file
- Functions exec() and eval() are less picky about whitespace/newlines
- New built-in functions getattr() and setattr() access arbitrary attributes
- More generic argument handling in built-in functions (see "./EXTENDING")
- Dynamic loading of modules written in C or C++ (see "./DYNLOAD")
- Division and modulo for long and plain integers with negative operands
  have changed; a/b is now floor(float(a)/float(b)) and a%b is defined
  as a-(a/b)*b.  So now the outcome of divmod(a,b) is the same as
  (a/b, a%b) for integers.  For floats, % is also changed, but of course
  / is unchanged, and divmod(x,y) does not yield (x/y, x%y)...
- A function with explicit variable-length argument list can be declared
  like this: def f(*args): ...; or even like this: def f(a, b, *rest): ...
- Code tracing and profiling features have been added, and two source
  code debuggers are provided in the library (pdb.py, tty-oriented,
  and wdb, window-oriented); you can now step through Python programs!
  See sys.settrace() and sys.setprofile(), and "../lib/pdb.doc"
- '==' is now the only equality operator; "../demo/scripts/eqfix.py" is
  a script that fixes old Python modules
- Plain integer right shift now uses sign extension
- Long integer shift/mask operations now simulate 2's complement
  to give more useful results for negative operands
- Changed/added range checks for long/plain integer shifts
- Options found after "-c command" are now passed to the command in sys.argv
  (note subtle incompatiblity with "python -c command -- -options"!)
- Module stdwin is better protected against touching objects after they've
  been closed; menus can now also be closed explicitly
- Stdwin now uses its own exception (stdwin.error)

New features in 0.9.5 (released as Macintosh application only, 2 Jan 1992):
- dictionary objects can now be compared properly; e.g., {}=={} is true
- new exception SystemExit causes termination if not caught;
  it is raised by sys.exit() so that 'finally' clauses can clean up,
  and it may even be caught.  It does work interactively!
- new module "regex" implements GNU Emacs style regular expressions;
  module "regexp" is rewritten in Python for backward compatibility
- formal parameter lists may contain trailing commas

Bugs fixed in 0.9.6:
- assigning to or deleting a list item with a negative index dumped core
- divmod(-10L,5L) returned (-3L, 5L) instead of (-2L, 0L)

Bugs fixed in 0.9.5:
- masking operations involving negative long integers gave wrong results


===================================
==> RELEASE 0.9.4 (24 Dec 1991) <==
===================================

- new function argument handling (see below)
- built-in apply(func, args) means func(args[0], args[1], ...)
- new, more refined exceptions
- new exception string values (NameError = 'NameError' etc.)
- better checking for math exceptions
- for sequences (string/tuple/list), x[-i] is now equivalent to x[len(x)-i]
- fixed list assignment bug: "a[1:1] = a" now works correctly
- new class syntax, without extraneous parentheses
- new 'global' statement to assign global variables from within a function


New class syntax
----------------

You can now declare a base class as follows:

	class B:			# Was: class B():
		def some_method(self): ...
		...

and a derived class thusly:

	class D(B):			# Was: class D() = B():
		def another_method(self, arg): ...

Multiple inheritance looks like this:

	class M(B, D):			# Was: class M() = B(), D():
		def this_or_that_method(self, arg): ...

The old syntax is still accepted by Python 0.9.4, but will disappear
in Python 1.0 (to be posted to comp.sources).


New 'global' statement
----------------------

Every now and then you have a global variable in a module that you
want to change from within a function in that module -- say, a count
of calls to a function, or an option flag, etc.  Until now this was
not directly possible.  While several kludges are known that
circumvent the problem, and often the need for a global variable can
be avoided by rewriting the module as a class, this does not always
lead to clearer code.

The 'global' statement solves this dilemma.  Its occurrence in a
function body means that, for the duration of that function, the
names listed there refer to global variables.  For instance:

	total = 0.0
	count = 0

	def add_to_total(amount):
		global total, count
		total = total + amount
		count = count + 1

'global' must be repeated in each function where it is needed.  The
names listed in a 'global' statement must not be used in the function
before the statement is reached.

Remember that you don't need to use 'global' if you only want to *use*
a global variable in a function; nor do you need ot for assignments to
parts of global variables (e.g., list or dictionary items or
attributes of class instances).  This has not changed; in fact
assignment to part of a global variable was the standard workaround.


New exceptions
--------------

Several new exceptions have been defined, to distinguish more clearly
between different types of errors.

name			meaning					was

AttributeError		reference to non-existing attribute	NameError
IOError			unexpected I/O error			RuntimeError
ImportError		import of non-existing module or name	NameError
IndexError		invalid string, tuple or list index	RuntimeError
KeyError		key not in dictionary			RuntimeError
OverflowError		numeric overflow			RuntimeError
SyntaxError		invalid syntax				RuntimeError
ValueError		invalid argument value			RuntimeError
ZeroDivisionError	division by zero			RuntimeError

The string value of each exception is now its name -- this makes it
easier to experimentally find out which operations raise which
exceptions; e.g.:

	>>> KeyboardInterrupt
	'KeyboardInterrupt'
	>>>


New argument passing semantics
------------------------------

Off-line discussions with Steve Majewski and Daniel LaLiberte have
convinced me that Python's parameter mechanism could be changed in a
way that made both of them happy (I hope), kept me happy, fixed a
number of outstanding problems, and, given some backward compatibility
provisions, would only break a very small amount of existing code --
probably all mine anyway.  In fact I suspect that most Python users
will hardly notice the difference.  And yet it has cost me at least
one sleepless night to decide to make the change...

Philosophically, the change is quite radical (to me, anyway): a
function is no longer called with either zero or one argument, which
is a tuple if there appear to be more arguments.  Every function now
has an argument list containing 0, 1 or more arguments.  This list is
always implemented as a tuple, and it is a (run-time) error if a
function is called with a different number of arguments than expected.

What's the difference? you may ask.  The answer is, very little unless
you want to write variadic functions -- functions that may be called
with a variable number of arguments.  Formerly, you could write a
function that accepted one or more arguments with little trouble, but
writing a function that could be called with either 0 or 1 argument
(or more) was next to impossible.  This is now a piece of cake: you
can simply declare an argument that receives the entire argument
tuple, and check its length -- it will be of size 0 if there are no
arguments.

Another anomaly of the old system was the way multi-argument methods
(in classes) had to be declared, e.g.:

	class Point():
		def init(self, (x, y, color)): ...
		def setcolor(self, color): ...
		dev moveto(self, (x, y)): ...
		def draw(self): ...

Using the new scheme there is no need to enclose the method arguments
in an extra set of parentheses, so the above class could become:

	class Point:
		def init(self, x, y, color): ...
		def setcolor(self, color): ...
		dev moveto(self, x, y): ...
		def draw(self): ...

That is, the equivalence rule between methods and functions has
changed so that now p.moveto(x,y) is equivalent to Point.moveto(p,x,y)
while formerly it was equivalent to Point.moveto(p,(x,y)).

A special backward compatibility rule makes that the old version also
still works: whenever a function with exactly two arguments (at the top
level) is called with more than two arguments, the second and further
arguments are packed into a tuple and passed as the second argument.
This rule is invoked independently of whether the function is actually a
method, so there is a slight chance that some erroneous calls of
functions expecting two arguments with more than that number of
arguments go undetected at first -- when the function tries to use the
second argument it may find it is a tuple instead of what was expected.
Note that this rule will be removed from future versions of the
language; it is a backward compatibility provision *only*.

Two other rules and a new built-in function handle conversion between
tuples and argument lists:

Rule (a): when a function with more than one argument is called with a
single argument that is a tuple of the right size, the tuple's items
are used as arguments.

Rule (b): when a function with exactly one argument receives no
arguments or more than one, that one argument will receive a tuple
containing the arguments (the tuple will be empty if there were no
arguments).


A new built-in function, apply(), was added to support functions that
need to call other functions with a constructed argument list.  The call

	apply(function, tuple)

is equivalent to

	function(tuple[0], tuple[1], ..., tuple[len(tuple)-1])


While no new argument syntax was added in this phase, it would now be
quite sensible to add explicit syntax to Python for default argument
values (as in C++ or Modula-3), or a "rest" argument to receive the
remaining arguments of a variable-length argument list.


========================================================
==> Release 0.9.3 (never made available outside CWI) <==
========================================================

- string sys.version shows current version (also printed on interactive entry)
- more detailed exceptions, e.g., IOError, ZeroDivisionError, etc.
- 'global' statement to declare module-global variables assigned in functions.
- new class declaration syntax: class C(Base1, Base2, ...): suite
  (the old syntax is still accepted -- be sure to convert your classes now!)
- C shifting and masking operators: << >> ~ & ^ | (for ints and longs).
- C comparison operators: == != (the old = and <> remain valid).
- floating point numbers may now start with a period (e.g., .14).
- definition of integer division tightened (always truncates towards zero).
- new builtins hex(x), oct(x) return hex/octal string from (long) integer.
- new list method l.count(x) returns the number of occurrences of x in l.
- new SGI module: al (Indigo and 4D/35 audio library).
- the FORMS interface (modules fl and FL) now uses FORMS 2.0
- module gl: added lrect{read,write}, rectzoom and pixmode;
  added (non-GL) functions (un)packrect.
- new socket method: s.allowbroadcast(flag).
- many objects support __dict__, __methods__ or __members__.
- dir() lists anything that has __dict__.
- class attributes are no longer read-only.
- classes support __bases__, instances support __class__ (and __dict__).
- divmod() now also works for floats.
- fixed obscure bug in eval('1            ').


===================================
==> Release 0.9.2 (Autumn 1991) <==
===================================

Highlights
----------

- tutorial now (almost) complete; library reference reorganized
- new syntax: continue statement; semicolons; dictionary constructors;
  restrictions on blank lines in source files removed
- dramatically improved module load time through precompiled modules
- arbitrary precision integers: compute 2 to the power 1000 and more...
- arithmetic operators now accept mixed type operands, e.g., 3.14/4
- more operations on list: remove, index, reverse; repetition
- improved/new file operations: readlines, seek, tell, flush, ...
- process management added to the posix module: fork/exec/wait/kill etc.
- BSD socket operations (with example servers and clients!)
- many new STDWIN features (color, fonts, polygons, ...)
- new SGI modules: font manager and FORMS library interface


Extended list of changes in 0.9.2
---------------------------------

Here is a summary of the most important user-visible changes in 0.9.2,
in somewhat arbitrary order.  Changes in later versions are listed in
the "highlights" section above.


1. Changes to the interpreter proper

- Simple statements can now be separated by semicolons.
  If you write "if t: s1; s2", both s1 and s2 are executed
  conditionally.
- The 'continue' statement was added, with semantics as in C.
- Dictionary displays are now allowed on input: {key: value, ...}.
- Blank lines and lines bearing only a comment no longer need to
  be indented properly.  (A completely empty line still ends a multi-
  line statement interactively.)
- Mixed arithmetic is supported, 1 compares equal to 1.0, etc.
- Option "-c command" to execute statements from the command line
- Compiled versions of modules are cached in ".pyc" files, giving a
  dramatic improvement of start-up time
- Other, smaller speed improvements, e.g., extracting characters from
  strings, looking up single-character keys, and looking up global
  variables
- Interrupting a print operation raises KeyboardInterrupt instead of
  only cancelling the print operation
- Fixed various portability problems (it now passes gcc with only
  warnings -- more Standard C compatibility will be provided in later
  versions)
- Source is prepared for porting to MS-DOS
- Numeric constants are now checked for overflow (this requires
  standard-conforming strtol() and strtod() functions; a correct
  strtol() implementation is provided, but the strtod() provided
  relies on atof() for everything, including error checking


2. Changes to the built-in types, functions and modules

- New module socket: interface to BSD socket primitives
- New modules pwd and grp: access the UNIX password and group databases
- (SGI only:) New module "fm" interfaces to the SGI IRIX Font Manager
- (SGI only:) New module "fl" interfaces to Mark Overmars' FORMS library
- New numeric type: long integer, for unlimited precision
	- integer constants suffixed with 'L' or 'l' are long integers
	- new built-in function long(x) converts int or float to long
	- int() and float() now also convert from long integers
- New built-in function:
	- pow(x, y) returns x to the power y
- New operation and methods for lists:
	- l*n returns a new list consisting of n concatenated copies of l
	- l.remove(x) removes the first occurrence of the value x from l
	- l.index(x) returns the index of the first occurrence of x in l
	- l.reverse() reverses l in place
- New operation for tuples:
	- t*n returns a tuple consisting of n concatenated copies of t
- Improved file handling:
	- f.readline() no longer restricts the line length, is faster,
	  and isn't confused by null bytes; same for raw_input()
	- f.read() without arguments reads the entire (rest of the) file
	- mixing of print and sys.stdout.write() has different effect
- New methods for files:
	- f.readlines() returns a list containing the lines of the file,
	  as read with f.readline()
	- f.flush(), f.tell(), f.seek() call their stdio counterparts
	- f.isatty() tests for "tty-ness"
- New posix functions:
	- _exit(), exec(), fork(), getpid(), getppid(), kill(), wait()
	- popen() returns a file object connected to a pipe
	- utime() replaces utimes() (the latter is not a POSIX name)
- New stdwin features, including:
	- font handling
	- color drawing
	- scroll bars made optional
	- polygons
	- filled and xor shapes
	- text editing objects now have a 'settext' method


3. Changes to the standard library

- Name change: the functions path.cat and macpath.cat are now called
  path.join and macpath.join
- Added new modules: formatter, mutex, persist, sched, mainloop
- Added some modules and functionality to the "widget set" (which is
  still under development, so please bear with me):
	DirList, FormSplit, TextEdit, WindowSched
- Fixed module testall to work non-interactively
- Module string:
	- added functions join() and joinfields()
	- fixed center() to work correct and make it "transitive"
- Obsolete modules were removed: util, minmax
- Some modules were moved to the demo directory


4. Changes to the demonstration programs

- Added new useful scipts: byteyears, eptags, fact, from, lfact,
  objgraph, pdeps, pi, primes, ptags, which
- Added a bunch of socket demos
- Doubled the speed of ptags
- Added new stdwin demos: microedit, miniedit
- Added a windowing interface to the Python interpreter: python (most
  useful on the Mac)
- Added a browser for Emacs info files: demo/stdwin/ibrowse
  (yes, I plan to put all STDWIN and Python documentation in texinfo
  form in the future)


5. Other changes to the distribution

- An Emacs Lisp file "python.el" is provided to facilitate editing
  Python programs in GNU Emacs (slightly improved since posted to
  gnu.emacs.sources)
- Some info on writing an extension in C is provided
- Some info on building Python on non-UNIX platforms is provided


=====================================
==> Release 0.9.1 (February 1991) <==
=====================================

- Micro changes only
- Added file "patchlevel.h"


=====================================
==> Release 0.9.0 (February 1991) <==
=====================================

Original posting to alt.sources.

--J/dobhs11T7y2rNN--



From LCortes at Flash.Net  Mon Jan 17 20:36:42 2000
From: LCortes at Flash.Net (Luis Cortes)
Date: Tue, 18 Jan 2000 01:36:42 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com> <slrn886oku.2ac.grant@grante.comtrol.com> <3883681A.DBD1B3C1@roguewave.com> <20000117131945.A2415@visi.com>
Message-ID: <20000118.1335900@negrita.brunner.org>

Actually, Mickysoft Exchange does allow pop3 connections, you have to 
get your system admin to turn it on for you though.  Also, fetchmail 
will move your ME mail down to your local machine if you ask real 
nice.

Regards,
Luis.

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 1/17/00, 12:19:46 PM, grante at visi.com wrote regarding Re: Sucking 
e-mail from Exchange Server vis HTTP?:


> On Mon, Jan 17, 2000 at 12:06:03PM -0700, bjorn wrote:

> > > >> I just found the http server page on Exchange server that
> > > >> allows access to e-mails via a web browser.  What I'd really
> > > >> like is to write a Python program that just sucks the mail out
> > > >> of the inbox and delivers it locally.
> > >
> > > My other option would be to write a MAPI client that runs under
> > > Win32 and then gives the mail to the SMTP server on my Linux
> > > machine.  There's a MAPI module for Python, isn't there?
> > > Perhaps that would be easier -- but I've never programmed under
> > > Win32 before...
> >
> > If you do this, let us know (I know I would be interested ;-)

> I'd rather get mail from the Exchange server directly, but
> apparently the protocol used between Exchange Server and
> Outlook isn't documented anywhere (it's rumored to be some form
> of LDAP-like thing).  I might try getting mail via the Web
> interface first.  I can play with that without having to
> install Windows. ;)

> --
> Grant Edwards
> grante at visi.com





From kc5tja at garnet.armored.net  Tue Jan 11 18:14:04 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 11 Jan 2000 23:14:04 GMT
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <1264508924-1480880@hypernet.com> <85g09q$o7e$1@nntp3.atl.mindspring.net>
Message-ID: <slrn87nemb.qla.kc5tja@garnet.armored.net>

In article <85g09q$o7e$1 at nntp3.atl.mindspring.net>, Aahz Maruch wrote:
>Hmmm...  So it's kind of like sparse overloading?  ;-)

An excellent description!  Consider it stolen.  :)

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From maccgi at bellsouth.net  Mon Jan 31 00:13:25 2000
From: maccgi at bellsouth.net (Richard Gordon)
Date: Mon, 31 Jan 2000 00:13:25 -0500
Subject: thread_test crashes
Message-ID: <v04205504b4bac9bbb35a@[192.168.0.2]>

I'm having a problem with threads using Python 1.5.2 and mkLinux 
(DR3, kernel upgraded to GENERIC #7). The short version is that I 
initially got eveything configured, built & installed without any 
trouble, then decided to try installing Zope which was unhappy 
because I didn't include threads when I configured Python.

So I start over and this time use the --with-thread switch. Again, no 
problems with the installation, but when I run regrtest.py, 
test_thread crashes:
(HandlerBCalled : (2, <frame object at 19675e0>).

I haven't gotten around to investigating this any further and was 
wondering if anyone has any thoughts about what's going on here and 
how to fix it?

Thanks.

Richard Gordon
--------------------
Gordon Consulting & Design
Database Design/Scripting Languages
mailto:richard at richardgordon.net
http://www.richardgordon.net
770.971.6887 (voice)
770.216.1829 (fax)



From robin at jessikat.demon.co.uk  Tue Jan 25 09:09:06 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 25 Jan 2000 14:09:06 +0000
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com>
 <$WXiIAAO+Dj4Ew2W@jessikat.demon.co.uk> <388C7D4A.D873AA@tismer.com>
 <zMBtEAA$jIj4Ewm7@jessikat.demon.co.uk> <388C9733.B4F60DC0@tismer.com>
Message-ID: <7PUz+AAC6aj4Ewne@jessikat.demon.co.uk>

In article <388C9733.B4F60DC0 at tismer.com>, Christian Tismer
<tismer at tismer.com> writes
>
>
>Robin Becker wrote:
>> 
>> In message <388C7D4A.D873AA at tismer.com>, Christian Tismer
>> <tismer at tismer.com> writes
>
.....

>Shouldn't be so. I do not change the object model
>at all. It appears that I missed to give another
>reference to an argument.
>SLP is designed to work exactly as standard Python,
>if the old API calls are used. I have to hunt it later.
>
>ciao - chris
>


OK I've reproduced the Zope bug. It seems that stackless doen't quite
hack try finally properly

In stackless this produces None in standard it produces 1

def test():
        try:
                z=1
                return z
        finally:
                pass

print test()

-- 
Robin Becker


From news at dorb.com  Wed Jan  5 23:54:33 2000
From: news at dorb.com (Darrell)
Date: Wed, 5 Jan 2000 23:54:33 -0500
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net> <N6Vc4.610$hJ1.7143@news.rdc1.tx.home.com>
Message-ID: <JbVc4.279$mj7.6038@typhoon.nyroc.rr.com>

http://www.webaxxs.net/
They have Python 1.4 installed but I installed 1.5 into my account.
Since they give me 500meg of storage that's not a problem.

--
--Darrell
"Paul E. Bible" <pbible at home.com> wrote in message
news:N6Vc4.610$hJ1.7143 at news.rdc1.tx.home.com...
> Interland.com supports Python on a Unix server (but not on Windows/NT).
>
> Paul
>
> "Chuck Esterbrook" <echuck at usit.net> wrote in message
> news:3873EE03.1CE5F8CC at usit.net...
> > Seems like many web host providers, such as MindSpring, offer CGI with
> > Perl, but not other languages, notably Python.
> >
> > Does anyone know who would host a CGI Python site?
> >
> > -Chuck
> >
> >
>
>
> --
> http://www.python.org/mailman/listinfo/python-list




From tjg at avalongroup.net  Mon Jan 17 15:11:08 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Mon, 17 Jan 2000 12:11:08 -0800
Subject: Tkinter Tab orders and keybindings
Message-ID: <3883775C.9369120@exceptionalminds.com>

Hi all,

How does one control the tab order of Tkwidgets? Is it even possible?

also, if I bind the Tab key to a widget, how do I get it to perform its
original function as well as my function?

Thanks.


-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From lvirden at cas.org  Sat Jan 22 15:18:17 2000
From: lvirden at cas.org (Larry W. Virden)
Date: Sat, 22 Jan 2000 15:18:17 -0500 (EST)
Subject: Looking for Solaris 2.6/Python 1.5.2 make test fix
In-Reply-To: Your message of Fri, 21 Jan 2000 23:50:34 +0100
Message-ID: <0001221518.AA16327@cas.org>

From: Thomas Wouters <thomas at xs4all.net>
> Something probably went wrong with building your Python binary. Can you
> start the python executable ? do you get the prompt ? Can you execute code ?

Yes, I can start the executable and I get the prompt.  I don't know about
the execute code - as far as I can tell.  Only the one Lib test fails -
I assume there's more python code than that that executes during make test.

> computer, or you built a buggy module statically. How were you building it ?
> Which version of python was it, how did you run 'configure' and what did you
> change to Modules/Setup ? (You *do* need to change a few things in there.)

I don't know what "how was I building it" means.  If you mean "what
arguments to configure did I provide", it was:
$ /ldatae/Misc/python/Python-1.5.2/configure  --program-prefix=g --target=sparc-sun-solaris2.6 --prefix=/projects/gnu/sparc-sun-solaris2.6 --exec-prefix=/projects/gnu/sparc-sun-solaris2.6 --with-gnu-gettext --with-rx --with-curses=/projects/gnu/sparc-sun-solaris2.6/lib --with-shared --with-profile --enable-nls --x-includes=/usr/openwin/include --x-libraries=/usr/openwin/lib --with-x-toolkit=motif --with-x --with-threads


As for 'changes to Modules/Setup', I modified the Modules/Setup.local - do
you really want me to send that whole thing to the list?  It's 176 lines!

> > lvirden at cas.org wrote:
> > > 
> > > When I configure then do a make all and then make test, Python core dumps:
> > > 
> > > rm -f /ldatae/Misc/python/Python-1.5.2/Lib/test/*.py[co]
> > > PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py
> > > Bus Error - core dumped
> > > make: [test] Error 138 (ignored)
> > > PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py
> > > Bus Error - core dumped
> > > make: *** [test] Error 138
> 
> The same goes for solaris, though i cannot be 100% certain that it works
> there -- haven't installed Python on the one solaris box we have (but
> that'll change if we start using it) but it would strike me as preposterous
> if it should fail to work ;) The same goes here, however. Did you build any
> things statically ? What happens if you build them dynamically ? Can you
> possibly then start python, but not import that specific module ? 

I typed make distclean; configure; make all.  If anything 'made statically'
it would be something that configure wanted built that way.  Same for the
info about 'make dynamically' - I did nothing to the Makefiles - so however
they were built, that's what got compiled.  As for 'not importing that
specific module' - I'm a novice to Python, so I don't know what 'module'
that test is invoking.  Seems to be related to regular expressions (perhaps
GNU's rx library?)

> Other helpful hints towards the problem: Did you get any warnings during
> compilation ? (Especially the optional modules -- you may have given the
> wrong library directory for linking, for instance, or linking with the wrong
> library version, compared to the include files.) If either of you can't
> figure it out and need a hand, feel free to mail me with the additional info
> (or the list, if you think the wonderful spam is suitable for this bunch o'
> vikings ;)

No warnings during compilation.
-- 
Larry W. Virden <mailto:lvirden at cas.org><http://www.usenix.org/events/tcl2k/>
<URL: http://www.purl.org/NET/lvirden/> <*> O- 
Unless explicitly stated to the contrary, nothing in this posting should 
be construed as representing my employer's opinions.
-><-



From gerrit.holl at pobox.com  Fri Jan 21 11:31:50 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 17:31:50 +0100
Subject: function "overloading"
In-Reply-To: <3888787d_2@goliath.newsfeeds.com>; from jblazi@netsurf.de on Fri, Jan 21, 2000 at 04:24:41PM +0100
References: <3888787d_2@goliath.newsfeeds.com>
Message-ID: <20000121173150.A1498@stopcontact.palga.uucp>

Janos Blazi wrote on 948468281:
> I have 3 classes called Tline, Tsegment and Tpoint, repectively. I'd like to
> write a function "subset" that cheks if one of the instances of the above
> classes  is a subset of another instance (as long as this is sensible).
> 
> I am afraid I cannot overload the name "subset" and additionally when subset
> has control it cannot decide the type of its arguments: Does this mean I
> have to write several *different* subset functions having different names?

>>> print issubclass.__doc__
issubclass(C, B) -> Boolean

Return whether class C is a subclass (i.e., a derived class) of class B.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From cpatti at black-racer.atg.com  Thu Jan 27 12:14:44 2000
From: cpatti at black-racer.atg.com (chris patti)
Date: 27 Jan 2000 12:14:44 -0500
Subject: A couple of (minor) things I'd like to see in the Python _environment_
Message-ID: <yu14sbza12z.fsf@black-racer.atg.com>

Firstly, let me just say that coming from a fairly Perl-centric
background, learning Python has been a breath of fresh air - things
actually MAKE SENSE more of the time than not :)

With that out of the way, there are a couple of things I genuinely
miss that I think would benefit Python tremendously:

* A command line driven documentation browsing system, similar to
 'perldoc' in the Perl world.  (Python's entities have doc strings -
  maybe those could be leveraged?)

* An interactive debugger invokable with a command line option to the
  Python interpreter. - I know about PDB, and it does in fact work,
  but oh how I miss the ability to simply do:

perl -d foo.pl

and from there be able to set breakpoints, examine the script in
operation etc.


Sorry if this steps on anyone's toes or hearkens back to Holy Wars
already fought long ago - these just _feel_ to me like limitations
that could (at least in the doco case) easily be fixed.

Thanks, 

-Chris P.




From mikael at isy.liu.se  Mon Jan 24 03:59:18 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Mon, 24 Jan 2000 09:59:18 +0100 (MET)
Subject: ANN: Python training in Sweden!
In-Reply-To: <000901bf6579$7bf53960$f29b12c2@secret.pythonware.com>
Message-ID: <XFMail.000124095918.mikael@isy.liu.se>

On 23-Jan-00 PythonWare wrote:
 >      http://www.pythonware.com/training/
 >      (in swedish only)

Which returns as

  http://www.pythonware.com/training/blue.css

which is a Not-Found-Booboo.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    24-Jan-00
Time:    09:58:03

This message was sent by XF-Mail.
-----------------------------------------------------------------------



From jamarijr at hotmail.com  Wed Jan  5 11:47:01 2000
From: jamarijr at hotmail.com (Arinte)
Date: Wed, 05 Jan 2000 16:47:01 GMT
Subject: Need help with PyRun_String
Message-ID: <84vsi0$3sb$1@nnrp1.deja.com>

How do you uses
PyObject* PyRun_String (char *str, int start, PyObject *globals,
PyObject *locals);

I don't know what to put in for the params (except str).

Also, I need tips on handling errors, like how can I get the trace
string thingie.

TIA


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


From news at dorb.com  Thu Jan 13 15:09:40 2000
From: news at dorb.com (Darrell)
Date: Thu, 13 Jan 2000 15:09:40 -0500
Subject: generating .pyc without running .py?
References: <85lb6i$js2$1@news1.tc.umn.edu>
Message-ID: <wfqf4.3019$hN5.92400@typhoon.nyroc.rr.com>

3.23 py_compile -- Compile Python source files.
The py_compile module provides a single function to generate a byte-code
file from a source file


--
--Darrell
"Brian Langenberger" <brian at brian.cbs.umn.edu> wrote in message
news:85lb6i$js2$1 at news1.tc.umn.edu...
> Is there a way to generate Python bytecode files (.pyc)
> without executing the original (.py) script?  I'd like
> to automate the generation process, but can't find a
> command line parameter that seems appropriate.
> --
> http://www.python.org/mailman/listinfo/python-list




From hei at adtranzsig.de  Thu Jan 27 03:38:19 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 27 Jan 2000 09:38:19 +0100
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com> <388F140B.C0AA1C99@erols.com> <388f88ca.1925884@news.erols.com>
Message-ID: <948966042.615175@news.adtranzsig.de>

Roey Katz schrieb in Nachricht <388f88ca.1925884 at news.erols.com>...
>The 'virtual classes' that Sweeney describes look to me an awful lot
>like something like this:
Looks very interesting!
By now, i solved the same problem (i think) by passing a class to the
constructor of some class in a framework, so as to have a sort
of "flexible factory":

class Base:
  def __init__(self,instclass)
    self.someinstance = instclass()

Maybe your approach makes it more readable... gotta try it out.

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


Roey Katz schrieb in Nachricht <388f88ca.1925884 at news.erols.com>...
>The 'virtual classes' that Sweeney describes look to me an awful lot
>like something like this:
>
>
> class myClass:
>     def __init__( self ): pass
>     def doSomething( self ): print 'instantiating'
>
> class myClass2:
>     def __init__( self ): pass
>     def doSomething( self ): print 'instantiating (2)'
>
> class Base:
>
>     virtualClass = None
>
>     def __init__( self ):
>          self.inst = self.virtualClass()
>          self.inst.doSomething()
>
>
> class Derived1( Base ):
>     virtualClass = myClass
>
> class Derived2( Base ):
>     virtualClass = myClass2
>
>
>-----------------
>Isn't this the same thing?  I noticed that the virtual classes in
>Sweeney's code sample were all of the same base class;  to
>me, it seems that Python's object model is one of policy, not rules:
>if an object responds to a message, so be it, regardless of what
>*kind* of object it is -- so we have a little more freedom with
>Python.  Correct me if I'm redundant?
>
>I enjoy these constructs because they allow me to design flexible
>interfaces in the base class that I don't have to change at all as I
>add new features (sounds coherent?) I'm using this for a role playing
>game.
>
>
>Roey




From cwr at crash.cts.com  Wed Jan 26 07:11:37 2000
From: cwr at crash.cts.com (Will Rose)
Date: 26 Jan 2000 12:11:37 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <948807929.472741@news.adtranzsig.de> <Thjj4.6690$S64.195053@nnrp1.uunet.ca> <948887438.236049@news.adtranzsig.de>
Message-ID: <86mo9p$sf$3@thoth.cts.com>

Dirk-Ulrich Heise <hei at adtranzsig.de> wrote:
: Warren Postma schrieb in Nachricht ...
:>Rather bloody minded of me to point this out, but hey "to each his own"...
:>
:>#define BEGIN {
:>#define END     }

: Of course; now convince people to *use* this throughout their programs.
: Until a committee can declare "{" and "}" as obsolete.

:>
:>Can I do the same somehow in Python?  ;-)
:>

: No. If you need optical block markers, i suggest this:

: def blah():
:   #begin
:   ...
:   #end

: Makes the code look nearly like Pascal; except for you don't get
: the "semicolon"-errors. Ain't that cute.

This is a very bad idea.  You see one set of delimiters, and the
interpreter sees another - not cool.


Will
cwr at cts.com



From BgPorter at NOacmSPAM.org  Mon Jan 31 12:30:50 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Mon, 31 Jan 2000 17:30:50 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <874sby5m6l.fsf@iname.com> <20000131153329.B19725@xs4all.nl> <Teil4.12357$t_.270545@news.rdc1.nj.home.com> <20000131171255.A1124@stopcontact.palga.uucp>
Message-ID: <eFjl4.12535$t_.271008@news.rdc1.nj.home.com>

"Gerrit Holl" <gerrit.holl at pobox.com> wrote in message
news:20000131171255.A1124 at stopcontact.palga.uucp...
> Brett g Porter wrote on 949330467:
> >
<<deletia>>
> > The original study was printed in the Nov 1983 Communications of the
ACM.
>
> Tell Linus!
Lemme at him....I'll take him on with one paw behind my back....





From tim_one at email.msn.com  Tue Jan 18 23:28:36 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 18 Jan 2000 23:28:36 -0500
Subject: Pardon my ignorance, data conversion
In-Reply-To: <38851FC9.E5CBC0E9@callware.com>
Message-ID: <000301bf6235$a71abf80$9a2d153f@tim>

[Michael P. Reilly sez "floor does funny things to negative
 numbers", Ivan defends its honor]

> I disagree.  They're not funny, only unexpected if you've
> been working with C or some other language that doesn't
> floor() or ceil() the same way Python does.  It's merely
> a different perspective on what's right.

The floor and ceiling functions predate computers, and there's no
disagreement about their definitions (coincidence or plot?  you be the judge
<wink>).  Python's floor & ceil *are* C's -- like most everything else in
Python's math module, they're tissue-thin wrappers around the platform C
library functions of the same names.  Modulo floating point endcases
(infinities & NaNs & such), their behavior is the same across platforms.

moving-left-on-the-number-line-is-a-matter-of-perspective-
    only-in-the-presence-of-drugs-ly y'rs  - tim





From skip at mojam.com  Fri Jan 28 18:05:20 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 28 Jan 2000 17:05:20 -0600 (CST)
Subject: existence of a variable
In-Reply-To: <86t4fu$s6u$1@nnrp1.deja.com>
References: <86t4fu$s6u$1@nnrp1.deja.com>
Message-ID: <14482.8368.71856.613717@beluga.mojam.com>

    Martin> What is the best way to determine existence of a variable in
    Martin> Python (I need to check if a variable is defined and if so,
    Martin> delete it)?

One option is

    try:
        del v
    except NameError:
        pass

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098
"Languages that change by catering to the tastes of non-users tend
not to do so well." - Doug Landauer



From bitbucket at isomedia.com  Wed Jan  5 14:03:21 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Wed, 05 Jan 2000 19:03:21 GMT
Subject: Python and ASP - foo.asp (1/1)
References: <_ji34.8719$523.343739@news.chello.at> <bBk34.1288$Cl3.6689@news-server.bigpond.net.au> <slrn84t6n0.q3j.aa126@fan1.fan.nb.ca> <BZA34.2308$Cl3.11615@news-server.bigpond.net.au> <387351CA.1EA77B30@gssec.bt.co.uk>
Message-ID: <38739579.333032435@mail.isomedia.com>

begin 644 foo.asp
M/"5`($Q!3D=504=%(#T at 4'ET:&]N("4^#0H-"CPE#0ID968 at 2%1-3%1A8FQE
M1G)O;4-O;&QE8W1I;VX@*&]#;VQL96-T:6]N+"!S0V]L;&5C=&EO;DYA;64I
M. at T*"7-(5$U,(#T@)SQT86)L92!B;W)D97(^7&X\='(^/'1H(&-O;'-P86X]
M,CXE<SPO=&@^/"]T<CY<;B5S/"]T86)L93Y<;B<-"@ES4F]W4&%T(#T@)SQT
M<CX\=&0^)7,\+W1D/CQT9#XE<SPO=&0^/"]T<CY<;B<-"@ES0F]D>2`]("<G
M#0H)9F]R('-+97D@:6X@;T-O;&QE8W1I;VXZ#0H)"7-";V1Y(#T@<T)O9'D@
M*R!S4F]W4&%T("4@*'-+97DL(&]#;VQL96-T:6]N("AS2V5Y*2D-"@ER971U
M<FX@<TA434P@)2`H<T-O;&QE8W1I;VY.86UE+"!S0F]D>2D-"B4^#0H-"CPE
M/2!(5$U,5&%B;&5&<F]M0V]L;&5C=&EO;B`H4F5Q=65S="Y397)V97)687)I
M86)L97,L("=297%U97-T+E-E<G9E<E9A<FEA8FQE<R<I("4^#0H\)3T at 2%1-
M3%1A8FQE1G)O;4-O;&QE8W1I;VX@*%)E<75E<W0N0VQI96YT0V5R=&EF:6-A
M=&4L("=297%U97-T+D-L:65N=$-E<G1I9FEC871E)RD@)3X-"CPE/2!(5$U,
M5&%B;&5&<F]M0V]L;&5C=&EO;B`H4F5Q=65S="Y#;V]K:65S+"`G4F5Q=65S
M="Y#;V]K:65S)RD@)3X-"CPE/2!(5$U,5&%B;&5&<F]M0V]L;&5C=&EO;B`H
M4F5Q=65S="Y1=65R>5-T<FEN9RP@)U)E<75E<W0N475E<GE3=')I;F<G*2`E
#/@T*
`
end


From senn at maya.com  Mon Jan 31 10:44:33 2000
From: senn at maya.com (Jeff Senn)
Date: 31 Jan 2000 10:44:33 -0500
Subject: Python Palm Pilot Hotsync Conduit
In-Reply-To: <mmillikan@vfa.com>'s message of "Mon, 31 Jan 2000 10:09:03 -0500 (EST)"
Message-ID: <wkoga2w8im.fsf@SNIPE.maya.com>

All this talk about Christian's SP and Python for the Palm Pilot
reminded me:

I have a (mostly complete) wrapper for the low-level Palm SyncManager
routines that would allow one to write HotSync conduits in Python (on
Windows).

I paused in the development (read: got distracted) so there are some
loose ends.  If anyone is interested in picking it up for the 'greater
good', I'll contribute my code...

-- 
-Jas
---------------------------------------------------------    / / |-/ \ / /|
Jeff Senn           412-488-2900      MAYA Design Group     /|/| |/ o | /-|
Chief Technologist  412-488-2940 fax  2100 Wharton Street  Taming Complexity
Head of R&D         senn at maya.com     Pittsburgh, PA 15203   www.maya.com




From eddyw at lsls4p.lsl  Wed Jan 12 10:58:18 2000
From: eddyw at lsls4p.lsl (Edward Welbourne)
Date: Wed, 12 Jan 2000 15:58:18 GMT
Subject: class methods ?
Message-ID: <200001121558.PAA15809@lsls4p.lsl>

Not in 1.3 (what I have available at work to test), and I doubt it in
1.5 also.  Basically, if a class has an attribute with a callable value,
the only way you can call it is by calling it as a method of an
instance, so it has to have a first argument for that purpose, even if
this is to be ignored.

You can, of course, cheat - using introspection.  In python 1.3, for
instance,

>>> class dummy:
... 	def __init__(self):
... 		try: count = dummy.invoked['__init__']
... 		except KeyError: count = 0
... 		dummy.invoked['__init__'] = 1 + count
... 	def printer(): print dummy.invoked
...	invoked = {}
... 
>>> dummy.printer()
TypeError: unbound method must be called with class instance 1st argument
>>> dummy.printer.im_func()
{}
>>> dummy()
<dummy instance at aeed8>
>>> dummy.printer.im_func()
{'__init__': 1}

though I think the introspection name (.im_func) is different in 1.5
(and you could use dummy.invoked.get('__init__', 0) in 1.5 to obtain
count, instead of needing the try...except).

On the other hand, I have a plan which will make it easy to have class
methods (among other tasty goodies) in python 2, if I can persuade the
Great, the Good and, most of all, the Guido in a fortnight's time, but I
have to finish turning it into something somewhat more intelligible ...
sadly, writing for humans isn't my forte.  More news by the end of
Sunday, I promise (I hate deadlines, but they're good for productivity).

	Eddy.



From fdrake at acm.org  Wed Jan 19 17:16:30 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Wed, 19 Jan 2000 17:16:30 -0500 (EST)
Subject: To decref or not to dercref that is my question???
In-Reply-To: <gXph4.77$N4.3733@newsread1.prod.itd.earthlink.net>
References: <gXph4.77$N4.3733@newsread1.prod.itd.earthlink.net>
Message-ID: <14470.14270.813845.558603@weyr.cnri.reston.va.us>

Arinte writes:
 > When using PyObject_GetItem, when I am done getting the value that I need
 > from it should I Py_XDECREF it?

  Yes.
  I'm working at getting all the required information into the API
reference manual.  If you are using the CVS version of Python, you can 
look in the file Doc/api/refcounts.dat; the comments at the top of the 
file explain the format.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From secret at email.com  Fri Jan 21 08:34:51 2000
From: secret at email.com (PSU boss)
Date: Fri, 21 Jan 2000 08:34:51 -0500 (EST)
Subject: Python Rocks!
In-Reply-To: <8671ub$77k$2@newshost.accu.uu.nl>; from m.faassen@vet.uu.nl on Thu, Jan 20, 2000 at 01:18:03PM +0000
Message-ID: <200001211334.IAA28640@python.org>

Martijn Faassen wrote on 948370683:
> Someone get the Python Secret Underground to do something about him. :)

He won't hurt us anymore.

regards,
PSU boss.




From claird at starbase.neosoft.com  Mon Jan 31 13:55:09 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 31 Jan 2000 12:55:09 -0600
Subject: Why Perl do But Python can't?
References: <200001310128.UAA02858@poverty.bloomington.in.us> <8737c1$76d$1@news.cz.js.cn>
Message-ID: <058AFBA1CD7F1621.791F88D00D532EC1.0128BF9429776D7B@lp.airnews.net>

In article <8737c1$76d$1 at news.cz.js.cn>, ???? <jfzhou at 188.net> wrote:
>Thank you help.
>But How I do it to solve the follow quertion:
>
>1:through the WWW to manage user,the CGI need root uid.
>2:how to read or write files that need some privilege ie. root ,admin etc...
>
>some one tell me
>1:use another programme to communite with CGI useing socket,but it is
>complex.
>2:use expect mod ,but python dosn't have.
>
>is there better solution?
			.
			.
			.
As best I can understand,
  use expect mod ,but python dosn't have.
refers to Expect.pm, and has no particular relation to
WWW, suid, sgid, or other elements of this post.

In fact Python does have analogues to Expect.pm.  The
little research I've done indicates that they're not
well-researched.  I've been considering writing up a
tutorial on their use and status ...  In the meantime,
I recommend those whom this topic interests visit
  ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/System/expy-0.4a.README
  ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/System/pipe-1.0.README
  ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/All/pyexpect.README

Thanks to Barry Warsaw for his help with these data.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From darrell at dorb.com  Fri Jan  7 15:49:24 2000
From: darrell at dorb.com (Darrell)
Date: Fri, 7 Jan 2000 15:49:24 -0500
Subject: XML to dictionary?
References: <FGALACFJFLICBBAA@my-deja.com> <01b901bf58b8$fadac010$0100a8c0@rochester.rr.com> <387631FF.5B10CDB8@prescod.net>
Message-ID: <0bcc01bf5950$ae8264a0$0100a8c0@rochester.rr.com>

From: "Paul Prescod" <paul at prescod.net>
>
> I'm a little surprised to see this comment in your code:
> 
> """Convert this function to 'C' for real speed."""
> 
> Too masochistic to write it yourself only once?? :)
> 
Ha!
If I though someone would actually use it, I'd love to do the 'C'.
Heck, I'd clean it up and write some doc if that happened.

As it is all of my posted code is just for fun so the masochism has limits.

--Darrell





From wtanksle at hawking.armored.net  Thu Jan 20 22:57:01 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 21 Jan 2000 03:57:01 GMT
Subject: What about REBOL (www.rebol.com)
References: <ngQh4.17853$pb2.1435429@tw11.nn.bcandid.com>
Message-ID: <slrn88fmtd.5fj.wtanksle@hawking.armored.net>

On Fri, 21 Jan 2000 03:20:51 GMT, Jayr Magave wrote:
>Sorry to introduce a strange subject to this group. Please, no flames.
>In fact, it isn't really strange -- I woul like to know if someone evaluated
>the new script language REBOL. It looked very powerful and high level to me.
>But lacks library support (specially database support).

Actually, I thought the nicest thing about it was its library support.  I
especially like how well thought out the libraries are -- they tend to
make their own syntax appropriate to the problem at hand, so solutions
drawn up using them tend to be rather transparent.

On the other hand, the language itself lacks expression delimiters.  This
makes complex expressions trippy -- it's hard to tell when you're calling
a function and when you're getting the value of a variable.  Functions
which take multiple parameters look _really_ odd.

For instance, here's a nasty Rebolism:

print sum 3 sum sum 6 7 4 5

How many arguments does 'sum' take?  What will this print?

Of course, the practical answer to this problem is "so don't do that".  If
you take up coding Rebol I'm certain that you'll find it reasonably easy
to stop yourself whenever you're tempted to solve a problem that way ;-).

>I am about to take my time to learn an useful multiplataform script language
>and Python looked very fine,  until I took a look at REBOL.

Ideally, Python will continue to look fine.  Rebol's a very nice language,
but so's Python.

Heck, not even discovering Joy
(http://www.latrobe.edu.au/www/philosophy/phimvt/j00syn.html), surely the
best language ever designed, has not sullied my happiness with Python.

(If you're looking for a useful scripting language, don't bother -- Joy's
pretty primitive in that respect.  However, if you want a really cool
programming language which will make you wish you'd discovered it before
you'd ever heard the word "lambda" -- read the pages NOW.  The best
algebraic introduction to a language I've EVER seen, and the simplest
proofs.)

Am I sick, or what?

>Jayr Magave

-- 
-William "Billy" Tanksley, in hoc signo hack


From scarblac-spamtrap at pino.selwerd.nl  Mon Jan 31 07:50:00 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 31 Jan 2000 12:50:00 GMT
Subject: sending keystroke to a dos application
References: <000501bf6be0$a8397f50$3acbd9c2@peridot.optichrome.com>
Message-ID: <slrn89b11p.ool.scarblac-spamtrap@flits104-37.flits.rug.nl>

Adrian Eyre wrote in comp.lang.python:
> > I have a dos application that runs in a dos window under NT or win98 that
> > takes interactive input. Basically numbers to choose from menu items and
> > return keys. I have not found a method to script this application by
> > sending keystrokes to the dos window. Is there a any possibility 
> > to do this with python?
> 
> Providing you don't actually need this app to run on MS-DOS, but just
> 98/NT in a DOS box, you can use:
> 
> import msvcrt
> key = msvcrt.getch()
> 
> ...to get a single keystroke.

I believe that's not what he meant. He has an existing application
(probably non-Python) that runs in a DOS box. He wants to script it
with Python by sending keystrokes to it.

I don't think that's possible, but I'm not a Windows expert.
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From alex at magenta.com  Sat Jan  8 08:55:57 2000
From: alex at magenta.com (Alex Martelli)
Date: Sat, 8 Jan 2000 14:55:57 +0100
Subject: RunPy (was py2bat (was: How do I make a Python .bat executable file?))
References: <6D8A17398E28D3119F860090274DD7DB4B3DAF@pces.cadlab.it>
Message-ID: <00f801bf59fd$99fdac80$c02b2bc1@martelli>

Piet van Oostrum writes:

> For perl there is a runperl.exe which, if copied to myscript.exe and run,
> calls perl myscript.pl with all arguments passed unchanged. It would
> probably be simple to clone this for python. I have never seen the source,
> however, and I am too lazy to write it myself.

Haven't seen the program you mention, either, but I've had a go
at writing a version for Python, just in case some should find it
useful.  http://aleax.supereva.it/Python/runpy.zip (2477 bytes)
has both the source (a single runpy.cpp file, to be compiled as a
console application) and the executable (runpy.exe, which is
compiled and linked to use the MSVCRT.DLL version of runtime).

This runpy.exe should do exactly what you mention; renamed to
foobar.exe, it looks for foobar.py along the PATH and, if found,
runs python.exe on it (assuming _that_ is along the PATH, too!)
with the same, unchanged arguments.

If anybody does download and try it, I'd be grateful for any
feedback, of course.

One can, of course, easily think of a few useful variations: e.g.,
a version that is a window app rather than a console one (and
which runs pythonw rather than python), options to look for
the script elsewhere than on the PATH (e.g., in the same
directory as the renamed runpy.exe itself, or on PYTHONPATH),
ways to pass option-flags to python.exe.  I'm quite willing to
entertain any such suggestions if anybody has any use for
them -- it's all really very simple (the current runpy.cpp is
77 lines including error diagnosis and reporting, for example).


Alex





From pinard at iro.umontreal.ca  Mon Jan 24 20:46:22 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 24 Jan 2000 20:46:22 -0500
Subject: Multidimensional Dictionaries: Possible?
In-Reply-To: w6uv@hotmail.com's message of "Tue, 25 Jan 2000 00:02:31 GMT"
References: <slrn88ppsg.49t.w6uv@voyager.home.com>
Message-ID: <oqpuuqx6s1.fsf@titan.progiciels-bpi.ca>

w6uv at hotmail.com (Jerry Gardner) writes:

> Are multidimensional dictionaries possible in Python? I'm sure they're
> possible, but I can't figure out the syntax.

> Here's how I do it in Perl:
> %dict = ();  # create empty dictionary
> $dict{$key1}{$key2} = $string;       # make an entry
> $newstring = ${dict{$key1}{$key2}};  # access an entry

dict = {}
dict[key1, key2] = string
newstring = dict[key1, key2]

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





From boud at rempt.xs4all.nl  Mon Jan  3 14:51:27 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 3 Jan 2000 19:51:27 GMT
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl> <84plci$ojt$1@news.fsr.net>
Message-ID: <84qujv$j9e$1@news1.xs4all.nl>

Mark 'Kamikaze' Hughes <kamikaze at kuoi.asui.uidaho.edu> wrote:
<...>

Like you, I'm not advanced enough to know whether I need garbage
collection or not and would dearly like a good folding editor (but not
to the point of writing one myself). But I'd like a really advanced
editor to play with, too - one that interprets all kinds of things in
commands, can do intelligent things with the relations that are buried
in my code. I've always found class browsers confusing, though. They
never seem to give enough information. 

-- 

Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt


From skaller at maxtal.com.au  Sat Jan  8 08:55:15 2000
From: skaller at maxtal.com.au (skaller)
Date: Sun, 09 Jan 2000 00:55:15 +1100
Subject: control structures (was "Re: Sins")
References: <Pine.LNX.4.10.9912261534560.8355-100000@palin.python>
Message-ID: <387741C3.B5A21926@maxtal.com.au>

Phil Jensen wrote:
 
> I forget if I posted this before, but what I'd like to
> see is the Zahn construct discussed in Knuth's article -
> Pythonically:
> 
>     loop until "element found" or "search failed":
>         ...
>         if blah1 == blah2:
>             "element found"
>         ...
>     when "element found":
>         ...
>     when "search failed":
>         ...

	I've read that article too, and tend to agree.
How about:

	with Found, NotFound:
		initialiser ...
	loop:
		loop body ...
		signal Found ..
	case Found:
		print "found it"
	case NotFound:
		print "couldn't find it"
	finally:
		cleanup ..

Semantics: each label in the 'with' part must be
given as a case. No exit from the loop exists,
except for the listed cases.

Implementation:

	with:
		class Found: pass
		class NotFound: pass
		..
	do:
		try:
			try:
				while 1:
					.. raise Found
			except Found: ...
		finally: ..


> IMHO, "break 3" would be awful.

	Agree. Shades of Tcl. :-(

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From kelly at poverty.bloomington.in.us  Sun Jan 16 21:52:38 2000
From: kelly at poverty.bloomington.in.us (Kelly Lynn Martin)
Date: Sun, 16 Jan 2000 21:52:38 -0500
Subject: Breaking out of walk() 
In-Reply-To: Your message of "Sun, 16 Jan 2000 16:36:10 GMT."
             <3881f20c.844046140@news1.on.sympatico.ca> 
Message-ID: <200001170252.VAA09211@poverty.bloomington.in.us>

On Sun, 16 Jan 2000 16:36:10 GMT, rjroy at takingcontrol.com (Robert Roy) said:

>Make sure you modify the list in-place. Don't do names=[] since that
>creates a reference to a different object. You can also choose not to
>recurse through certain directories.

That doesn't break out of walk entirely; it will still continue to
recurse through higher-level directories.  It just prunes.

If you want a complete abort, throwing an exception is the only way I
can think of to get out.

Kelly



From mlh at vier.idi.ntnu.no  Tue Jan  4 18:11:25 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 05 Jan 2000 00:11:25 +0100
Subject: Simple cartesian product
References: <iacpuvhqwdp.fsf@vier.idi.ntnu.no> <iacn1qlqvw7.fsf@vier.idi.ntnu.no>
Message-ID: <iack8lpqvoi.fsf@vier.idi.ntnu.no>

mlh at vier.idi.ntnu.no (Magnus L. Hetland) writes:

> def product(a,b):
>     a = reduce(lambda l, x: l + [x]*len(b),
>                [[]] + a)
>     b = b * len(a)
>     return map(None, a, b)
> 

Or, rather:

def product(a,b):
    a2 = reduce(lambda l, x: l + [x]*len(b),
               [[]] + a)
    b  = b * len(a)
    return map(None, a2, b)

(This is getting worse by the minute ;)

--

  Magnus
  Lie
  Hetland


From gerrit.holl at pobox.com  Tue Jan 18 16:58:26 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 18 Jan 2000 22:58:26 +0100
Subject: **kwargs
In-Reply-To: <862j09$o1k$1@nnrp1.deja.com>; from sp00fD@yahoo.com on Tue, Jan 18, 2000 at 08:38:36PM +0000
References: <862j09$o1k$1@nnrp1.deja.com>
Message-ID: <20000118225826.A2833@stopcontact.palga.uucp>

sp00fD wrote on 948224316:
> Does **kwargs represent a dictionary?  If so, can I have a method like
> such
> 
> def foo(self, foo=None, bar=None, **kwargs)
> 
> and then do something like
> 
> this_dict = {}
> ..insert stuff into this_dict..
> 
> my.foo(foo="something", bar="something else", this_dict)
> 
> ? How would I do something like this ?

Use apply:

>>> def f(foo=None, bar=None, **kwargs):
...     print "foo is: %s" % foo
...     print "bar is: %s" % bar
...     print "kwargs is: %s" % kwargs
...
>>> f(foo="something", bar="something else", fun="cool")
foo is: something
bar is: something else
kwargs is: {'fun': 'cool'}
>>> apply(f, (), {"foo": "something", "bar": "something else", "fun": "cool"})
foo is: something
bar is: something else
kwargs is: {'fun': 'cool'}
>>> print apply.__doc__
apply(function, args[, kwargs]) -> value

Call a function with positional arguments taken from the tuple args,
and keyword arguments taken from the optional dictionary kwargs.

Library reference:
   apply (function, args[, keywords])
          The function argument must be a callable object (a user-defined
          or built-in function or method, or a class object) and the args
          argument must be a sequence (if it is not a tuple, the sequence
          is first converted to a tuple). The function is called with
          args as the argument list; the number of arguments is the the
          length of the tuple. (This is different from just calling
          func(args), since in that case there is always exactly one
          argument.) If the optional keywords argument is present, it
          must be a dictionary whose keys are strings. It specifies
          keyword arguments to be added to the end of the the argument
          list.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From garry at sage.att.com  Fri Jan  7 13:46:09 2000
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Fri, 7 Jan 2000 18:46:09 GMT
Subject: CGI parameters
References: <8557si$10l$1@nnrp1.deja.com>
Message-ID: <38763471.E86F39F8@sage.att.com>

sp00fD wrote:
> 
> How do you query CGI parameters from radio buttons, urls, etc in
> python?  In perl it would look something like
> 
> $query = new CGI;
> $foo = $query->param("foo");

query = cgi.FieldStorage()
foo = query["foo"].value

you'll probably want to wrap the latter statement in a
try/except pair or precede it with a test for query.has_key( 'foo' )

-- 
Garry Hodgson			Every night 
garry at sage.att.com		a child is born
Software Innovation Services	is a Holy Night.
AT&T Labs			 - Sophia Lyon Fahs


From jand at easics.be  Sun Jan 16 14:42:24 2000
From: jand at easics.be (Jan Decaluwe)
Date: Sun, 16 Jan 2000 19:42:24 GMT
Subject: conflict between my package and os standard module
References: <Dkcg4.19933$zU5.216301@wbnws01.ne.mediaone.net>
Message-ID: <38823BF3.40D3AB96@easics.be>

Ben Wittner wrote:
> 
> I'm looking for help.
> 
> I have a bunch of extensions to standard modules that are pretty
> specific to the needs of my project.
> 
> I chose to create a package for my project, let's call it my_proj.

Good! Packages seem to be an underused feature in python. In the
recent "Learning Python" book they are even classified as "advanced".
 
> In there, I chose to create modules corresponding to standard
> modules (e.g., my_proj.os, my_proj.re) to hold the project-specific
> extensions of the standard modules.
> 
> I put the directory containing my_proj on my PYTHONPATH,
> but I did NOT put my_proj on my PYTHONPATH.
> 
> In my_proj/os.py I put the following line:
> 
> import os
> 
> Since my_proj is NOT on my PYTHONPATH, I thought that the line above
> would cause the standard module os to be imported.
> Instead, it seems that my_proj/os.py is imported.

You are victim of the "intra-package reference shortcut". The current
directory is always searched first for modules, hence the problem.

This was actually my first frustration when I started using 
python - fortunately also one of the last. I tried exactly the
same thing following the same - in my opinion solid - reasoning.

The reason for the shortcut is that packages are a relatively
new feature in python. Thanks to the shortcut, existing source
directories could easily be put into packages - without 
having to modify module importing code. OK - makes sense.

> Of course, I could rename my extension modules to things like
> os_x.py and re_x.py, but it seems like that should be unecessary
> since I'm already indicating that it's part of my project related package
> by prepending with "my_proj.".
> 
> Any ideas before I rename all my extension packages and change all
> references to them?

You could play tricks with __import__, but I don't recall exactly
how and I wouldn't recommend it.

For now, I would go for the renaming option (that's what I did.)

As a good solution, I have proposed to define a (possibly virtual)
package for the standard modules, so that you could say:

  from standard import os

to avoid the ambiguity. This proposal received some favorable
response, but I have no idea what the current status is.

Regards, Jan

-- 
Jan Decaluwe	       Easics              
Design Manager         System-on-Chip design services  
+32-16-395 600	       Interleuvenlaan 86, B-3001 Leuven, Belgium
mailto:jand at easics.be  http://www.easics.com


From claird at NeoSoft.com  Mon Jan 24 07:46:08 2000
From: claird at NeoSoft.com (Cameron Laird)
Date: 24 Jan 2000 06:46:08 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jan 24)
Message-ID: <2AD24EA06FAAB69A.57E72A251C97A8D0.B532FEFC6E43402C@lp.airnews.net>

The 8th International Python Conference begins with much-below-freezing
temperatures outside and an unprecedented growth of pre-registrants inside.
Expect a *very* productive week
        http://www.python.org/workshops/2000-01/

    Christian Tismer endows Python with continuations and co-rou-
    tines while trimming its stack and time weight.  It's even
    promised for a Palm
	http://www.deja.com/=dnc/getdoc.xp?AN=576789536

    John Grayson announces the appearance of the first book focused
    on Tkinter
        http://www.deja.com/=dnc/getdoc.xp?AN=575925861

    Michael P. Reilly taxonomizes four distinct approaches to
    dealing with a pty
	http://www.deja.com/=dnc/getdoc.xp?AN=574772821

    Aahz March provides an example of threading which is not only
    suitable for beginners but even useful for Web workers
	http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=570016307&fmt=text

    Edward Welbourne's excellent treatise on proper use of exceptions
	http://www.deja.com/=dnc/getdoc.xp?AN=575782124
    Note Fred L. Drake's stylistic follow-up (not yet URLized)

    Michael Hudson and Tim Peters preach better Python through
    self-discipline
	http://www.deja.com/getdoc.xp?AN=576276046

    jim kraai summarizes .ini knowledge beyond ConfigParser
	http://x35.deja.com/getdoc.xp?AN=576839546

    Do your Python applications want to know operating system version
    information, which libc is in charge, how many processors are run-
    ning, and so on?  M.-A. Lemburg provides answers
	http://starship.python.net/crew/lemburg/platform.py

    Telling testimony to Python's virtues (note the last sentence)
        http://www.deja.com/=dnc/getdoc.xp?AN=575774205        


========================================================================

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

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      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 <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From peter.stoehr at weihenstephan.org  Sat Jan 15 08:42:06 2000
From: peter.stoehr at weihenstephan.org (Dr. Peter Stoehr)
Date: Sat, 15 Jan 2000 14:42:06 +0100
Subject: Python and Tkinter Programming by John E. Grayson
Message-ID: <3880792E.79A874B6@weihenstephan.org>

Hi out there,

is the book mentioned in the subject already available ?

Greeting from Bavaria
    Peter
-- 
---------------------------------------------------------------------------
   Dr. Peter Stoehr --- Teisenbergweg 6 --- 85435 Erding --- 08122/47232
      http://www.weihenstephan.org/home/ak/support/team/jpg/peter.jpg
---------------------------------------------------------------------------
		I'm the terror that flaps through the night


From tjg at avalongroup.net  Thu Jan 20 02:52:07 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Wed, 19 Jan 2000 23:52:07 -0800
Subject: Mapping Column Names to results using DBI
References: <200001200705.XAA06134@netcom.com>
Message-ID: <3886BEA7.DD11FFE3@exceptionalminds.com>

Aahz Maruch wrote:

> The point is that if you are choosing the order of column names in your
> SELECT, then it doesn't matter if the underlying table changes
> structure.  It is now safe to index by ordinal position.  Moreover, if
> you want to get really fancy, you could make up a list of field names,
> string.join() it for the SELECT, then map the list to a dict to get the
> ordinal position for each field name.

Sorry for being thick as a brick on that one. Thanks for the
clarification much appreciated!

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From mfletch at tpresence.com  Mon Jan 31 13:22:29 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Mon, 31 Jan 2000 13:22:29 -0500
Subject: Whitespace delimiters suck
Message-ID: <CC59868A0A76D311B1F50090277B321C1F30D6@VRTMAIL>

Using voice dictation on my machine in PythonWin, this sequence becomes:

"class test colon new-line 
py-def test underscore function open-paren close-paren colon new line
print open-quote foxtrot oscar oscar close-quote new-line"

and then I hit the backspace twice. Of course, I don't actually use the
(very awkward, I agree) navigation features (or even the
character-by-character spelling for that matter). Using the voice dictation
for basic input, using plain-English spellings, and creating a few simple
macros (e.g. py-def (def), py-equals (==), triple-quote
(\No-Space'''\No-Space), doublex (wx\No-Space \cap)) and redefining
underscore to have no spacing around it, while in an auto-indenting editor,
gives you a decent programming environment (though certainly not perfect).
You still need your hands, but you can let them hang at your side most of
the time. Compared to the pain of typing, a considerable improvement, but
still a long way from perfect.

This is using Dragon NaturallySpeaking and a little application for
customising its vocabulary, incidentally.

Enjoy,
Mike

-----Original Message-----
From: Eric S. Johansson [mailto:esj at harvee.billerica.ma.us]
Sent: Monday, January 31, 2000 12:42 PM
To: python-list at python.org
Subject: Re: Whitespace delimiters suck
...
which is spoken:
"class test colon new-line indent delta echo foxtrot test no-space
underscore 
nospace function nospace left- paren right-paren indent new-line print
open-quote 
foxtrot oscar oscar new-line dedent new-line dedent"
...



From claird at starbase.neosoft.com  Tue Jan 25 09:37:34 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 25 Jan 2000 08:37:34 -0600
Subject: Big languages, little languages, room for all (was: tabs do WHAT?)
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k78t$thg@newton.cc.rl.ac.uk>
Message-ID: <57BC36B7CC8A3F4D.7467FEC6D4AAD55D.83A104115B9D7A8F@lp.airnews.net>

In article <86k78t$thg at newton.cc.rl.ac.uk>,
Richard Brodie <R.Brodie at rl.ac.uk> wrote:
>
>"Thomas Hamelryck" <thamelry at vub.ac.be> wrote in message news:86joac$l2s$1 at mach.vub.ac.be...
>
>> BTW, people here seem to have adopted the idea that if you don't like
>> a particular feature of a language, you should abandon it (e.g., "You
>> don't like indentation? Use Perl!").
>
>Adding features to a language always seems easy but that is the best
>reason for not doing so. There is the temptation to make the language
>the union of all the desired features. It's all been talked around before,
>and Guido has said what his feelings are. In the end, I think you've
>either got to respect that decision, or fork the language.
>
>> You can find a very nice description of this process with respect to C++
>> in "The Evolution of C++ : Language Design in the Marketplace of Ideas",
>> by Jim Waldo (Editor), James Waldo (Editor).
>
>C++ has to be the classic case of a language that evolved rather than
>was designed: PDP-11 assembler with OO bits tacked on. If we can
>learn something from its evolution, it should be not to do the same
>again.
>
>'Oh, that way madness lies; let me shun that.' - King Lear
>
>
Some embrace madness, and wisely.

I think it's my turn this month to argue for there being
more than one way to do things in more than one way.  I
happen to agree that Python is good at being clean,
comprehensible, more-minimal-than-maximal, conservative
in its core, and so on.  C++ certainly has problems; I
happen to believe, though, that it's OK for it to be an
evolved-rather-than-designed monster, that there's a
place for such committee work, and that our discipline
might even correctly generate such devil spawn again
(don't kid yourself about Java's coherence).  I'm just
thankful that Guido has fulfilled *his* vision so
expertly.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From wtanksle at hawking.armored.net  Sat Jan 29 01:21:33 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 29 Jan 2000 06:21:33 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <86sfmd$ktl$1@mach.vub.ac.be>
Message-ID: <slrn8952eo.gfa.wtanksle@hawking.armored.net>

On 28 Jan 2000 16:21:33 GMT, Thomas Hamelryck wrote:

>If braces are used, you can choose between a myriad of tools to get 
>consistent indentation (there are lots of C and C++ beautifiers out
>there). However, if the indentation in python is screwed up you 
>are lost completely. No tool can correct that (yes and I do know
>pindent).

The whitespace-eating nanoviruses which so persistently plague so many
people who have never seriously used Python are only a tiny mutation away
from eating curlybraces as well, so I wouldn't be so complacent.  If they
can eat a tab and destroy a single line of Python, they can just as easily
eat a curly brace and destroy an entire function of C.

(Actually, I have spent a few hours cleaning up C code where the braces
had gotten confused and someone had deleted too many -- it took a while to
figure out which one went where.)

>Thomas Hamelryck    Institute of Molecular and Structural Biology

-- 
-William "Billy" Tanksley, in hoc signo hack


From tseaver at aftershock.neosoft.com  Tue Jan  4 12:23:05 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 4 Jan 2000 11:23:05 -0600
Subject: Python Conference -- Early Bird Registration ends Jan 5!
References: <200001031737.MAA24342@eric.cnri.reston.va.us> <slrn872g4o.39e.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <949AD391BED458EE.D5B7C2AA267CBF06.1E724640DD75E83B@lp.airnews.net>

In article <slrn872g4o.39e.scarblac-spamtrap at flits104-37.flits.rug.nl>,
Remco Gerlich <scarblac-rt at pino.selwerd.cx> wrote:
>Guido van Rossum wrote in comp.lang.python:
>> Make the Python conference the first conference you attend in the new
>> millennium!
>
>But doesn't the new millennium staD^&**FGSF&())KJDFKDSSAACNSD28947
>
>NO CARRIER

Perl is off-topic here. :)

But-before-swine?'ly,

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From emile at fenx.com  Mon Jan  3 09:14:35 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 3 Jan 2000 06:14:35 -0800
Subject: setting time?
Message-ID: <019901bf55f4$df18f840$01ffffc0@worldnet.att.net>

On my RH6.1, when logged in with the appropriate permissions, this
works:

import os
os.popen('date -s "Jan 03, 2000 06:06"')

HTH

--

Emile van Sebille
emile at fenx.com
-------------------


<moonseeker at my-deja.com> wrote in message
news:84ptud$vfh$1 at nnrp1.deja.com...
> Hi again!
>
> Noone an advice to set the system time?
>
> Mike
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list
>
>






From n89553 at stud.upb.de  Mon Jan 31 12:24:06 2000
From: n89553 at stud.upb.de (Mirko Liss)
Date: Mon, 31 Jan 2000 18:24:06 +0100
Subject: Namespace pollution
References: <949332864.553076@news.adtranzsig.de>
Message-ID: <3895C536.B4E4DADB@stud.upb.de>

Dirk-Ulrich Heise wrote:

> I've got one problem with all this stuff:
> I end up with dozens of helper functions that are
> called exactly one time. They pollute the namespaces
> of the module or class they're defined in. I hate polluted
> namespaces.
> 
> How do others cope with this problem (that does not only
> apply to Python)?
Use double underscores to hide any object you don?t 
want to be imported from your module.

regards,

Mirko

-- 
M Li?,  <n89553 at squid.upb.de>


From arcege at shore.net  Wed Jan 19 12:51:27 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 19 Jan 2000 17:51:27 GMT
Subject: Interacting with external program -pty?
References: <3DB0284E2B7AD21189A400805F19DB280129E238@se-hsl-mail010.hassle.se.astra.com>
Message-ID: <zQmh4.3922$Ef6.1039939@news.shore.net>

Niklas.Blomberg at hassle.se.astra.com wrote:
: Dear all,

: I would appreciate some advice on how to script an external interactive
: process(program) using python. In addition to sending commands (which is
: straightforward) I need to be able to capture the output (which can be
: extensive) for further downstream processing. Since the program expects a
: terminal pipes don't work - the output is buffered - I have started to look
: at the pty module. This is uncharted territory for me and any advice both on
: possible solutions and perhaps availiable example code would be most
: appreciated.

: I'll be happy to post a summary.

There are a number of techniques you can use:

  1)  standard library module "pty"
      >>> import os, pty
      >>> master_fd, master_pty = pty.master_open()
      >>> slave_fd = pty.slave_open(master_pty)
      >>> os.write(master_fd, "hello\n")
      6
      >>> os.read(slave_fd, 6)
      'hello\012'
      >>> os.close(master_fd)
      >>> os.close(slave_fd)
    This needs to be wrapped with other code; there is a specialized
    "fork" function in the method.

  2)  the ExpectPy module (<URL: http://starship.python.net/~arcege/>)
      >>> import ExpectPy
      >>> spawn_id = Expectpy.spawn('/bin/cat', 'cat')
      >>> spawn_id.send('hello\r\n')
      >>> spawn_id.expect((ExpectPy.GLOB, "*\n", None))
      >>> spawn_id.match
      'hello\015\012hello\015\012'
      >>> spawn_id.close()
    The message prints twice because it would in /bin/cat at the
    terminal (cf. half versus full duplex).

  3)  depending on what you want: the standard library module "telnetlib"
    This module is to be used to interact with a telnet session (not
    with the telnet program).

  4)  fork, create pipes and exec
    This is the low-level stuff.
      >>> import os
      >>> def myspawn(*args):
      ...   (cld_reader, par_writer) = os.pipe()
      ...   (par_reader, cld_writer) = os.pipe()
      ...   pid = os.fork()
      ...   if pid:  # parent
      ...     os.close(cld_reader)
      ...     os.close(cld_writer)
      ...   else:    # child
      ...     os.close(0); os.dup(cld_reader)  # stdin
      ...     os.close(1); os.dup(cld_writer)  # stdout
      ...     os.close(2); os.dup(cld_writer)  # stderr
      ...     os.close(cld_reader); os.close(par_writer)
      ...     os.close(par_reader); os.close(cld_writer)
      ...     os.execvp(args[0], args)
      ...     os.exit(2)  # if we canot exec()
      ...   return (pid, (par_reader, par_writer))
      ...
      >>> (pid, pipe) = myspawn('cat')
      >>> os.write(pipe[1], 'hello\n')
      >>> os.read(pipe[0], 6)
      'hello\012'
      >>> os.close(pipe[3])
      >>> os.waitpid(pid, 0)
      (82572, 2)
      >>
    Notice that the stderr goes the to same location as stdout, just
    like pty's and terminals.  This can be changed tho. ;)

I hope this helps.
  -Arcege



From neelk at brick.cswv.com  Wed Jan 19 19:25:18 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 20 Jan 2000 00:25:18 GMT
Subject: Python Rocks!
References: <000401bf623e$16e712c0$9a2d153f@tim>
Message-ID: <slrn88cnpm.7ts.neelk@brick.cswv.com>

Tim Peters <tim_one at email.msn.com> wrote:
> [tye4 and Tom Culliton bump heads over whitespace -- who
>  could have imagined that the absence of something could
>  leave such a dent <wink>? ]

Wouldn't that be a dedent? :)

The only time indentation has been a problem for me is when using
eval/exec, and encoding programs as strings is basically the wrong
thing to do anyway. This fact has occasionally tempted me to write a
PythonGen module along the same lines as Robin Friedrich's HTMLgen. It
would make writing things like code walkers and program transformers
much more elegant. It seems like an idea obvious enough that someone
else must have beaten me to the punch, though.


Neel


From thomas at xs4all.net  Fri Jan 28 11:37:54 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 28 Jan 2000 17:37:54 +0100
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
In-Reply-To: <388FD0F5.B890FCFD@callware.com>; from ivanlan@callware.com on Wed, Jan 26, 2000 at 10:00:37PM -0700
References: <816010E2456BD111A48700805FBBE2EE01C60408@ex-quebec-u1.baan.com> <388FC511.C776341A@one.net.au> <00a101bf6882$aa7cb540$9600a8c0@execpc.com> <388FD0F5.B890FCFD@callware.com>
Message-ID: <20000128173750.A16093@xs4all.nl>

On Wed, Jan 26, 2000 at 10:00:37PM -0700, Ivan Van Laningham wrote:

> > How did this message make it into the list in the first place?

> > Did the spammer fake his/her sending address as that of a valid
> > list member, or is the list not sent to allow only registered members
> > to post?

> Because it's a gatewayed newsgroup, the Python mailing list can't reject
> non-subscribers.

> <sad-sad-story>-ly y'rs,

Actually, I think it would be a Big Mistake(tm*) or at least a Not Very
Smart Thing(c*) to block the list for non-subscribers. It would require
people to subscribe before they can ask their questions, and it would be
nasty for people who use more than one email address, or post with a
different one than they receive on. I manage a couple of closed lists,
company-lists for employees only, and even with less than 100 subscribers it
gets annoying as hell.

Much, much better to be open, and to deal with spam when it arrives -- by
examining the Recieved: lines and complaining to the origin of the spam,
plus possible relayers inbetween.

*)-copyrights-and-trademarks-of-Microsoft-ofcourse-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From steve_w at localhost.localdomain  Sun Jan 30 07:34:14 2000
From: steve_w at localhost.localdomain (steve_w at localhost.localdomain)
Date: Sun, 30 Jan 2000 12:34:14 GMT
Subject: [Newbie Problems] AttributeError from CGI script?
References: <5650A1190E4FD111BC7E0000F8034D26A0F2A1@huina.oceanic.com>
Message-ID: <slrn898d3h.86h.steve_w@localhost.localdomain>

I checked and I don't appear to be assigning anything to a variable
call string.  Thanks for the suggestion though.  The server I'm
working from is using Python 1.4 on Solaris 2.6.  They promised
me they'd upgrade a few weeks ago so it'd match my documentation 
and local version but so far no dice.

To make matters worse I'm now getting some different error which seems
to be from the CGI Wrapper software.

Steve Wall


Doug Stanfield wrote:
>Its just a guess, but make sure you haven't made an assignment to string
>someplace in your code.  In other words, I'm guessing you've done the
>equivalent of:
>
>[dougs at lawehana dougs]$ python
>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 string
>>>> line = "the placeholder"
>>>> string = "gum up"
>>>> line = string.replace(line,"placeholder","works")
>Traceback (innermost last):
>  File "<stdin>", line 1, in ?
>AttributeError: 'string' object has no attribute 'replace'
>
>I'm guessing because my error message is somewhat different.  It sometimes
>helps to know what platform and version you're using, and usually adding the
>least amount of code that duplicates the error helps also.
>
>Hope this helps.
>
>-Doug-
>
>> 
>> 
>> I'm trying to use Python in a CGI script for the first time and 
>> have run into a problem I don't understand.  I get the subject
>> error message when trying to do a string.replace using one of
>> the values returned by the form that trigers the script.
>> 
>> The line of code looks like:
>> 
>> line = string.replace(line,"placeholder",form["target"].value)
>> 
>> I get an error message back saying:
>> 
>> Traceback (innermost last):
>> File "[my script's file name]"
>>    line = string.replace(line,"placeholder",form["target"].value)
>> AttributeError: replace
>> 
>> Could someone help me figure out what this error message means?
>> 


From tjdurso at slip.net  Sun Jan 23 02:31:38 2000
From: tjdurso at slip.net (Thomas D'Urso)
Date: Sat, 22 Jan 2000 23:31:38 -0800
Subject: New install. Getting ImputError
Message-ID: <388AAE5A.B20A7BCE@slip.net>

Just installed / started with Python and reading Learning Python. When I
try to execute this code:

   import test
   from test import vtest    # <-- error on this line
   print vtest

file "test" contents:

   vtest = 'this is a test'

I get debugger error for 2nd line:    ImputError: cannot input name
vtest
The global variable says ImputError is:     Input can't fine module, or
can't find name in module.

It give no other errors as far as opening the file. Working on a Mac G4
with OS 9. Python 1.5.2.
Doing the "import test.autotest" reports no errors.

TIA Thomas






From samschul at pacbell.net  Wed Jan 19 21:30:02 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Thu, 20 Jan 2000 02:30:02 GMT
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <388670D2.2FA842F6@starmedia.net> <Pine.SOL.3.91.1000120001532.29235B-100000@sunse>
Message-ID: <865rur$5ud$1@nnrp1.deja.com>

>
> Yes, i'm saying that 'eq', 'ne' are more readable then '==' '!=',
'<>',
> 'next' shorter then 'continue' and '..' operator can be a good thing
to
> consider.
>
> evguenii
>
>

I like the following syntax for readability, and feel the this should be
implimented in Python:

	sub	ax,bx		# Lets keep the python comment symbol
	jne	Mylabel         # We also need labels
        rtn			# Notice all commands are three chars
Mylabel
	add	00h,ax

The above really simplifies things.

I love taxonomy, and want to see more confusing options in a language.

Sam Schulenburg



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


From tjg at avalongroup.net  Thu Jan 20 20:49:37 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Thu, 20 Jan 2000 17:49:37 -0800
Subject: DBI Commentary
Message-ID: <3887BB31.A9B8E90F@exceptionalminds.com>

Hi all,

Thought I would take an item that I have been working on and commenting
on on the main list and move it here in the hope of being more on-topic.
Please forgive this newbie for jumping in on the deepend of a discussion
that has probably been going on for a long time with many minds more
familiar with the subject than I.

There has been some discussion on the main list regarding the
retrieving  of column headings using DBI databases. As far as I can tell
from the spec, associating column names with data retrieved from a
SELECT does not happen automatically.

Thanks particularly go to Gordon, Aahz and Lance as well as a number of
others in my journey to understanding DBI. Now that I think I understand
it, I'd like to make some recommendations (As I said last night, my PSA
membership form is on its way today so I can fully participate in the DB
sig).

First, embedding SQL in Python, even using the """SQL""" method is
really ugly especially while keeping <tye4>indentation</tye4> consistent
and it get's phenomenally ugly if the queries get at all complex. I
don't think there is a solution to this problem, and I would hazzard a
guess that no matter what language is in use it would be equally ugly.
That said, anything that can be done to keep the ugliness to a minimum
would be a good thing.

Aahz made the very valid point that specifying columns in your query
*guarantees* tuple return order. I had not fully appreciated that, and I
have put that into use in a number of places. However, I have to agree
with Lance when he says that "SELECT * FROM table" is not as often a bad
thing as Aahz claimed. My reasons for that are far less technical than
Lance's, my reasoning is just that it adds to the ugliness mentioned in
the prior paragraph. Using "SELECT field1, field2, field3 FROM table"
simply becomes *far* to cumbersome and difficult to read/debug if you
get beyond five or so fields. If you are dealing with a large number of
columns (my current project has one table with 33). Debugging an error
becomes an exercise in frustration (especially when using the """SQL"""
method as the interpreter points the error message at the first line of
the multi-line execute statement.

Lance posted some very nice code to build dictionaries out of the
results from a query... 

dsc = cur.description
l = len(dsc)
lst = cur.fetchall()
rtn = []
for data in lst:
    dict = {}
    for i2 in range(l):
        dict[string.lower(dsc[i2][0])] = data[i2]
        rtn.append(dict)
return rtn

It seems to me that the DBI spec *should* mandate that the results of a
fetchXXX() return not a tuple, but a dictionary, or a list of
dictionaries. The problem is that a change like this would break a large
amount of existing code. So, I'm trying to think of an appropriate
extension (perhaps a new argument to the fetchXXX()'s)

It is possible that I am missing something already specified that has
not been implemented in my tools, but I have been unable to find it.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From robin at jessikat.demon.co.uk  Mon Jan 24 12:16:47 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 24 Jan 2000 17:16:47 +0000
Subject: Stackless 1.01 Homepage
In-Reply-To: <388C7D4A.D873AA@tismer.com>
References: <388BE68E.B26985CA@tismer.com>
 <$WXiIAAO+Dj4Ew2W@jessikat.demon.co.uk> <388C7D4A.D873AA@tismer.com>
Message-ID: <zMBtEAA$jIj4Ewm7@jessikat.demon.co.uk>

In message <388C7D4A.D873AA at tismer.com>, Christian Tismer
<tismer at tismer.com> writes
>
>
>Robin Becker wrote:
>> 
>> In article <388BE68E.B26985CA at tismer.com>, Christian Tismer 
><tismer at tismer.com> writes
>> 
>> I'm very impressed. here are the pybench timing differences (absolute and 
>expressed as % of
>> original python) for my 333Mhz win32 machine. here negative is bad for 
>stackless, positive is
>> good. My only disappointment is that Zope wouldn't start up.
>
>Whow. I guess there is a bug in exec. In any case, it cannot be a
>conceptual but but just a failure of mine. Idle, PyWin, pystone,
>Metakit, they all work fine.
>
>Will look into it after the conference.
>Thanks for giving the measures.
>
....

Zope uses several deep changes to standard python eg ExtensionClass,
Aquisition which may be the cause of the problem.
-- 
Robin Becker



From hanche at math.ntnu.no  Mon Jan 24 12:26:49 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 24 Jan 2000 12:26:49 -0500
Subject: tabs do WHAT?
References: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net> <86h5ep$iq$1@mach.vub.ac.be> <948721268.245257@news.adtranzsig.de>
Message-ID: <pcoya9f4bza.fsf@math.ntnu.no>

+ "Dirk-Ulrich Heise" <hei at adtranzsig.de>:

| Thomas Hamelryck schrieb in Nachricht <86h5ep$iq$1 at mach.vub.ac.be>...
| >People here (including Guido Himself) get extremely picky when the
| >use of indentation is criticised.
| 
| Go into a C++ newsgroup and suggest they get rid of the curlys.
| Or go into a Lisp group and tell them you love Lisp to death
| except for the brackets.
| I think you'll harvest some responses like "Go away and use a
| different language." What do you expect?

Perhaps more to the point, a good editor is absolutely essential with
almost any modern language.  Lisp is truly a pain without a good
editor.  With a good one, like emacs, I not only get the editor to
show me mathcing parentheses, but it can indent any program fragment
automatically, immediately highlighting any mistake in the parentheses
by creating an indentation different from what I thought the program
structure was like.  Likewise, in python, I never use the space bar
for indentation; emacs (or IDLE, now that I started using it) does it
for me.  The editor also has features for indenting or outdenting
blocks of code to the appropriate level for those circumstances when
you change the structure of the code or copy blocks of code from one
place to another place with a different indentation leve.  Use these
features, and all concerns about the unfriendliness of this or that
indentation scheme, parentheses or curly brackets etc melt away like
icicles in the sun.
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From tim_one at email.msn.com  Tue Jan 25 02:34:24 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jan 2000 02:34:24 -0500
Subject: tabs do WHAT?
In-Reply-To: <86h5ep$iq$1@mach.vub.ac.be>
Message-ID: <000301bf6706$99fca7c0$632d153f@tim>

[Warren Postma]
> If Guido reads this, please note my personal feature request
> for Python 2.0:  Make the "whitespace" stuff optional, for
> people who like weird syntactical features only. I'll love]
> you forever. Thanks.

[Thomas Hamelryck]
> This will not happen.

Good bet.

> People here (including Guido Himself) get extremely picky
> when the use of indentation is criticised.

Guido hasn't responded one way or the other to indentation criticism in
years, and I doubt you'll find one of his old responses being "extremely
picky" either (unbending, yes, but conviction isn't pickiness).

> They do not realize that this feature is an important obstacle
> for many people

Neither is Guido an idiot <0.9 wink>.  Of *course* he knows it's a
show-stopper for some people.  But so are many other things.  Every language
designer knows this, and the successful ones figure out early there's no
future in trying to make everyone happy.  There is a future in keeping a
language's philosophy consistent for as long as possible, and the use of
indentation to denote block structure wasn't an arbitrary choice in Guido's
view of the world.  He realizes it's an important obstacle; do "they"
realize it's an important manifestation of Python's philosophy?

> (and since these people have stopped using python they are
> obviously not posting here).

Then what keeps this thread going year after year <wink>?

> Indentation is great, but should not be part of the syntax
> of a language.

like-sex-is-great-but-should-not-be-part-of-life-ly y'rs  - tim





From esj at harvee.billerica.ma.us  Mon Jan 31 12:42:23 2000
From: esj at harvee.billerica.ma.us (Eric S. Johansson)
Date: 31 Jan 2000 17:42:23 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com> <86np80$jhp$1@harvee.billerica.ma.us> <86ntlg$2tc$1@news.ycc.yale.edu> <86pi88$n3n$1@harvee.billerica.ma.us> <38918EE7.38A43687@electricorb.com>
Message-ID: <874hhv$dnf$1@harvee.billerica.ma.us>

Glen Starchman <glen at electricorb.com> wrote:
> class test:
> indent
> def test_function():
> indent
> print "foo"
> dedent
> dedent

which is spoken:
"class test colon new-line indent delta echo foxtrot test no-space underscore 
nospace function nospace left- paren right-paren indent new-line print open-quote 
foxtrot oscar oscar new-line dedent new-line dedent"

now, if you really want to get ugly go change the function name from test_function
to test_subroutine when the only navigation capability you have is by counting 
lines and characters

believe me, it gets old real fast...

-- 
Eric S. Johansson    esj at inguide.com	esj at harvee.billerica.ma.us
This message was composed almost entirely using NaturallySpeaking


From mal at lemburg.com  Mon Jan 24 11:16:39 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 24 Jan 2000 17:16:39 +0100
Subject: Bug in Calendar Module
References: <22E71DAEC504D111B78100805FFE9DC72F03CAD4@pfs21.ex.nus.edu.sg> <86agsj$gst$1@nntp3.atl.mindspring.net>
Message-ID: <388C7AE7.2A335A4F@lemburg.com>

Aahz Maruch wrote:
> 
> In article <22E71DAEC504D111B78100805FFE9DC72F03CAD4 at pfs21.ex.nus.edu.sg>,
> Liew Pol Yee <polyp at cyberway.com.sg> wrote:
> >
> >It seems to be unable to accept dates after January, 2038. Thanks for any
> >help!
> 
> I think mxDateTime might work better for you.  Don't remember whether it
> has the equivalent of calendar in it.

It doesn't, but adding one on the basis of calendar.py should
be easy: all the needed APIs are there.

BTW, mxDateTime can handle dates from -5851455-01-01 00:00:00.0
to 5867440-12-31 00:00:00.00 (on 32-bit machines). Should be
enough for most apps, I guess ;-) ... ok, just showing off a little...

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




From Lance.Ellinghaus at avnet.com  Thu Jan 20 18:13:13 2000
From: Lance.Ellinghaus at avnet.com (Lance Ellinghaus)
Date: Thu, 20 Jan 2000 15:13:13 -0800
Subject: Python Rocks! - get rid of colons
Message-ID: <8825686C.007F912D.00@avnet.com>

If you had bothered to look into how Python works, you would find that there are
two parts to it:

1) A compiler that generates byte-code (Similar to the old UCSD p-Code PASCAL
compilers used to do. Great fun to decode those!, and what the javac program
does)
2) A byte-code processing machine (Virtual Machine).

Java has both and UCSD PASCAL had both. They are/were two separate and distinct
steps.

With Python, the first step is done automaticly for you if needed, but skipped
if not needed. Kinda like auto compiling the .java files into .class files if
they needed it and if they were available.

So can we drop the "If you had bothered to use Python...." thread.. ok? This is
wasting way too much bandwidth.


-------From: "tye4"  on 01/20/2000 4:51:48 PM-------

From: "tye4" <tye4 at yahoo.com>
To: python-list
cc:
Subject: Re: Python Rocks! - get rid of colons







Ivan Van Laningham <ivanlan at callware.com> wrote in message
news:388774C7.1784B662 at callware.com...
> Hello all--
>
> If you had bothered to use Python, you would already know that it is not
> a compiler but an interpreter.
>
If you have bothered to use Python, you should know that it is neither a
compiler nor an interpreter - but something in between: a byte-code
compiler. That's why it's fast and not sluggish like many BASIC
interpreters.

-tye4




--
http://www.python.org/mailman/listinfo/python-list





From donn at u.washington.edu  Mon Jan  3 12:24:08 2000
From: donn at u.washington.edu (Donn Cave)
Date: 3 Jan 2000 17:24:08 GMT
Subject: Finding where Python is on a Unix server through Telnet?
References: <slrn86uhb3.b66.thantos@chancel.org> <386EA0EF.F7A72441@jslove.net> <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com>  <84ncfb$e5q$0@216.39.151.169>
Message-ID: <84qlvo$bkm$1@nntp6.u.washington.edu>

Quoth thantos at chancel.org (Alexander Williams):
| On 2 Jan 2000 11:23:23 GMT, Donn Cave <donn at oz.net> wrote:
| >"locate" is a new one on me, not available on Digital UNIX.
|
| On Tru64 (*grumble grumble*) try 'whereis python'.  

Well, on Digital UNIX I also find whereis:

  $ whereis python
  python:
  $ python
  Python 1.5.2 (#1, Apr 14 1999, 13:46:13) [C] on osf1V4
  Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
  >>>

Tru64 may have revised this utility and improved the odds it will
find anything, but the point that emerges is ``what is the question?''

We're looking for python, for sure, but ... The official Python
installation from DEC (wouldn't that be nice)?  Use "whereis".
Python in Suzie's home directory, or a mutant Python installed
in an obscure place by some system programmer?  Use "locate"
(if it works as described in another followup.)  Python already
running?  Use "sys.executable" (and cross your fingers.)  Python
that would run if invoked from bash?  Use "type".  Python that
would run if invoked by csh?  Use "which".

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




From rdudfield at my-deja.com  Mon Jan 17 04:19:08 2000
From: rdudfield at my-deja.com (rdudfield at my-deja.com)
Date: Mon, 17 Jan 2000 09:19:08 GMT
Subject: Best way to work in debug code?
Message-ID: <85umq4$tlr$1@nnrp1.deja.com>

Hello,


Just wondering how you people write debug code in your functions
/methods?

I personally write a "if (DEBUG_LEVEL >= 5):" around the code.  However
this reduces readability.

To fix the readability problem I usually delete debug code all the
time.  Only having to rewrite it again ( and debug the debug code ).

I sometimes use cvs to find that old debug code, but that's time
consumming.




The way that would be perfect ( for me ) would be for the editor I use(
jed ) to be able to toggle on and off the viewing of debug code, at
different levels of course.  Then have the python interpreter skip over
the debug code, or optimise it out.  One way to do this would be to do
this:

  for x in range(0,100):
    doNonDebugstuff()
  debug 5:
    print len(listOfObjects)
    writeGraphOfObjectsToImage(listOfObjects,"graph.png")
  doMoreNonDebugStuff()

Where debug 5 means "if (DEBUG_LEVEL <= 5):".

You could then call the python interpreter with a debug flag or maybe
call a builtin function to change the debug level.

It would be good to be able to choose to debug at level 4 and below, or
just level 4.



Additions to this might include using strings as debug tags. Like:

debug "functionName":
  doDebugStuff()

Or both.

debug "functionName" 4:
  doDebugStuff()

Being able to tell the interpreter to run all debug code in a particular
function/method/object would be good too.


Rene.




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


From Windholtz at acm.org  Wed Jan 19 21:50:09 2000
From: Windholtz at acm.org (Mark Windholtz)
Date: Wed, 19 Jan 2000 21:50:09 -0500
Subject: [ANN] Cincy OO Users Group Meeting - XP !
Message-ID: <38867653_2@news2.one.net>

  * * *   M E E T I N G :     Tuesday 15 Feb 2000                   * * *
  * * *   Cincy Object Tech Users Group (COTUG) Meeting    * * *

  TOPIC:  eXtreme Programming
  ----------------------------------------
      Contents
  ----------------------------------------
  * Topic details
  * Agenda
  * Fee

  If you plan to attend the meeting please register for the eGroup
  and answer the POLL QUESTION: "Feb 2000 Attendance"

----------------------
   Topic details
----------------------
 The next two monthly meetings will be on eXtreme Programming (XP)

 15 Febuary 2000 (Tuesday)
       --> The 12 essential practices of XP.
       --> Refactoring: Comparing Java & the Refactoring Browser.

 14 March 2000 (Tuesday)
       --> Hands-On eXtreme Hour demo (with a load factor of 2.5 :-).
       --> You will experience XP through development iterations.

Sign onto the eGroup at:  http://w3.one.net/~objwind/st_users


----------------------------------------------------
   Agenda
----------------------------------------------------
       5:30 doors open
  6:00-6:30 Social (and Pizza for those who register)
  6:30-8:00 eXtreme Programming essential Practices
  8:00-9:00 Refactoring demos

  Directions
  ------------
  Map and directions on the homepage http://w3.one.net/~objwind/cinmaps.htm
  Note: the building is on the south side of Merchant st.
  Mapquest incorrectly places the star on the north side
  of the street.

  Fee
  -----------------------------
  Fee: $5  (covers the pizza)

  If you plan to attend the meeting please register for the eGroup
  and answer the poll question about "Feb 2000 Attendance"

 -Mark Windholtz
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 | Cincy Object Tech Users Group
 | http://w3.one.net/~objwind/st_users
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~







From mgushee at havenrock.com  Thu Jan  6 08:11:10 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 06 Jan 2000 08:11:10 -0500
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com> <38744512.377982379@news.isomedia.com>
Message-ID: <m3n1qj5or5.fsf@havenrock.com>

bitbucket at isomedia.com (Eugene Goodrich) writes:

> On Wed, 5 Jan 2000 23:52:46 -0500, "Darrell" <news at dorb.com> wrote:
> 
> >from filename import *
> >
> >Doesn't create another namespace. Avoid this in most cases.
> >
> 
> What are the implications of your statement for Tkinter?  Is that one
> of the exceptions?

Some packages, including Tkinter, are designed to be 'safe' to use
with 'from filename import *'. All that really means is that they are
careful not to export any unnecessary names, and that the names they
do export are unlikely to conflict with other commonly-used names.

E.g., when you do

from Tkinter import *

you have class names like Frame and Canvas in the global namespace --
which are unlikely to be found except in another GUI module.

However, I've seen code that does

import Tkinter

and qualifies all the names. Guess it depends what you're doing.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From thor at localhost.localdomain  Mon Jan 31 19:08:03 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 1 Feb 2000 00:08:03 GMT
Subject: MiniBloquiHeader
Message-ID: <slrn89ccg8.ilo.thor@localhost.localdomain>

For those who use LaTeX I've made a little utility that
uses python,pstricks and fancyhdr :

http://www.ctv.es/USERS/irmina/MiniBloquiHeader.html
-- 
MGA


From sameerc1 at yahoo.com  Thu Jan  6 04:12:20 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Thu, 6 Jan 2000 01:12:20 -0800
Subject: python with ASP
Message-ID: <003c01bf5826$23f7b0e0$db75fea9@desktop>

when I run it using PWS under win98, I get the following error:

Python ActiveX Scripting Engine error '80020009' 

invalid syntax 

/pytest.asp, line 3 


Mark Hammond wrote in message ...
>What exactly are you asking - "why doesnt this work?"? - what are the
>symptoms?  It should work...
>
>Mark.
>
>"Sameer Chowdhury" <sameerc1 at yahoo.com> wrote in message
>news:007c01bf57f7$eb3941c0$db75fea9 at desktop...
>> how could the following work.  It's supposed to either display a
>form or
>> display "Got Title".
>>
>> <%@ LANGUAGE = Python %>
>> <%
>> if REQUEST.Form("title"):
>> %>
>>  <html>
>>  <b> Got Title</b>
>>  </html>
>> <% else: %>
>>  <html>
>>  <FORM ACTION="/pytest.asp" method="POST">
>> <b>Title:&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</b><BR>
>>  <input type="text" value="" size=60 name="title"><P>
>>  <input type=submit value="Submit">
>>  </form>
>>  </html>
>>
>>
>> __________________________________________
>> NetZero - Defenders of the Free World
>> Get your FREE Internet Access and Email at
>> http://www.netzero.net/download/index.html
>>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000106/468a25bb/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pytest.asp
Type: text/asp
Size: 345 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000106/468a25bb/attachment.bin>

From echuck at usit.net  Wed Jan  5 20:21:07 2000
From: echuck at usit.net (Chuck Esterbrook)
Date: Wed, 05 Jan 2000 20:21:07 -0500
Subject: Who hosts CGI/Python?
Message-ID: <3873EE03.1CE5F8CC@usit.net>

Seems like many web host providers, such as MindSpring, offer CGI with
Perl, but not other languages, notably Python.

Does anyone know who would host a CGI Python site?

-Chuck




From phd at phd.russ.ru  Thu Jan 13 11:25:11 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 13 Jan 2000 16:25:11 +0000 (GMT)
Subject: Cookies
In-Reply-To: <I=p9OJtYnt3u+UxxIZ0WBCMmSX4=@4ax.com>
Message-ID: <Pine.LNX.4.21.0001131623300.18609-100000@fep132.fep.ru>

On Thu, 13 Jan 2000, Anders M Eriksson wrote:
> >> kaka = Cookie.Cookie(os.environ["HTTP_COOKIE"])
> 
> Then I get a KeyError: HTTP_COOKIE
> 
> Which probably means that the HTTP_COOKIE environment variable doesn't
> exist!?

   This should be caught by your try/except! You wrote universal except -
for all exceptions, no?

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From aahz at netcom.com  Mon Jan 17 14:58:48 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jan 2000 19:58:48 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com>
Message-ID: <85vs9o$sbh$1@nntp9.atl.mindspring.net>

In article <slrn886kml.284.grant at grante.comtrol.com>,
Grant Edwards <grant at nowhere.> wrote:
>
>I just found the http server page on Exchange server that allows access
>to e-mails via a web browser.  What I'd really like is to write a
>Python program that just sucks the mail out of the inbox and delivers
>it locally.

No can do unless you implement a JavaScript engine.  :-(  (Trust me, I'd
*love* to use Lynx for my mail.)
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From dworkin at ccs.neu.edu  Wed Jan 19 22:58:26 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 19 Jan 2000 22:58:26 -0500
Subject: a small problem
References: <20000112015923.A8450@xs4all.nl> <000701bf5cb4$c72320a0$da2d153f@tim> <20000112145107.B8450@xs4all.nl> <20000112165932.A2758@stopcontact.palga.uucp> <20000113130514.A21652@xs4all.nl> <20000113153612.A1808@stopcontact.palga.uucp> <864o8l$dv6$1@newshost.accu.uu.nl>
Message-ID: <vnd4sc92y3h.fsf@betelgeuse.ccs.neu.edu>

m.faassen at vet.uu.nl (Martijn Faassen) writes:

> I have scant knowledge of Perl (being Saved by Python just in time), but
> as far as I recall the $ @ stuff can be used to somehow make a particular
> variable behave like a number, list, or whatever. So I imagine it allows
> arbitrary arguments which you then may analyze and then treat in the way
> you desire. Or something. My mind never really wrapped around the concept.

You've pretty much got it, but it's even scarier than you may be thinking.

Arguments to a subroutine are generally not declared, but rather are
automatically assigned to the special variable @_ inside the scope of
the function.  Thus, the first argument is referenced by $_[0], the
second by $_[1], etc.  

This has nothing to do with the $_ variable, by the way, which has an
entirely different meaning.

reminded-why-I-program-with-Python-instead-of-magical-line-noise-ly y'rs,

-Justin

 



From soundhack at hotmail.com  Sat Jan 22 19:01:44 2000
From: soundhack at hotmail.com (Robert)
Date: Sat, 22 Jan 2000 19:01:44 -0500
Subject: Registering COM Servers
Message-ID: <388a4488$0$22470@senator-bedfellow.mit.edu>

I'm trying to register a python class as a COM server, and have followed the
instructions given
in the "Quick Start to Server side COM and python" However I'm finding
(using dcomcnfg on NT4.0) that my COM object
is being registered as a "remote server" as opposed to a local server.  Can
someone tell me what I need to do to get it registered as a local server?





From MHammond at skippinet.com.au  Tue Jan 18 12:49:32 2000
From: MHammond at skippinet.com.au (MHammond at skippinet.com.au)
Date: Tue, 18 Jan 2000 17:49:32 GMT
Subject: WIN32 : how to list the current processes by their executable na me?
References: <326536345498D311B3BC00105A39802A07457E@newsintern.dspace.de> <CxKdRDAwDEh4Ewzj@jessikat.demon.co.uk>
Message-ID: <862936$g33$1@nnrp1.deja.com>

In article <CxKdRDAwDEh4Ewzj at jessikat.demon.co.uk>,
  Robin Becker <robin at jessikat.demon.co.uk> wrote:

> this fails for me with win95.

Unfortunately, win32pdh only works on NT/2000.  I dont know how to do
the same thing on Win95/98...

Mark.


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


From bjorn at roguewave.com  Tue Jan 25 13:29:16 2000
From: bjorn at roguewave.com (bjorn)
Date: Tue, 25 Jan 2000 11:29:16 -0700
Subject: Multidimensional Dictionaries: Possible?
References: <slrn88ppsg.49t.w6uv@voyager.home.com> <oqpuuqx6s1.fsf@titan.progiciels-bpi.ca> <388DE028.ED24A3B1@nipltd.com>
Message-ID: <388DEB7B.4D25A94E@roguewave.com>

> What's the quickest way to do the equivalent of the following Perl in Python,
> again, using the above example:
>
> $dict{$key1}{$key2}++
>
> I can only think of the following, I hope there is a better way! :
>
> a =  dict[key1, key2]
> a = a + 1
> dict[key1, key2] = a

How about:

class Counter:
    def __init__(self, n):
        self.n = n

    def inc(self):
        self.n = self.n + 1
        return self.n

dict[key1, key2] = Counter(0)
dict[key1, key2].inc()

-- bjorn




From ngps at madcap.dyn.ml.org  Thu Jan 13 09:52:23 2000
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 13 Jan 2000 14:52:23 GMT
Subject: MySQLdb, blobs, and inserting
References: <85itgj$mlc$1@nnrp1.deja.com> <85k0fd$hv2$1@news1.xs4all.nl> <387de600.0@news.cyberway.com.sg>
Message-ID: <387de6a7.0@news.cyberway.com.sg>

According to Ng Pheng Siong <ngps at post1.com>:
>   c.execute('insert into items (story) values ("%s") % file.read())
>                                                ^^^^
>                                                Note this!

Doh! Need a single quote to close the string, of course:

  c.execute('insert into items (story) values ("%s")' % file.read())
                                                     

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



From wjk at wjk.mv.com  Mon Jan  3 22:15:23 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Mon, 03 Jan 2000 22:15:23 -0500
Subject: How to use intern()? - newbie
Message-ID: <387165CB.44DA3AD3@wjk.mv.com>

Hi:

I am Curious about this:

intern("ASTRING") -- puts that string internally in a table...
1. How do you access any string put away using intern().
2. How would you put this function to good use / or what is
    its purpose.
3. Python Quick Reference - says interned strings are "immortals".
    Does that mean for the duration of running program only?

Thanx



From matthewm at es.co.nz  Wed Jan  5 18:29:21 2000
From: matthewm at es.co.nz (Matthew Miller)
Date: Thu, 06 Jan 2000 12:29:21 +1300
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk>
Message-ID: <3873D3D1.7DFC4360@es.co.nz>

Hi,

I can only comment on Learning Python which I find succinct and pretty
well structured. Some lengthier eamples would have been nice but the
book's approach is to expose some of the functionality of alot of the
modules through short scripts. On the occassions when I got stuck with a
language feature, I blamed the book but ultimately had to point the
finger at myself. For example, when catch multiple exceptions I was
using:-

try:
    statements
except exception1, exception2:
    staements

where the correct syntax is:-

try:
    statements
except (exception1, exception2):
    staements

The point I'm making is that their was only 1 reference (and possibly 1
example) which hinted at how-to catch multiple exceptions and if you're
skipping through the book quickly, then it can lead to problems. I would
give it an 8/10 as a learning source. Whichever book you go for, make
sure to get The Python Pocket Reference - it's invaluable. Good luck.
Matthew.

Bala wrote:

> Hi I'm new to Python but not to programming. I currently have:
>         Programming Python (Nutshell Handbook)
>         by Mark Lutz
> But I find it a little too heavy for my liking.
>
> I'm now thinking of getting either
>         Python Essential Reference (OTHER NEW RIDERS)
>         by David M. Beazley, Guido Van Rossum
> or
>         Learning Python (Help for Programmers)
>         by Mark Lutz, David Ascher, Frank Willison (Editor)
> Which one of these two would you recommend?
>
> Thanks.
>
> -Bala



From guy-murphy at easynet.co.uk  Wed Jan  5 00:43:14 2000
From: guy-murphy at easynet.co.uk (Guy Murphy)
Date: Wed, 5 Jan 2000 05:43:14 -0000
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl>
Message-ID: <84ukj7$2kdu$1@quince.news.easynet.net>

Hi.

With reference to the articles comments...

"Finally, if a program's source is defined using <method>, <parameter>, and
<block> tags, then individual programmers can choose whatever superficial
appearance they want."

..One might consider the direction that Microsofts
HTAs/Scriptlets/Behaviours are taking, as a step in this direction.

Yes one might say that it's just another rehash of an IDL, but I believe
that its aim is slightly different...I'll avoid commenting further, as I'm
likely to move into vague touchy feely territory :)

I agree with the desirablitity of seperating logic from structure and
perhaps what we are looking at is the rather simple solution of providing
clear means to describe structure within the language. Perhaps the same
attention to detail in this area that is traditional given to logic and flow
description is necessary, especially in the area of interface defination.

No bold assertions being made on my part, as I'm a self confessed
lightweight wandering into this area from Web development.

Also as a sidenote while musing in this direction (and probably because of
my development background) I'd like to see better support event-like
behaviour in a language. Not entirely sure what, maybe something like...

WHEN this_thing > that_thing:
    do_something()

...<shrug> seems to me that respone to occurences (aka events :) is a pretty
major concern in a lot of areas, and is poorly addressed. To my mind it's
this sort of thing that would warrant the title VHLL, where a major area of
repetitive handling is addressed by the language.

I'll shut up now before I get way over my head :P

Cheers
    Guy

> He made two striking
> suggestions - being able to draw part of the logic, in the sidebar of the
> code, as it were, and having a sort of underlying representation of the
> code (he said xml, but that's what everyone is saying these days), so that
> a particular development tool could show the code in various helpful ways.





From cv601 at torfree.net  Tue Jan  4 17:38:30 2000
From: cv601 at torfree.net (William Park)
Date: Tue, 4 Jan 2000 22:38:30 GMT
Subject: Quick Python v.s. Learning Python
References: <387152EB.10DD0E17@Lextek.com>
Message-ID: <Fnu1K6.1nz.0.bloor@torfree.net>

Art Pollard (Pollarda at Lextek.com) wrote:
: I just purchased Programming Python.  As I am 
: totally new to Python, I learned from this book
: that I want to learn more.  Unfortunely, I don't 
: think this is the best book for a Python newbie.
: (Though from what I've seen, it makes a good 
: language reference.)

: So right now, I am looking for another learning
: book.  The two which popout are:

: Quick Python (In my local bookstore)
: Learning Python (Not in stock but obtainable)

: Does anyone have any preferences / recommendations???

: I have extensive C/C++ experience but would like a good
: book to help me break through the wall and reach that
: Ahhh Haaaa experience when the language becomes clear.
: (I still remember the Ahhhh Haaa from many years ago 
: when I figured out C's pointers... What a day. :-)

: -Art

I recommend "The Quick Python Book" --- the layout and the typesetting are 
better.  So is the content, but it's biased opinion.

William Park

-- 


From watchcity at my-deja.com  Tue Jan  4 20:50:01 2000
From: watchcity at my-deja.com (Warren)
Date: Wed, 05 Jan 2000 01:50:01 GMT
Subject: Newbie: Printing under Windoze/DOS
Message-ID: <84u804$u1l$1@nnrp1.deja.com>

I can figure out how to print under Linux, but how can I print out text in
the MS environment? I want to print out all the fonts available on a machine
as examples. The same question for tcl/tk as well, but python/Tkinter is my
priority.  Thanks in advance!  Warren


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


From randy_shaffer at my-deja.com  Sat Jan  1 16:40:47 2000
From: randy_shaffer at my-deja.com (randy_shaffer at my-deja.com)
Date: Sat, 01 Jan 2000 21:40:47 GMT
Subject: Python and GUI options
Message-ID: <84lreq$hib$1@nnrp1.deja.com>

Hello,

I am building a testing harness for the components in my company?s
softswitch product.  I chose Python for this project because the
language and the libraries seemed very well suited for such a task.
The application is coming together nicely and I believe my choice of
tools was a good one.

The application?s architecture has fleshed out into a kind of model-
view-controller scheme.  I have implemented the majority of the model
and controller and now am considering implementation options for the
view (gui).

The softswitch application lives on Unix, as does the testing harness
implementation at this time.  The developers who are my customers will
be using the test harness via their NT workstations, which means that
the test harness gui will be served up by a soft Xserver (Exceed).

I have experimented with tkinter and the PMW library.  Things seem to
operate a bit slow.  The other options/considerations I?ve pondered are:

Move the gui to the Win platform ? I?m not very proficient in MFC-land
and I?m worried.  Is there Python portability to Windows
problems/issues as well?

Can Viper help out here?  I get the feeling that it is not quite ready
for prime time.  Am I wrong?

What about JPython and the fabulous Java Swing library?  Am I asking
for trouble here?

What are your thoughts, please?

Randy Shaffer
Software Engineer
Vsys


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


From eddyw at lsl.co.uk  Mon Jan 17 06:59:05 2000
From: eddyw at lsl.co.uk (Edward Welbourne)
Date: Mon, 17 Jan 2000 11:59:05 GMT
Subject: [PSA MEMBERS] Re: [Types-sig] Suggestions for python 2
In-Reply-To: <200001170221.VAA10441@eric.cnri.reston.va.us>
References: <m129zpL-00006JC@vortigen.demon.co.uk>
            <200001170221.VAA10441@eric.cnri.reston.va.us>
Message-ID: <200001171159.LAA17397@lsls4p.lsl>

> I'm not a functional programmer, and Python is not a functional
> programming language; and as far as I'm concerned it never will be
> one.

and that's just fine with me.  The reason for my list of `things this
will empower' is: there are folk asking for stuff that would mangle
python so that they can have those things.  My belief is that we don't
need to mangle python to achieve what I have in mind - and it'll
incidentally give them those things, in so far as they insist on having
them, without the rest of us having to notice.  In particular, safe
tunnels (and keyword-only arguments) are just *too* nice to not ask for.
And far nicer than asking for built-in functional tools.

> (Practical comment: the syntax <generator> <name> [<bases>]: <suite>
> won't fly; you seem to want <generator> to be any expression, and an
> expression followed by a name is too fragile a construct to be
> comfortably parseable (even if it might be unambiguous).)

No problem: always ask for more than need, so can back-pedal later.  The
generator could happily be a simple name (even one which has previously
been the subject of an statement which says `I intend to use this as a
generator' if desired).  Likewise, bases-tuples could be as at present
(rather than arbitrary expression yielding tuple, as asked for).

> Maybe you'll make the case better in person
I'm looking forward to trying ... ;^)

See you next week,

	Eddy.



From tim_one at email.msn.com  Wed Jan 26 00:39:21 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 26 Jan 2000 00:39:21 -0500
Subject: Python startup seems slow... (NT)
In-Reply-To: <MPG.12f7f19a8fa3be15989680@or.news.verio.net>
Message-ID: <000201bf67bf$b27527a0$592d153f@tim>

[Jason Taylor]
> Under NT, it seems that starting up python to process even
> a trivial script is pretty slow.

Under all installations of NT, or some specific one that you tried?  Don't
laugh <wink>.  One thing to watch out for is fragmentation under NT.  Get a
defrag utility (e.g., Diskkeeper Lite is free) and check that out.

> With the following python program, helloworld.py:
>
>   print "Hello World"
>
> it takes a second or more for "python helloworld.py" to execute
> on my machine.  It doesn't get any faster the second or third
> time around.
>
> Curious, I used the same program with Perl.  It had basically
> no delay.

One way in which Python differs is that part of Python is written in Python,
and so Python does more poking around on the disk (looking for Python files)
than Perl does at startup.  This involves filesystem stat calls, which
aren't speedy under Windows.  Alas, the people who know the most about this
are at the Python Conference now, so your timing is off.

After defragging your disks, a couple things to check out:

1. Python looks for and executes site.py at startup.  You can
   disable that by passing the -S switch.  Depending on what's
   in site.py, running it can consume an arbitrary amount of
   time.  Specifying -S saves significant startup time on my
   old P5-166 Win95 box -- which seems nevertheless faster
   than your NT box!

2. Python looks for and loads exception class definitions from
   exceptions.py.  You can disable that by passing the -X
   switch.  This is a Bad Idea (the -X switch will go away
   somdeday), but timing the difference with & without will
   at least shed some light.

Doing both of those speeds startup a lot for me -- but Perl startup is still
obviously quicker on my machine.

One other thing:  trim sys.path to the minimum.  This won't help startup,
but can save oodles once the app gets going.

And one final thing:  if python.exe is resolved later in your PATH than
perl.exe, it takes Windows that much longer to find it to begin with
(Windows isn't much quicker at finding things than Python is <wink>).
Fastest under Windows is to specify a full path (including the trailing
".exe").

the-death-of-a-thousand-cuts-ly y'rs  - tim





From skip at mojam.com  Mon Jan 24 12:27:43 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 24 Jan 2000 11:27:43 -0600 (CST)
Subject: OFFTOPIC: *-ly y'rs, 
In-Reply-To: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com>
References: <pcobt6c635a.fsf@math.ntnu.no>
	<Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com>
Message-ID: <14476.35727.977056.378091@beluga.mojam.com>

    derek> I have only ever seen these sign-offs on this list:

    derek> *-ly y'rs

    derek> Where did they come from and are they patented :) How can one
    derek> join the signoff-ly club?

Ask the timbot.  It knows.

    derek> these-are-super-fun-ly y'rs,

welcome-to-the-club-ly y'rs,

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From cheselka at tucana.noao.edu  Tue Jan 11 13:33:04 2000
From: cheselka at tucana.noao.edu (Matt Cheselka)
Date: 11 Jan 2000 18:33:04 GMT
Subject: ioctl
Message-ID: <85ft10$1kdg$1@noao.edu>

Hello all again,

I'm now attempting to use ioctl to talk to a piece of hardware that I'm working with and have again gotten stuck.  Here's how I'm doing it:

	>>> import fcntl
	>>> fd = open ("/dev/random", "rb")	// the random number generator
	>>> RNDGETENTCNT = -2147200512		// get entropy count
	>>> val = fcntl.ioctl (fd.fileno(), RNDGETENTCNT, xxx)
	Traceback (innermost last):
	  File "<stdin>", line 1, in ?
	NameError: xxx

so I try this:

	>>> xxx = 0
	>>> val = fcntl.ioctl (fd.fileno(), RNDGETENTCNT, xxx)
	Traceback (innermost last):
	  File "<stdin>", line 1, in ?
	IOError: (14, 'Bad address')

So it appears as though I'm not passing the correct value of 'xxx' since it
looks like fcntl.ioctl is interpreting 'xxx=0' to be the address of where to
place the result, which is in the ROM and not legal.

So I've got two questions:

	1. is this essientially the correct way to do ioctl stuff or is there
	   a better way?
	2. how do I pass this method the correct address?

TIA,

Matt Cheselka
cheselka at noao.edu


From lvirden at cas.org  Fri Jan 21 06:25:58 2000
From: lvirden at cas.org (lvirden at cas.org)
Date: 21 Jan 2000 11:25:58 GMT
Subject: Tkinter coverage in books? [Was: Re: Python and Tkinter Programming by John E. Grayson
References: <3880792E.79A874B6@weihenstephan.org>
Message-ID: <869fo6$jcs$1@srv38.cas.org>


Other than Programming Python and this new reference, what other books
provide the Python/Tkinter programmer substantial information?  That is
to say, does more than a few sentence mention that 'oh, by the way, there's
this extension that a number of Pythonians use...' 
-- 
<URL: mailto:lvirden at cas.org> <URL: http://www.usenix.org/events/tcl2k/>
<*> O- <URL: http://www.purl.org/NET/lvirden/>  Tcl2K - Austin, Texas, US
Unless explicitly stated to the contrary, nothing in this posting
should be construed as representing my employer's opinions.


From grant at nowhere.  Thu Jan 27 13:52:35 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 27 Jan 2000 18:52:35 GMT
Subject: Python script monitoring another process
References: <389089E8.32BB7710@cs.brandeis.edu>
Message-ID: <slrn8914vi.hc8.grant@grante.comtrol.com>

In article <389089E8.32BB7710 at cs.brandeis.edu>, Magnus Bjornsson wrote:

>I'm new to Python so this is probably a typical newbie question...
>Anyway, I'm trying to build a Python script which does the following:
> 1. starts a process (that's easy)
> 2. checks the output from the process (from its stdout)
> 3. when a certain token is found in the process' output the script 
>    sends a new token to the process' stdin

The popen2 module lets you do this.

-- 
Grant Edwards                   grante             Yow!  Edwin Meese made me
                                  at               wear CORDOVANS!!
                               visi.com            


From JamesL at Lugoj.Com  Tue Jan 25 00:28:08 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Mon, 24 Jan 2000 21:28:08 -0800
Subject: tabs do WHAT?
References: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net>
Message-ID: <388D3468.7083A347@Lugoj.Com>

Warren Postma wrote:
[A whole lot of misinformed patronizing and opinionating elided.]
> Even BASIC, ugly as it is from a functionality point of view, is miles ahead
> from the point of view of useability by people who are skilled technical
> people, yet who are not "programmers".

You don't know ugly. And what you think is BASIC really isn't. Python
indentation requirement is one of the two most beautiful and satisfying
aspects of the language.

Been programming for 25+ years. Wrote the following BASIC program to display
lines of electric potential on a home-brew display device on a "fast" 4Mhz
Cromemco Z-80 system with 16K byte ram using TDL BASIC. At least I think
that is correct. It was written back when I was studying physics and did not
consider myself a "programmer". Now tell me if you can figure out the block
structure (I had to do this all the time, since most BASIC interpreters
would not preserve indentation.):

10 DEF FNED(K)=(K<0) OR (255<K)
20 INPUT "# LINES";S
30 PI=3.1415926535*2
40 SS=PI/S
45 REM CLEAR DISPLAY
50 OUT 0,2 : OUT 0,1
60 INPUT"# CHARGES";N
70 DIM QM(N),X(N),Y(N)
80 FOR I=1 TO N
90 PRINT I;
100 INPUT QM(I),X(I),Y(I)
110 IF FNED(X(I)) OR FNED(Y(I)) THEN 90
115 REM DRAW POINT AT CHARGE LOCATION X(I),Y(I)
120 OUT 2,Y(I) : OUT 1,X(I)
130 NEXT I
140 FOR I=1 TO N
150 A=0
160 FOR C=1 TO S
170 A=A+SS
180 X=X(I) : Y=Y(I)
190 X=X+COS(A)
200 Y=Y+SIN(A)
210 IF FNED(X) OR FNED(Y) THEN 390
215 REM DRAW NEXT LINE POINT AT X,Y
220 OUT 2,Y : OUT 1,X
230 XS=0 : YS=0
240 FOR J=1 TO N
250 XD=X-X(J)
260 YD=Y-Y(J)
270 R1=XD*XD+YD*YD
280 IF R1<.99 THEN 390
290 R1=SQR(R1)
300 K=QM(I)*QM(J)/(R1*R1*R1)
310 XS=XS+K*XD
320 YS=YS+K*YD
330 NEXT J
340 R=SQR(XS*XS+YS*YS)
350 IF R=0 THEN 390
360 X=X+XS/R
370 Y=Y+YS/R
380 GOTO 210
390 NEXT C
400 NEXT I
410 END


From m.faassen at vet.uu.nl  Thu Jan 20 21:34:30 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 21 Jan 2000 02:34:30 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com>
Message-ID: <868gjm$nmq$1@newshost.accu.uu.nl>

tye4 <tye4 at yahoo.com> wrote:
[snip Tim]
> I don't oppose indentation either. It's good for program readability, and
> readability only. It shouldn't be part of the syntax.

But it is -- see below. In just about any language.

Why should it be only for readability? Since programming for readability
is good programming practice, and since it's easy to do without block
delimiters, why not simply take the next step and get rid of this redundancy?

Taking away the whitespace from the redundancy doesn't work -- code becomes
unreadable. So the delimiters were taken away.

> Whitespace end
> delimiters are invisible and hence confusing and prone to error.

                    <---- Lots of invisible whitespace
          ^
          |--- can you see it? :)

Indentation is easier to see than delimiters. It's easy to prove; take away
the indentation and see if you can see which delimiters match which. Now
take away the delimiters and leave the indentation. You know where to put
the delimiters again -- it's usually instantly obvious. If this was not the
case, nobody would use indentation in their code at all.

> I've used some stupid Unix shell languages which fail if I add whitespace:

> Without spaces (this works):
>     $x=10
>     $y=2+2

> With spaces (doesn't work):
>     $x  =  10
>     $y  = 2 + 2

Python is not a unix shell language. That's like saying you've used Lisp and
balancing the braces was difficult so now you think any use of braces 
in programming languages is evil. :)

> I thought they gotten rid of all such languages until I came across Python.
> Good programming languages should be whitespace tolerant, and most of them
> are.
> (C++, Java, Ada, BASIC, HTML, Eiffel, Perl, Pascal, ML etc... really long
> list.)

Why? Why should good programming languages be ignoring whitespace? Many of
these don't, don't, actually -- they use whitespace to split up things into
keywords and variable names and such: <imgsrc="foo"> is not going to work.

Not that HTML is a programming language.

> Why is my question about whitespace raise so many eyebrows? This feature is
> commonplace.

It is because many Python users like this use of whitespace. Just because
most other languages do it differently doesn't mean it is better; if people
reasoned like that we'd still be using lots of balancing braces in every
language, and not just in Lisp (it being one of the first high level
languages). 

Anyway, not your question about whitespace is raising the eyebrows; it's
your sometimes rather inflammatory presentation of this question. You called
a feature of Python that most of us like a lot and think essential to Python
_stupid_, for instance. Generally this is not a good approach to newsgroups.

Regards,

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


From ivanlan at callware.com  Wed Jan 19 19:33:00 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 19 Jan 2000 17:33:00 -0700
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu> <865ij9$ia4$1@newshost.accu.uu.nl> <38865072.DBD8B4E5@callware.com> <865k4o$iaa$1@newshost.accu.uu.nl>
Message-ID: <388657BC.49FD9EE9@callware.com>

Hi All--

Martijn Faassen wrote:
> 
[snip]
> > Therefore, I plan to redesign
> > Python so that you will all have to read and write Mayan glyphs in order
> > to write any programs.
> 
> Sounds great! Should it be part of Numpy or the core language?
> This is a math related extension, right?
> 

No, no, it *is* the core language.  No more def.  Instead, you will have
to use the glyph for God K.  You will have to write your variable names
in the Mayan syllabary.  All numbers will be modified base 20 by default
(if you prefix a number with the "elephant" glyph, however, you will be
able to use Roman numbers).  No more reserved words, only reserved
glyphs.

Instead of colons at the end of lines, you will have to use the
"spectacle" glyph (properly oriented, of course).

> Though-dutch-glyphs-would've-been-better-ly yours,
> 

Possibly, but you said that you were leaving the redesign of Python to
those who weren't attending IPC8.  You'll have to wait till next year
for your opportunity to apply a little Dutch Cleanser to Python'-)

<asta-la-vista-bebe>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From s.schwarzer at ndh.net  Thu Jan  6 11:50:01 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 06 Jan 2000 17:50:01 +0100
Subject: bdb: which class?
References: <B9AA1D5DA95ED211900100805F6FFF9C02F3D8F6@exbhbhmb01.us.dbisna.com>
Message-ID: <3874C7B9.106A5730@ndh.net>

Hello,

"Zakon, Stuart" schrieb:
> I would like to have the class name for methods of classes which are being
> called but I didn't see it in the documentation for code objects.
> Anybody know if this can be found?

If I understand your question: What about

class where_am_i:
    def name( self ):
        'return the name of the class to which self belongs'
        return self.__class__.__name__

? This should also work, if the object identified by self belongs to
a class Derived, i. e. 'Derived' should be returned.

Stefan


From skip at mojam.com  Tue Jan 25 22:05:58 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 25 Jan 2000 21:05:58 -0600 (CST)
Subject: Python startup seems slow... (NT)
In-Reply-To: <MPG.12f7f19a8fa3be15989680@or.news.verio.net>
References: <MPG.12f7f19a8fa3be15989680@or.news.verio.net>
Message-ID: <14478.25750.656580.780253@beluga.mojam.com>

    Jason> Under NT, it seems that starting up python to process even a
    Jason> trivial script is pretty slow.  With the following python
    Jason> program, helloworld.py:

    Jason>   print "Hello World"

    Jason> it takes a second or more for "python helloworld.py" to execute
    Jason> on my machine.  It doesn't get any faster the second or third
    Jason> time around.

I rather suspect you have a problem of some sort.  On my Linux system I time
the following command at under 0.05s elapsed time:

    python -c "print 'hello world'"

run from the command line.

    Jason> Is there anything I can do to improve the performance?  Is this a
    Jason> known problem?

Not a known problem that I'm aware of.  If simple scripts like that with no
imports took a second to run in general on NT, I'm sure we'd have heard
about it by now.  (Not criticizing, just suggesting you turn over a few
rocks on your system.)  I seem to recall that NT has a pretty nice system
resource monitoring tool.  Are you running your script on a completely
quiescent system?

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From scarblac-spamtrap at pino.selwerd.nl  Sun Jan 30 14:43:06 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 30 Jan 2000 19:43:06 GMT
Subject: Choosing a language
References: <01bf6b45$ff345b40$412b69ce@default>
Message-ID: <slrn8994sf.ndn.scarblac-spamtrap@flits104-37.flits.rug.nl>

Patrick Cooper wrote in comp.lang.python:
> I want to start learning computer programming and from what I have found so
> far I have chosen to start with Python.  I am going to be focusing on CGI
> first but eventually want to work with desktop applicates.  Python seems to
> be the most versitle and easiest to learn.  Before I go in any more depth
> does anyone think I should start with a different language?  I want to
> eventually learn C, but think Python is the best place to start.

It's what I usually recommend to people. It's easy to learn, and
it can be used for a *lot* of things, including CGI. Other languages
are usually more specialistic.

There are two complaints about Python as a first language that I've heard:
1) Basically all the documentation is in English. Not a problem for you.
2) It may spoil you. Other languages will be *ugly* compared to Python.

Besides, this is one of the best newsgroups about a programming language
on Usenet. It's actually useful :-)

-- 
Remco Gerlich,  scarblac at pino.selwerd.nl

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From tim_one at email.msn.com  Wed Jan 19 00:28:58 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 19 Jan 2000 00:28:58 -0500
Subject: Python Rocks!
In-Reply-To: <862jqd$l9s@news.or.intel.com>
Message-ID: <000401bf623e$16e712c0$9a2d153f@tim>

[tye4 and Tom Culliton bump heads over whitespace -- who
 could have imagined that the absence of something could
 leave such a dent <wink>?
]

[tye4]
> Are you insinuating that this is not a problem since I'm
> not as experienced as Guido?  I don't need to design a
> language to detect flaws in it.

Indeed not, but it would help your case had you *used* Python for quite a
while before asserting your position so absolutely.  Whitespace flames are
historically common on c.l.py, and for every one who comes along insisting:

> This problem needs to be fixed.

at the start, there are many more who change their minds a few months later.

A long time ago I wrote a program to tabulate the distribution of function
sizes in the std Python library (a great deal of which is written in
Python); the median (not counting blank lines and comments) was six(!)
lines.  Python is a very high-level language, and when functions don't
slobber over pages (as they too often do in, say, C or Java), indentation is
dandy for denoting block structure.  I would hate to try maintaining C
without curlies (or Pascal without "end"s, or ...), but well-written Python
is a joy without the noise.

> That's because indentation is highly dependent on what
> text editor you use. Some editors convert tabs to spaces
> some don't. Only way around that would be not use tabs...

You will find comfortable ways to live with it, provided only you're not
determined to let this stop you <0.7 wink>).  From Notepad to Emacs, it's
just not a real problem in practice.

> ...
> Unfortunately, it's hard to change basic syntax of a language.

At this point, flatly impossible.  Guido's last word on the subject can be
found in Tools/Scripts/pindent.py.

often-mentioned-never-used<wink>-ly y'rs  - tim





From gerrit.holl at pobox.com  Tue Jan  4 14:11:08 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 4 Jan 2000 20:11:08 +0100
Subject: Considering using python for something, and need a bit of info :)
In-Reply-To: <0.3903b60a.25a2245f@aol.com>; from PartyG2816@aol.com on Mon, Jan 03, 2000 at 11:12:15AM -0500
References: <0.3903b60a.25a2245f@aol.com>
Message-ID: <20000104201108.A1842@stopcontact.palga.uucp>

PartyG2816 at aol.com wrote on 946894335:
> Allo list,
>     Another concern of mine is: how secure can it be? IE can i restrict 
> certain aspects of it based upon a condition?

Yes, you can use the rexec.py module for that.

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From bitbucket at isomedia.com  Sat Jan  8 01:47:25 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Sat, 08 Jan 2000 06:47:25 GMT
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <852jbk$9tt$1@nntp9.atl.mindspring.net> <3875099A.2CF9333C@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net> <3876B79F.30439283@mdlink.de>
Message-ID: <3876dcee.547930242@news.isomedia.com>

On Sat, 08 Jan 2000 05:05:51 +0100, Helge Hess <helge.hess at mdlink.de>
wrote:

>Aahz Maruch wrote:
>> 
>> Not directly, no.  I think it breaks compatibility with Python's
>> extremely simple object model and I would be surprised if Guido
>> disagrees with me.
>
>Hm, I wonder why it is simpler to have two separate kinds of methods
>instead of a consistent behaviour for all methods ?
>

I suppose if it bothers you that much, you could just stop using
__init__. :)

Seriously, when I was first learning Python, I didn't know about any
of the "magic" functions, so I lived without __init__.  As far as I
knew, all functions really were alike.  I side with Aahz on keeping it
simple enough to sucker, uh, induct, no, uh, attract new users to
Python.

     -Eugene
import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From aahz at netcom.com  Fri Jan 28 10:43:47 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 28 Jan 2000 15:43:47 GMT
Subject: html parsing (htmllib)
References: <280120000015041211%zippy@cs.brandeis.edu>
Message-ID: <86sdfj$nsa$1@nntp9.atl.mindspring.net>

In article <280120000015041211%zippy at cs.brandeis.edu>,
Patrick Tufts  <zippy at cs.brandeis.edu> wrote:
>
>I want to grab the text from a set of web pages.  htmllib seems like
>what I want, except that want to assign the parsed text to a variable
>instead of printing it to stdout or a file.

Use cStringIO.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From dworkin at ccs.neu.edu  Sat Jan 29 21:54:37 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 29 Jan 2000 21:54:37 -0500
Subject: Creating a local variable with a parameterized name
In-Reply-To: Nic Williams's message of "Sun, 30 Jan 2000 01:43:20 GMT"
References: <38937075.18B75753@rochester.rr.com> <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu> <8704vo$soj$1@nnrp1.deja.com>
Message-ID: <vndaelomfpu.fsf@betelgeuse.ccs.neu.edu>

Nic Williams <nic at csee.uq.edu.au> writes:

> Michael if you really need to insert the variable into the local name
> space then be aware of the built-in functions global() and locals()
> which return dictionaries holding the global and local symbol tables,
> respectively. There is also vars([object]) which gives an object's
> symbol table or (when no parameter is given) the local symbol table.
> 
> The result isn't much more attractive than Justin's but you do insert
> the variable into the 'official' local space:
> 
> def f(name):
> 	locals()[name]=123
> 	print locals()[name]

This is why I didn't recommend that approach, from the Python documentation:

locals () 
       Return a dictionary representing the current local symbol table. 
       Warning: the contents of this dictionary should not be modified; 
       changes may not affect the values of local variables used by 
       the interpreter. 

Yes, I know that it works right now, but it is also quite clearly
marked as something which could stop working without warning in a
future version of Python.

If you must recommend something like that, it is only polite to point
out that it is strictly incorrect code and should not be relied upon.

-Justin

 



From bedatse at hknet.com  Sat Jan  8 22:29:52 2000
From: bedatse at hknet.com (Beda Tse)
Date: Sun, 9 Jan 2000 11:29:52 +0800
Subject: Hope this is the mailing list
Message-ID: <005601bf5a52$1a37dcf0$0000fea9@wyvern>

Please help...
how i can extract this four variable
gecko
jing at kirk.com
toad
ray at lex.com
from this two
"gecko" <jing at kirk.com>
"toad" <ray at lex.com>
by using regular expression????
I want to remove the bracket and quote sign
please help me and reply to bedatse at hknet.com
Thanks all !!




From rupole at compaq.net  Tue Jan 18 00:50:32 2000
From: rupole at compaq.net (Roger Upole)
Date: Tue, 18 Jan 2000 05:50:32 GMT
Subject: WIN32 : how to list the current processes by their executable name?
References: <85vur9$1prj$1@news5.isdnet.net>
Message-ID: <IaTg4.7699$kv.517394@news.easynews.com>

There is a win32pdhutil module in the win32 library, with a function
FindPerformanceAttributesByName.
Given an executable name, it will return a list of matching pids, which are
then at your mercy.

"Yann SAMAMA" <ysamama at cybercable.fr> wrote in message
news:85vur9$1prj$1 at news5.isdnet.net...
> Hi everybody!!
>
> Is there a simple way to implement an os.kill() that works on all WIN32
> platforms?
> I know about the FAQ article, but I want the kill function to take the
name
> of the executable file as parameter, not the Process ID.
>
> It should work like this :
>
>
> def kill(exe_file):
> # assumes we're on a Win 32 platform
>     import win32api
>
>     mylist = getcurrentprocesses(exe_file)  # that's the key function, it
> retrieves a list of corresponding PID
>
> # if exec_file is "myprog.exe", we kill all instances of this program that
> are actually running...
>
>     for pid in mylist:
>         handle = win32api.OpenProcess(1, 0, pid)   # that's the FAQ
> implementation of kill function
>         win32api.TerminateProcess(handle, 0)
>
>
> Any idea about how writing this function "getcurrentprocesses()" ?
>
>
> I've heard about a POSIX port of "kill" on NT that accepts a file name
> with -f option, but I'm afraid I won't be allowed to install it on my NT
> platform...
>
> Thanks in advance,
>
> Yann.
>
>




From bwarsaw at python.org  Wed Jan 26 13:55:45 2000
From: bwarsaw at python.org (Barry Warsaw)
Date: Wed, 26 Jan 2000 13:55:45 -0500 (EST)
Subject: Announcing JPython 1.1 final
Message-ID: <14479.17201.455555.755125@anthem.cnri.reston.va.us>

Greetings from the Python conference!

I'm pleased to announce the release of JPython 1.1 final, the 100%
Pure Java implementation of the Python language.

Below is a summary of recent changes.  See

    http://www.jpython.org/NEWS

for more details on what's changed.  For more information on JPython
in general and for links to the downloads, please see

    http://www.jpython.org/

Enjoy,
-Barry

<P><A HREF="http://www.jpython.org">JPython 1.1 final</A> - a 100%
Pure Java implementation of the Python language.  (26-Jan-2000)

-------------------- snip snip --------------------
26-Jan-2000 JPython 1.1 final released

  - Only one small CPython compatibility patch to os.mkdir() and
    os.makedirs() -- optional `mode' argument, which is ignored.

21-Jan-2000 JPython 1.1 release candidate 1 released

  New JPython features and bug fixes
    - Race conditions when creating multiple PythonInterpreter objects
      in an embedded multithreaded Java application have been fixed.
    - Recursive print of a module's namespace no longer overflows the
      JVM stack.
    - Fixed backslashing of quotes inside triple quoted strings (Finn
      Bock).
    - Fixes to read(int) method on file objects, which used to return
      the empty string before EOF was seen.
    - Class-based exceptions are used in frozen (jpythonc) modules.
    - Duplicate argument names in a function definition raise the
      appropriate SyntaxError.
    - os.path.getsize() has been added.
    - mutability of tuples under some situations has been fixed.
    - dictionary insertion order affecting dictionary equality has
      been fixed (Finn Bock).
    - Using the --package option with jpythonc has been fixed.

  Improved CPython 1.5.2 compatibility
    - optional sizehint argument implemented for readlines() method on
      file objects.
    - Better compatibility in %g formatting of floats.
    - sys.stdout.softspace is now writeable.
    - re.VERBOSE/re.X flags are implemented.
    - traceback object's reprs are CPython conformant.
    - In module os, os.error is equivalent to the builtin OSError.
      Also, listdir(), mkdir(), makedirs(), remove(), rename(), and
      rmdir() raise OSError directly.
    - os.stat() raises an OSError if the file is missing.



From wjk at wjk.mv.com  Sat Jan 22 15:30:02 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Sat, 22 Jan 2000 15:30:02 -0500
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <86a46g$9c3$1@nnrp1.deja.com>
Message-ID: <388A1349.7C730718@wjk.mv.com>

Reminds me of Benjamin Franklin's famous quote:

"Thepenismightierthanthesword" .... white space becomes
kinda important... don't ya think!

Justus Pendleton wrote:

> > I thought they gotten rid of all such languages until I came across
> Python.
> > Good programming languages should be whitespace tolerant, and most of
> them
> > are.
>
> Haveyouevertriedtore adEng lishthatdoesn 'tuse whitespaceproperly?
>
> Most human languages aren't really whitespace tolerant (well, at least
> the ones I know :-).  When I was learning Japanese, one of the hardest
> things about reading it was the fact that they don't use whitespace to
> delineate words.
>
> How come now one is out there advocating that we "fix" English by
> removing the mandated whitespace?  :')
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From Christian.Eibl-ObjectVision at kst.siemens.de  Mon Jan 31 08:06:21 2000
From: Christian.Eibl-ObjectVision at kst.siemens.de (Eibl Christian Object Vision)
Date: Mon, 31 Jan 2000 14:06:21 +0100
Subject: performance question when using embedding/extending
Message-ID: <1237B133D9B4D311947400005A40FDD4639E3D@knk51.kst.siemens.de>

hi,

i need to call a python function from my c++ app and this python function in
turn
accesses an instanciated c++ object (i pass the pointer to the object as a
parameter).
in the python function i have approx. 20-30 calls to the object (simple
member accessing).

the goal:
provide a scripting interface for computation of certain parameters
depending on the
state of a changing object...

when i'm doing a simple performance test (just calling the function 100.000
times in c++)
it turns out that on my 300MHZ PII the function is called 32.000 times per
second if
i don't access any attributes via python and it goes down to 900 times per
second if
i place 30 access calls in the python function.

i "quantified" the application an see that almost all the cpu-time is used
in "PyEval_CallObject".
the application i'm building needs at least 10-times the performance it has
now...

my question:
 - is this just the (slow) way it is ?
 - is there a faster way to use embedding/extending (details see below) ?
 - is python not suitable for this task (the old system was also an
interpreter, but specifically
    written for this task) ?


the details::

i'm using the following code in my app to call a python function in a
module:

<snip>
 _pythonCaller_p->runWithObject( "i", &finalResult, somePtr_p,
"_SomeClass_p" )
</snip>

the "runWithObject" method is defined as follows:
NOTE: i'm using SWIG with shadow classes to generate python interfaces for
my C++ classes!

<snip>
bool PythonCaller::runWithObject( char* resultFormat, void* result_p, void*
pyObj_p, char* pyType)
{      
      PyObject* func_p = 0;
      PyObject* args_p = 0;
      PyObject* resu_p = 0;
      if( _doReload ) {
         func_p = loadAttribute( _moduleName, _functionName );
      }
      else {
         func_p = _pyFunction;
      }
      if( func_p==0 ) {
         return false;
      }
      char aTmp[256];
      SWIG_MakePtr( aTmp, pyObj_p, pyType );
      args_p = Py_BuildValue( "(s)", aTmp );
      if( args_p==0 ) {
         return false;
      }
      resu_p = PyEval_CallObject( func_p, args_p );
      if( resu_p==0 ) PyErr_Print();
      if( _doReload ) {
         Py_DECREF(func_p);
      }
      Py_DECREF(args_p);
</snip>

the python function called with this code is defined as follows:

import Mc
import Mcc

def calcAttrFinalCoding(mp) :
   mpi = Mc.MailpiecePtr( mp )
   mpc = mpi.fetch()

   if mpc.getPbrInfoStatus() == Mcc.barcodeOk :
      return Mcc.attrCodBc
   if mpi.getBvInfoStatus() == Mcc.barcodeOk :
      return Mcc.attrCodBc
   if mpi.getPbrInfoStatus()  == Mcc.barcodeOk :
      return Mcc.attrCodBc
   if mpi.getBvInfoStatus() == Mcc.barcodeOk :
      return Mcc.attrCodBc
   if mpi.getPbrInfoStatus()  == Mcc.barcodeOk :
      return Mcc.attrCodBc
   return 1

thanx for any help!
Chris

--
Christian Eibl
 Object Vision Software AG, M?nchen
 Christian.Eibl at ovsag.de | www.ovsag.de
 mobile: +49.171.7623844 or +43.676.3060031

 pgp-fingerprint: D681 0321 30BE FA99 AD03
                  FBE7 38C5 E788 2846 25DB




From composzioni at tiscalinet.it  Mon Jan 17 01:05:37 2000
From: composzioni at tiscalinet.it (Giovanni Riccardi)
Date: Mon, 17 Jan 2000 07:05:37 +0100
Subject: Bugs in IDLE environment (??)
Message-ID: <85uc1m$5i8$1@pegasus.tiscalinet.it>

Just a repost of the message, because I had no
replies within a week.
------
I wrote a small script that uses input() to get
the user input,get the values, make some calculation
and return calculated value.
The problem is that it works when the script is executed directly by the
interpreter, but not in IDLE environment where after the user has entered
some values the script doesn't return to the main loop.
For example:

print "Enter the name:"
x = input("--> ")
print "Enter the age:"
y = input("--> ")

doesn't continue after the first input().

Some help??
thanx in advance
Giovanni


--
-------------------------------------------
GIOVANNI RICCARDI * TERRACINA (LT) * ITALY

EMAIL: composizioni at tiscalinet.it
  ICQ: 40670340
-------------------------------------------

*******************************************
TAKE A LOOK TO MY OWN PROJECTS ON THE WEB
*******************************************
*         INFORM ITALIAN WEB SITE         *
*(a design system for Interactive Fiction)*
*                                         *
*        http://inform.freeweb.org        *
*******************************************
*******************************************
*         c o m p o s i z i o n i         *
*        digital  art & technology        *
*                                         *
*     http://composizioni.freeweb.org     *
*******************************************







From hei at adtranzsig.de  Tue Jan 18 05:37:29 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 18 Jan 2000 11:37:29 +0100
Subject: Call Python functions from C[++] as macros
References: <38837344@news.nwlink.com>
Message-ID: <948195173.979768@news.adtranzsig.de>

You want to do "embedding and extending":


http://www.python.org/doc/current/ext/ext.html

That means, your app will have to embed a Python
interpreter, and you will want to extend the Python
by new modules that control your C++ code.

I have a ~100K ZIP that patches the Python source code
so it will become BCB3-compilable, i think the zip
contains a small demo app.

If you want it, drop me a line. I hesitate to bomb your mailbox
without your permission.

When you're doing something financial (you mention
"business rules"), this group may be of interest to you:

http://www.egroups.com/list/python-finance/info.html

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


Glen Parker schrieb in Nachricht <38837344 at news.nwlink.com>...
>Hi all,
>
>Just getting started with Python (just downloaded the Win32 version last
>night :-).  I need to figure out how to get some reasonably advanced things
>working.  I'm writing a large app in C++ (BCB4.0) and I want to use Python
>as an event handler for executing business rules etc.
[...]




From mlh at vier.idi.ntnu.no  Mon Jan 10 17:21:14 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 10 Jan 2000 23:21:14 +0100
Subject: JPython serialization problem
References: <387989FF.658A8F68@xrce.xerox.com>
Message-ID: <iac66x1r2jp.fsf@vier.idi.ntnu.no>

Emmanuel Pietriga <emmanuel.pietriga at xrce.xerox.com> writes:

> 
> At first I thought the paths were defined incorrectly, but if I create
> an instance of the same class directly on the python side using its
> constructor, it works!  Therefore JPython knows where to find the class
> definition.

Have you tried importing the class before deserializing the object?
(Just a hunch - may be useless...)

> 
> Anyone can help?
> 
> Thanks.
> 

--

  Magnus
  Lie
  Hetland                                mailto:<anything>@hetland.org


From mlh at vier.idi.ntnu.no  Tue Jan  4 18:14:34 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 05 Jan 2000 00:14:34 +0100
Subject: Simple cartesian product
References: <iacpuvhqwdp.fsf@vier.idi.ntnu.no> <iacn1qlqvw7.fsf@vier.idi.ntnu.no> <iack8lpqvoi.fsf@vier.idi.ntnu.no>
Message-ID: <iachfgtqvj9.fsf@vier.idi.ntnu.no>

mlh at vier.idi.ntnu.no (Magnus L. Hetland) writes:

> mlh at vier.idi.ntnu.no (Magnus L. Hetland) writes:
> 
> > def product(a,b):
> >     a = reduce(lambda l, x: l + [x]*len(b),
> >                [[]] + a)
> >     b = b * len(a)
> >     return map(None, a, b)
> > 
> 
> Or, rather:
> 
> def product(a,b):
>     a2 = reduce(lambda l, x: l + [x]*len(b),
>                [[]] + a)
>     b  = b * len(a)
>     return map(None, a2, b)
> 
> (This is getting worse by the minute ;)

Sorry everybody... (I guess I should have gone to bed by now... ;)

Here is a tested version (not much better, but at least it works)


>>> def product(a,b):
...     a2 = reduce(lambda l, x, n=len(b): l + [x]*n,
...                [[]] + a)
...     b = b * len(a)
...     return map(None, a2, b)
... 
>>> product([1,2,3],[4,5,6])
[(1, 4), (1, 5), (1, 6), (2, 4), (2, 5), (2, 6), (3, 4), (3, 5), (3, 6)]
>>> 


--

  Magnus
  Lie
  Hetland


From JoeSmith at bogusaddress.com  Tue Jan 25 04:46:29 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Tue, 25 Jan 2000 09:46:29 GMT
Subject: Using tabs effectively. Was: Why not use tabs?
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <20000118.19182100@robert.polytopic.com> <38854A89.B3FBFA12@Lugoj.Com> <38858B76.AC22E62B@bogusaddress.com> <3887DF0D.B102B357@Lugoj.Com>
Message-ID: <388D71CD.1C6E5F2D@bogusaddress.com>

I was assuming that that tabs were for getting to the first real (significant)
character in the line.  Spaces need to be used after the first character in the
line (I neglected to mention this).  This allows any tab setting to work.  So in
your case you would use spaces after the A.  I assume your example is for aligning
variables or and equal signs.
    xyz = abc
    x    =  dfk
    int  x;
    long y;

This netscape thing is not doing the right thing for me, but tab is the first
character and the two ='s should be vertically aligned and the x and the y should
be verticly aligned.

I am definitely not talking about anything other than programming type of code.

James Logajan wrote:

> Joe Smith wrote:
> >
> > If one is using all tabs, then putting a comment at the top of the file
> > indicating the tab setting seems like a waste.  If one is mixing tabs and
> > spaces, then it is needed.  Maybe the comment should be one of the following:
> > using all tabs, using all spaces or mixing with a tab setting of X.
>
> Unfortunately, this wont work in general. You may be forgetting that tabs
> may also be embedded. For example, consider a file that contains the
> following text lines, where ^I indicates a tab:
>
> A^I^Ix
> AA^I^Ix
> AAA^I^Ix
> AAAA^Ix
> AAAAA^Ix
> AAAAAA^Ix
> AAAAAAA^Ix
>
> The intent is to have that look like:
>
> A       x
> AA      x
> AAA     x
> AAAA    x
> AAAAA   x
> AAAAAA  x
> AAAAAAA x
>
> There is only ONE tab setting that will produce that result. You would have
> to be either very clever, plow through the possibilities, or be told (via a
> comment) for you to discover the original tab setting. (The discovery of the
> tab setting used in the example is left as an exercise for the reader.)
>
> Hopefully that explains the reason for my desire to see a comment at the top
> of all source files indicating the tab setting, even when spaces aren't
> being used.



From mn at altern.org  Tue Jan  4 12:38:59 2000
From: mn at altern.org (Mirko Nasato)
Date: Tue, 04 Jan 2000 17:38:59 GMT
Subject: exception arg
Message-ID: <slrn8749le.d0.mn@banger.iol.it>

I can't understand the following:

>>> 1/0
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ZeroDivisionError: integer division or modulo
>>> try:
...     1/0
... except ZeroDivisionError, arg:
...     if arg == "integer division or modulo":
...             print "This is what I expect."
...     else:
...             print 'Unexpected. "arg" is "%s".' % arg
... 
Unexpected. "arg" is "integer division or modulo".
>>>

The same seems to occur with every kind of exception, so maybe
there's an explanation...


-- Mirko Nasato



From AKiss at GEOMETRIA.hu  Mon Jan 24 13:07:22 2000
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Mon, 24 Jan 2000 19:07:22 +0100
Subject: global variable problem
Message-ID: <D148AB3A9D8FD311A99A008048C79D1104BE3A@MAIL>


volumeregeling <volumeregeling at hotmail.com> wrote in message
news:<388c496a.9657176 at 122.0.0.250>...
> Hi,
> 
> 
> I am writing a program without classes.
> 
> I have the following :
> 
> Debugging = 0
> 
> def remote_debug():
>       Debugging = 1

Try this:

def remote_debug():
	global Debugging
      Debugging = 1

Arpad

> 
> 
> 
> For some reasong the Debuggine in the function is not the same as the
> global one , can you tell me how i can update the global variable from
> within a function ????  
> so i would like to update the global one when i am in the function and
> not create a second Debugging which disapears when the functions dies.
> 
> 
> thakns,
> 
> Volumeregeling.
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From piet at cs.uu.nl  Fri Jan  7 16:56:18 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 07 Jan 2000 22:56:18 +0100
Subject: bsddb item size?
References: <mrE03AA5nOd4Ew2R@jessikat.demon.co.uk>
Message-ID: <wzd7rd4kcd.fsf@sunshine.cs.uu.nl>

>>>>> Robin Becker <robin at jessikat.demon.co.uk> (RB) writes:

RB> Anybody know if there are item size limitations with bsddb?

I don't think so. They specifically chose an algorithm with no limitations
(other than total memory size).
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From wlfraed at ix.netcom.com  Wed Jan 19 23:07:50 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Wed, 19 Jan 2000 20:07:50 -0800
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <388670D2.2FA842F6@starmedia.net> <Pine.SOL.3.91.1000120001532.29235B-100000@sunse>
Message-ID: <tIeGOHk6cMDxz41yCwn96Iy45Dgx@4ax.com>

On Thu, 20 Jan 2000 00:20:12 +0300 (MSK), Evguenii Smogailov
<sei at sunse.jinr.ru> declaimed the following in comp.lang.python:

>
> Yes, i'm saying that 'eq', 'ne' are more readable then '==' '!=', '<>',
> 'next' shorter then 'continue' and '..' operator can be a good thing to 
> consider.
> 
	Yet after over 30 years of ".ne.", ".eq.", ".gt.", etc. in
FORTRAN, Fortran 90 has just added "<>", "=" (or is it "=="), ">", etc.
to permit one to get away from those awful text strings.

	The only languages I've ever seen that use a text string for
those operators are FORTRAN and VMS DCL.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From ivanlan at callware.com  Tue Jan  4 18:20:21 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 04 Jan 2000 16:20:21 -0700
Subject: Python & Apache
References: <A4uc4.411$hJ1.3104@news.rdc1.tx.home.com> <387272A1.F036408C@callware.com> <38727518.9895C19F@rubic.com>
Message-ID: <38728035.D857AA6E@callware.com>

Hi All--

Jeff Bauer wrote:
> 
> Ivan Van Laningham wrote:
> > The biggest gotcha is that the Python scripts *must* start with
> >
> >         #!c:\Python\python.exe
> >
> > (or wherever Python actually lives). You can't use
> >
> >         #!/usr/local/bin/python
> >
> > because Apache won't find it, even if the path exists
> > on Apache's drive--it insists on the driveletter and
> > the backslashes.
> 
> Hi Ivan.
> 
> My notes document otherwise.  I could be wrong, since
> I don't use the latest Win32/Apache, but I just copy
> python.exe into C:\usr\local\bin and I can use
> portable (for some loose definition of "portability")
> python cgi scripts between Unix and NT.
> 
> http://starship.python.net/crew/jbauer/apachenotes/
> 

Well, whatdya know.  As long as apache and python live on the same
drive, it does work.

Last time I read the release notes from Apache, they said that it
*required* the windows-style pathname.  Looks like they relaxed that
requirement.

Cool.

Thanks, Jeff.

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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From kochmar at pitfiend.psc.edu  Tue Jan 18 17:11:52 2000
From: kochmar at pitfiend.psc.edu (John Kochmar)
Date: 18 Jan 2000 22:11:52 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de>
Message-ID: <862of8$57j$1@news.psc.edu>

In article <948195584.987404 at news.adtranzsig.de>,
 "Dirk-Ulrich Heise" <hei at adtranzsig.de> writes:
|> tye4 schrieb in Nachricht <860ftk$bas at news.or.intel.com>...
|> >Indentation is a confusing and lame way to separate what is inside a block
|> >and what's outside it.
|> 
|> I'm irritated by the subject.
|> The indentation complaint re-occurs every week,
|> but this is the first time it comes packaged with the
|> subject line "Python rocks!"
|> 
|> Get used to it, and don't use tabs.

Why not?  I use tabs all the time (I was indenting my pascal code in a
python-like style in the early 80's, and have carried that over to C,
shell scripts, etc.)  Heck, I even mix tabs and spaces (although, the
later is almost exclusively to add a try/except block to an existing
script indented with tabs ;^)

BTW, I'm a dinosour who likes to vi, and hates when an editor sees that
I've hit a tab and says to itself "gee, he hit a tab, so he must really
want N spaces, so I'll silently replace it for him..."  If I hit a tab,
darn it, I MENT a tab.

John 'I like tabs' K


From gmcm at hypernet.com  Thu Jan  6 19:17:35 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 6 Jan 2000 19:17:35 -0500
Subject: no return values for __init__ ??
In-Reply-To: <3875152E.98598E4E@mdlink.de>
Message-ID: <1264918239-7223828@hypernet.com>

Helge Hess wrote:
> Gordon McMillan wrote:
> > >   def __init__(self...
> > >      s = super.__init__(self...
> > 
> > OK, but it introduces a "gotcha". As in, what happened to the
> > stuff I set up before calling super.__init__, (which is not
> > only valid, but often useful).
> 
> I don't know what you mean with that. It is deallocated if
> another value is returned and available to super.__init__ as
> usual. super.__init__ can base it's decision about the value to
> return on state already set up in self as well. I cannot see a
> problem here.

Hmmm...

class Base:
  def __init__(self, arg):
    if type(arg) is type(''):
      s = String(arg)
    elif type(arg) is UnicodeType:
      s = Unicode(arg)
    # preserve any mods to the instance
    s.__dict__.update(self.__dict__)
    if self.__class__ is Base:
        return s  
    # oops, there are other classes involved
    # now rebuild the class hierarchy
    klass = self.__class__
    while klass:
        # assuming no MI, so we don't
        # have to get recursive
        parent = klass.__base__          
        if parent and parent is Base:
            klass.__base__ =  #oops, 
            # what do I do now?

Guess what? I need to derive a new class on the fly!

- Gordon



From gmcm at hypernet.com  Mon Jan 17 19:24:20 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 17 Jan 2000 19:24:20 -0500
Subject: Can TkInter be frozen?
In-Reply-To: <3883A044.F60D145F@lmco.com>
Message-ID: <1263967434-11066381@hypernet.com>

Chuck Meyers asks:

> Can TkInter be frozen?

Sort of.

> I understand the answer used to be no, has it changed?

Years ago there was a script that would pack up all the *.tcl 
files and then unpack them on the other machine.

If you're on Windows, my Installer will do it for you:
http://starship.python.net/crew/gmcm

Haven't tried to incorporate that in the Linux version, because 
of issues with finding the tcl and tk shared libs at runtime.

> Is it expected to change?

My guess would be that if stock Tcl/Tk incorporated a freeze-
like utility that enabled this, then incorporating it into Python's 
freeze would follow. But I doubt that anything depending on a 
Tcl/Tk extension will ever find it's way into the Python 
distribution.

- Gordon



From bwarsaw at cnri.reston.va.us  Tue Jan  4 12:11:42 2000
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 4 Jan 2000 12:11:42 -0500 (EST)
Subject: IPC8 Hotel Group Rate Extension
Message-ID: <14450.10702.755118.236656@anthem.cnri.reston.va.us>

For those of you who are coming to the 8th International Python
Conference and are planning to stay at the hotel, but have /not/ yet
made your room reservations, please note that the hotel has extended
the Python group rate of $129/night (plus tax) until close of business
(US East Coast time) on January 6th, 2000.  After this date, the least
expensive room the hotel will offer is $179/night (plus tax).

For more information on the conference see

    http://www.python.org/workshops/2000-01/

Hotel information is available at

    http://www.python.org/workshops/2000-01/hotel.html

See you there!
-Barry



From bob at no.com  Sat Jan 22 10:02:51 2000
From: bob at no.com (bob)
Date: Sat, 22 Jan 2000 15:02:51 GMT
Subject: Python and Tkinter Programming 
References: <3888E695.A929D7C2@home.com>
Message-ID: <vEji4.962$253.16250@news>

I have checked out the PDF's and they are great. Tkinter has long needed
this level of documentation!

Bob


John Grayson <johngrayson at home.com> wrote in message
news:3888E695.A929D7C2 at home.com...
> I know that this is a *shameless* plug, but anyway...
>
> My book, Python and Tkinter Programming, is shipping
> from the printer today. Copies should be available at IPC8.
>
> There are two example chapters, in PDF format, on
> the Manning web site. If you'd like to see them
> before deciding to buy or not, point your browser
> at http://www.manning.com/grayson/
>
>      Cheers!
>
>        :-j ohn




From artymiak at safenet.pl  Fri Jan  7 16:39:35 2000
From: artymiak at safenet.pl (Jacek Artymiak)
Date: Fri, 07 Jan 2000 21:39:35 GMT
Subject: CGI problem
References: <855k3e$agm$1@nnrp1.deja.com>
Message-ID: <38766371.725C9D67@safenet.pl>

sp00fD wrote:
> 
> I'm creating a CGI program, and I keep getting an error from apache:
> 
> "(8)Exec format error exec of /usr/local/apache/htdocs/...
> Premature end of script headers: /usr/local/apache/htdocs/..
> 
> I've put the print "Content-type: text/html\n" (with anywhere from 0 to
> 8 newlines after it) in there.
> 
> What am I doing wrong?

Try this first, then add your own bells & whistles

#! /usr/local/bin/python

print "Content-type: text/html"
print
print '''

<html>

        <head>

        <meta http-equiv="content-type" content="text/html">

        <title>Hello</title>

        </head>

        <body>

		<h1>Hello!</h1>

	<body>

</html>

'''


 

Jacek Artymiak

-------------------------------------------------------------------
 See how many people have registered at the Python Users Counter!
http://www.wszechnica.safenet.pl/cgi-bin/checkpythonuserscounter.py
-------------------------------------------------------------------
    Autor/Dziennikarz/Konsultant - Author/Journalist/Consultant
       artymiak at safenet.pl, http://www.wszechnica.safenet.pl
co-author:  StarOffice for Linux Bible (IDG Books Worldwide, Inc.)
http://www.amazon.com/exec.obidos/ASIN/0764533630/polskawszechnica
-------------------------------------------------------------------


From grant at nowhere.  Thu Jan  6 17:25:51 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 06 Jan 2000 22:25:51 GMT
Subject: Origin of double underscore convention?
References: <Y96d4.1764$oF2.5206@ozemail.com.au>
Message-ID: <slrn87a5jf.neh.grant@grante.comtrol.com>

In article <Y96d4.1764$oF2.5206 at ozemail.com.au>, Jason Stokes wrote:

>I'm relatively new to Python.  This is possibly a dumb question, but
>I'm wondering about the double underscore convention to mark an inbuilt
>or internal method.  I find it pretty ugly and I'm wondering if it was
>chosen for any historical reason, eg analogy to another language.  

Some C compilers used to prefix all user-declared variables
with "_".  That way they would never collide with assembly
language variables as long as the assy lang didn't use variables
that started with "_".  

Or maybe it was the other way 'round...

Anyway, the custom for using prefixed "_" to simulate separate
namespaces has been around for many, many years.

-- 
Grant Edwards                   grante             Yow!  Yow! We're going to
                                  at               a new disco!
                               visi.com            


From bjorn at roguewave.com  Wed Jan  5 19:49:29 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 05 Jan 2000 17:49:29 -0700
Subject: SUGGEST: os.path.validate()?
References: <20000105135116.A2974@stopcontact.palga.uucp>
Message-ID: <3873E699.AE8AC19A@roguewave.com>

I would suggest creating a "filesystem" module for this since there are several
different cdrom formats in addition to all the os formats that would be interesting
(and one os might have multiple filesystems)...

just my $.02 worth
-bjorn

Gerrit Holl wrote:

> Hello,
>
> I've got a suggestion: what about a function os.path.validate()?
> For posix, it's easy:
> def validate(p):
>     return not '\0' in p
>
> but for Dos/Windows, it's harder, and I've no idea what's a valid
> path on Mac.
>
> regards,
> Gerrit.
>
> --
> -----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
> Version: 3.12
> GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
> Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
> -----END GEEK CODE BLOCK-----
>
> --
> http://www.python.org/mailman/listinfo/python-list




From moonseeker at my-deja.com  Wed Jan 19 09:42:43 2000
From: moonseeker at my-deja.com (moonseeker at my-deja.com)
Date: Wed, 19 Jan 2000 14:42:43 GMT
Subject: no multithreading?
Message-ID: <864ih0$5is$1@nnrp1.deja.com>

Hi!

I have understood that the python interpreter switches to another thread if a
fixed amount of python commands is completed. Is this true? That means if a
task hangs on a command (i.e. file open), all tasks are locked? What happens
if I use sleep() in a task? And, that would be mean that the python
multitasking is no real multitasking...

Any comments are appreciated.

mfg,
Mike


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


From mfletch at tpresence.com  Sun Jan 23 08:01:30 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 23 Jan 2000 08:01:30 -0500
Subject: Win32Installer + Numeric, anyone succeeded?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3097@VRTMAIL>

Yes, that was it (loads and runs now).  No, there doesn't appear to be an
import _numpy in my numeric directory anywhere.  Could be it's only in the
CVS repository?

Thanks,
Mike

-----Original Message-----
From: Just van Rossum [mailto:just at letterror.com]
Sent: Sunday, January 23, 2000 7:51 AM
To: Mike Fletcher; Python Listserv (E-mail)
Subject: Re: Win32Installer + Numeric, anyone succeeded?
...
Numeric relies on _numpy.pyd, but _numpy was not imported in any Python
code, just C code, so freeze's modulefinder (which the Win installer also
uses I think) didn't notice it. And Numeric indeed crashes horribly when
the _numpy shared lib isn't available. I think this was later fixed by a
dummy "import _numpy" somewhere. Maybe you can check whether _numpy.pyd
gets included at all, and if your version of Numeric includes the dummy
import (in Numeric.py I believe).
...



From aahz at netcom.com  Fri Jan 21 10:32:46 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 15:32:46 GMT
Subject: olemsg32 module used in win32com exchange server example?
References: <slrn8897gs.3cm.grant@grante.comtrol.com> <slrn88ceoc.5ga.grant@grante.comtrol.com> <38863DB3.7FAEC820@callware.com> <slrn88e58v.67b.grant@grante.comtrol.com>
Message-ID: <869u6u$qtq$1@nntp9.atl.mindspring.net>

In article <slrn88e58v.67b.grant at grante.comtrol.com>,
Grant Edwards <grant at nowhere.> wrote:
>In article <38863DB3.7FAEC820 at callware.com>, Ivan Van Laningham wrote:
>>
>>I suspect that is true, but there's a gotcha.  Unless I'm wrong, both
>>NetScape Mail and Outlook use MAPI dll's, but they expect different
>>functions to be found there.  So, when I want to use Outlook to retrieve
>>mail, I can configure everything else correctly but I have to change the
>>MAPI.DLL to the one the Outlook expects or it barfs.  Then put it back
>>when I want to use NetScape. ...
>
><Boggle>

It's even worse than that.  AFAIK, NT *still* only loads one copy of a
DLL, so that even if you have DLLs in two different directories, it is
impossible simultaneously to run programs that use a same-named DLL if
they need even different versions.  And, of course, most DLLs still use
eight-character names....
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From quinn at mono.ugcs.caltech.edu  Fri Jan 28 18:03:43 2000
From: quinn at mono.ugcs.caltech.edu (Quinn Dunkan)
Date: 28 Jan 2000 23:03:43 GMT
Subject: bug in pdb?
Message-ID: <slrn894859.7f4.quinn@mono.ugcs.caltech.edu>

Apparently, pdb makes use of a module called repr (which I think is a
*terrible* name for a module).  When I try to call repr(foo) from pdb, I get
TypeError, and worse, when I try to step over code which _uses_ repr, the
program blows up.

% cat >bug.py
foo = 'baz'
print repr(foo)
^D
% /usr/local/lib/python1.5/pdb.py bug.py
> <string>(0)?()
(Pdb) n
> <string>(1)?()
(Pdb) 
TypeError: 'call of non-...(type module)'
> <string>(1)?()
(Pdb) 
Traceback (innermost last):
  File "/usr/people/web/bin/pdb", line 921, in ?
    run('execfile(' + `filename` + ')')
  File "/usr/people/web/bin/pdb", line 855, in run
    Pdb().run(statement, globals, locals)
  File "/usr/local/lib/python1.5/bdb.py", line 343, in run
    exec cmd in globals, locals
  File "<string>", line 1, in ?
  File "bug.py", line 2, in ?
    print repr(foo)
TypeError: call of non-function (type module)


From donn at oz.net  Sun Jan  2 06:23:23 2000
From: donn at oz.net (Donn Cave)
Date: 2 Jan 2000 11:23:23 GMT
Subject: Finding where Python is on a Unix server through Telnet?
References: <386EA0EF.F7A72441@jslove.net> <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com>
Message-ID: <84ncfb$e5q$0@216.39.151.169>

Quoth Jay Love <jsliv at jslove.net>:
| Tyler Eaves wrote:
|> 
|> What is the command?
|
|
| try 'which python'
| or for everything related to python 'locate python'

"which" is definitely the way to go from the csh or tcsh shells.

In bash or ksh, it may work, but it also may lie - if it's important
to find the actual path among more than one possible paths for the binary,
you should use "type" (or "whence") in these shells.

"locate" is a new one on me, not available on Digital UNIX.

	Donn Cave, donn at oz.net


From romtur at starnets.ro  Wed Jan 12 02:55:02 2000
From: romtur at starnets.ro (romtur at starnets.ro)
Date: Wed, 12 Jan 2000 07:55:02 GMT
Subject: PyGreSQL on RHL 6.0
Message-ID: <85hc0h$h25$1@nnrp1.deja.com>

Hello !

My name is Andrei Cenja, and I live in Bucharest, Romania.

I've downloaded PyGreSQL, but there were some problems
when I tried to install the module (under RedHat Linux 6.0)
I followed the steps in the Reame.linux file, but I ended up with
"Import Error: /usr/lib/libpq.so.2.0: undefined symbol: crypt" when I
try to import _pg.

Could you please issue some kind of opinion on what went wrong ?

Thanks.

Andrei



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


From sholden at bellatlantic.net  Fri Jan 14 16:44:37 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Fri, 14 Jan 2000 21:44:37 GMT
Subject: Tcl/Tk Path problems in Python 1.5.2 installation
References: <387F80E5.A0AF82A5@sourcesignal.com>
Message-ID: <387F98DE.EC006329@bellatlantic.net>

Had a similar problem myself.  I suspect that the directory which
the .dll's install into should be added to the PATH, and mine was
already very long (due to installing PowerBuilder, which puts
four or five directories on their).

On Windows NT I ficed things by copying the libraries into the
\WINNT\SYSTEM32 directory -- but of course they won't uninstall
if you ever (gasp!) decide you don't need Pythin any more -- and
they may leave a gotcha waiting to hit you after updates.

regards
 Steve

David Nichols wrote:
> 
> I just installed python 1.5.2 (running py152.exe and answering 'yes - I
> want to install Tcl/Tk') and have been unable to get the path set
> up correctly for graphical applications.  For instance, if I go to the
> idle directory (//E/Programs/Py152/Tools/idle) and try to launch it I
> get the following result.
> 
> ============================================================
> BASH.EXE-2.02$ python idle.py
> Traceback (innermost last):
>   File "idle.py", line 3, in ?
>     PyShell.main()
>   File "E:\Programs\Py152\Tools\idle\PyShell.py", line 611, in main
>     root = Tk()
>   File "E:\Programs\Py152\Lib\lib-tk\Tkinter.py", line 886, in __init__
>     self.tk = _tkinter.create(screenName, baseName, className)
> TclError: Can't find a usable init.tcl in the following directories:
>     {} ./lib/tcl8.0 . E:/Programs/Py152/tcl8.0/library
> E:/Programs/Py152/Tools/library
> 
> This probably means that Tcl wasn't installed properly.
> ============================================================
> 
> The directories in sys.path are as follows ...
> E:\PROGRAMS\PY152
> E:\PROGRAMS\PY152\DLLs
> E:\PROGRAMS\PY152\PIL
> E:\PROGRAMS\PY152\lib
> E:\PROGRAMS\PY152\lib\lib-tk
> E:\PROGRAMS\PY152\lib\plat-win
> E:\Programs\Py152\DLLs
> E:\Programs\Py152\Lib
> E:\Programs\Py152\Lib\lib-tk
> E:\Programs\Py152\Lib\plat-win
> 
> Tcl in installed in E:\PROGRAMS\Py152\\Tcl, and init.tcl exists in
> E:\PROGRAMS\Py152\Tcl\lib\tcl8.0 ... but this directory does not
> seem to be searched.
> 
> Do I need to set additional environmental variables or make changes
> to the registry beyond those made by the installer?
> 
> Thanks in advance for any help
> 
> David Nichols
> Source Signal Imaging
> dnichols at sourcesignal.com


From thamelry at vub.ac.be  Fri Jan 28 11:21:33 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 28 Jan 2000 16:21:33 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com>
Message-ID: <86sfmd$ktl$1@mach.vub.ac.be>

Glen Starchman <glen at electricorb.com> wrote:
[snip]
: Personally, there is nothing I hate more in the curly-brace languages than the fact
: that developers don't have to line up their braces to show groupings of code. I
: guess if every C/C++/Java, etc... developer wrote code like:

: for (i=0; i < 100; i++)
: {
:     do something;
: } /* notice how the braces line up and the body is indented!*/

: I would be happy with that approach. However, there is nothing better than forcing
: clean code... in my very humble opinion.
[snip]

If braces are used, you can choose between a myriad of tools to get 
consistent indentation (there are lots of C and C++ beautifiers out
there). However, if the indentation in python is screwed up you 
are lost completely. No tool can correct that (yes and I do know
pindent).

Cheers,

---
Thomas Hamelryck    Institute of Molecular and Structural Biology
Aarhus University   Gustav Wieds Vej 10C
DK-8000 Aarhus C    Denmark



From tim_one at email.msn.com  Sun Jan 23 23:45:58 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 23 Jan 2000 23:45:58 -0500
Subject: tabs do WHAT?
In-Reply-To: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net>
Message-ID: <000601bf6625$e80c3340$902d153f@tim>

[Warren Postma]
> Let's see how many languages I have run into that make
> whitespace and/or tabs important:

Whitespace?  Now *that's* a novel topic c.l.py hasn't seen enough of lately
<wink>.

> ...
> As an embedded language, Python needs to be usable beyond the
> ivory towers where functional programming or object oriented
> programming or metaclasses are fodder for discussion.

Python didn't come from an ivory tower, and indeed even Perl may be more
heavily used in those fabled lands.

> Python looked eminently useable by non-programmers until I found
> out that whitespace was meaningful.  Try to explain scope to
> someone who doesn't write programs for a living.

This one is ironic:  Python's treatment of block structure came from formal
usability studies (conducted with the ABC language), using subjects who
indeed didn't write programs for a living.  "Introductory colon +
indentation" is what worked best -- they tried the others too.  See recent
msgs in DejaNews for more about that.

> ...
> Has anyone built a version of Python that totally ignores
> tabs and whitespace,

Yes, a patch for that was posted many years ago; it didn't catch on.

> or a filter that you can use on top of Python?

See Tools/Scripts/pindent.py in the std distribution.

although-i'm-not-sure-it's-ever-been-run<wink>-ly y'rs  - tim





From tim_one at email.msn.com  Mon Jan 24 19:30:37 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jan 2000 19:30:37 -0500
Subject: What about REBOL (www.rebol.com)
In-Reply-To: <mm0j4.48027$Rj5.26574@dfw-read.news.verio.net>
Message-ID: <000e01bf66cb$670a8480$a42d153f@tim>

[posted & mailed]

[Tim]
> ... (despite that it *looks* nothing like Lisp, a REBOL
> "block" is a lexical closure, and the semantics are
> virtually identical to Scheme's)

[Joe Marshall]
> REBOL abandoned lexical closures in version 2.0, so the
> semantics have become quite different from Scheme's.

I eagerly yield to Joe on all matters REBOL -- he was the primary
implementer of the version I used most.  So, Joe, should the original
questioner learn Python or REBOL?

c.l.py-is-a-low-pressure-group<wink>-ly y'rs  - tim





From skaller at maxtal.com.au  Sat Jan  8 09:50:34 2000
From: skaller at maxtal.com.au (skaller)
Date: Sun, 09 Jan 2000 01:50:34 +1100
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com>
Message-ID: <38774EBA.40716D1E@maxtal.com.au>

Skip Montanaro wrote:
> 
>     Emile> Doesn't this do that now? or is there some objection to this usage?
> 
>     [ ... snip try/except used to implement state machine ... ]
> 
> I find nothing wrong with it.  Some people object to using try/except for
> anything but handling errors however.

My objection is twofold. First, the syntax is unweildy:

	try:
		for i in range(3):
			if ... : raise X
	except X:

requires at least one extra level of indentation.

[I'm interested in a 'generic' way to combine
python statements which avoids this (like 'elif' does)
Any ideas?]

	My second objection is more serious:
try .. raise .. except is too dynamic (for this usage): 
there is no way to match up the exceptions actually 
thrown with the handlers.

	The constructions under consideration
(variants on a construction proposed by Knuth)
would support compile time verification that
every loop exit had a corresponding handler
lexically following the loop.

	In particular, 'falling through'
such a loop is impossible: all exits must be explicit.

	Current looping constructions are so bad,
that the simplest of uses cannot be expressed
using them : searching a list for an element
cannot be written correctly:

	for i in X: 
		if X[i]=e: goto Found
	print "Not Found"
	goto next
	Found: ...
	next

YUKKKKKK!!! There is no natural way to do this.
We fall through the loop on failure, instead
of success. (using a flag, or testing anything
afterwards, is not acceptable as a solution).

Using exceptions:

	try:
		for i in X:
			if X[i]==e: raise Found
		raise NotFound
	except Found: ..
	except NotFound: ..

is much cleaner, but exceptions are excessively
dynamic for what is essential a static control
flow problem. A solution will resemble the
above exception handling implementation .. but
with the matching up of exits and handlers enforced
at compile time.

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From otis at my-deja.com  Sat Jan 15 16:26:58 2000
From: otis at my-deja.com (Otis Gospodnetic)
Date: Sat, 15 Jan 2000 21:26:58 GMT
Subject: universal DB module (a la Perl's DBI/DBD)
Message-ID: <85qomm$91i$1@nnrp1.deja.com>

Hello,

I need to write some Python apps (CGIs, actually) that talk to a DB
backend and I would like to use some existing modules for that.
Since the backend DB vendor is not yet known, I would love to use some
modules that will make my code DB-independent
I'm looking for something like Perl's DBI/DBD modules that make it
possible to use the same code for a number of different RDBMSes (Oracle,
Sybase, Informix, MySQL, etc.)

Are there such Python modules?

Thanks!

Otis


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


From rumjuggler at home.com  Tue Jan 18 22:45:00 2000
From: rumjuggler at home.com (Ben Wolfson)
Date: Wed, 19 Jan 2000 03:45:00 GMT
Subject: quinoid script
Message-ID: <t5da8scaemd9j3fnh0j49hj2vk4tuha718@4ax.com>

Got bored, wrote the below.  
It opens a file with the same name as itself but with an extra '1' before
the .py extension, then deletes the file that created it.
Just wanted to point out how easy it was to write.


-----------FILE STARTS HERE---------
import os, sys

s = """
import os, sys

s = %c%c%c%s%c%c%c
open((sys.argv[0][:-3]+'1'+sys.argv[0][-3:]),'w').write(s %%
(34,34,34,s,34,34,34))
try:
	os.remove((sys.argv[0][:-4]+sys.argv[0][-3:]))
except:
	pass
"""
open((sys.argv[0][:-3]+'1'+sys.argv[0][-3:]),'w').write(s %
(34,34,34,s,34,34,34))
try:
	os.remove((sys.argv[0][:-4]+sys.argv[0][-3:]))
except:
	pass

-- 
Barnabas T. Rumjuggler

Liberty is commonly thought of not as a constant, but as something which
must be made to expand indefinitely, lest it become boring.
 -- The Flummery Digest



From heinkel at dummy.embl-heidelberg.de  Mon Jan 10 05:53:49 2000
From: heinkel at dummy.embl-heidelberg.de (Ralph Heinkel)
Date: 10 Jan 2000 10:53:49 GMT
Subject: purify and python
Message-ID: <85cdnt$avq$1@beta.EMBL-Heidelberg.DE>

Hi, 

since I wrote some C, C++, and F77 extensions for python, I am planning
to use purify to test them for memory leaks and so on.
The problem is that I cannot get a purified version of python running.
Python seems to start and immediately exit with error code -1 (255),
without printing any error message (See below for log). I never get
to the python command line.

The way I linked python is shown below. Linking python without purify
(without any other recompilation) produces a working python version.

Any clue? 

I am running python 1.5.2 on IRIX 6.5 (R10000) with purify 4.1. Python
was configure with  --with-thread  and  --without-gcc

Ralph

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

Modules/Setup is:

# Pure module.  Cannot be linked dynamically.
# -DWITH_QUANTIFY, -DWITH_PURIFY, or -DWITH_ALL_PURE
WHICH_PURE_PRODUCTS=-DWITH_PURIFY
PURE_INCLS=-I/usr/pure/purify
PURE_STUBLIBS=-L/usr/pure/purify -lpurify_stubs -lquantify_stubs
pure puremodule.c $(WHICH_PURE_PRODUCTS) $(PURE_INCLS) $(PURE_STUBLIBS)

The (python) modules are linked statically, i.e. *shared* not used


This is how I linked it:

birdie:src/Python-1.5.2/Modules$ purify -windows=no cc -g  python.o \
 ../libpython1.5.a -DWITH_PURIFY -I/usr/pure/purify -L/usr/pure/purify \
 -lpurify_stubs_n32 -lpthread -lmpc -lm  -o python

ld32: WARNING 84: /usr/lib32/mips4/libmpc.a is not used for resolving 
any symbol.
Purify 4.1 IRIX6, Copyright (C) 1992-1997 Rational Software Corp. All 
rights reserved. 
.... Processing "/usr/pure/purify/libpurify_stubs_n32.so"
.... Processing "/usr/lib32/libpthread.so"
.... Processing "/usr/lib32/libm.so"
.... Processing "/usr/lib32/libc.so.1"
.... Processing "python"



This is the output:

birdie:src/Python-1.5.2/Modules$ ./python 
****  Purify instrumented ./python (pid 63792545 at Mon Jan 10 11:48:44 2000)
  * Purify 4.1 IRIX6, Copyright (C) 1992-1997 Rational Software Corp. 
All rights reserved. 
  * For contact information type: "purify -help"
  * Command-line: ./python 
  * Options settings: -purify -windows=no -purify-home=/usr/pure/purify 
  * Purify licensed to European Molecular Biology La
  * Purify checking enabled.

****  Purify instrumented ./python (pid 63792545)  ****
Current file descriptors in use: 5
FIU: file descriptor 0: <stdin>
FIU: file descriptor 1: <stdout>
FIU: file descriptor 2: <stderr>
FIU: file descriptor 26: <reserved for Purify internal use>
FIU: file descriptor 27: <reserved for Purify internal use>

****  Purify instrumented ./python (pid 63792545)  ****
Purify: Searching for all memory leaks...

****  Purify instrumented ./python (pid 63792545)  ****
SIG: Signal handled:
  * This is occurring while in:
        _x_open        [open.s:16]
        _exit          [syscalls.c:5478]
        exit_certain   [fork.c:202]
        _SGIPT_libc_exit [fork.c:253]
        ptctl          [libcthread.c:149]
        __exit         [cuexit.c:60]
  * Received signal 48 (SIGReserved48 - System reserved signal 48)
  * Handler function: 
        sched_handler  [vp.c:243]
  * Signal mask: (SIGReserved48) 
  * Pending signals: 
Memory leaked: 0 bytes (0%); potentially leaked: 2048 bytes (1.12%)

PLK: 2048 bytes potentially leaked at 0x1010c558
  * This memory was allocated from:
        _malloc        [malloc.c:886]
        vp_alloc       [vp.c:1214]

Purify Heap Analysis (combining suppressed and unsuppressed blocks)
                         Blocks      Bytes
              Leaked          0          0
  Potentially Leaked          1       2048
              In-Use          4     180236
  ----------------------------------------
     Total Allocated          5     182284

****  Purify instrumented ./python (pid 63792545)  ****
  * Program exited with status code -1.
  * 1 access error, 1 total occurrence.
  * 0 bytes leaked.
  * 2048 bytes potentially leaked.
  * Basic memory usage (including Purify overhead):
     2179072 code
      169108 data/bss
      195436 heap (peak use)
        4960 stack
  * Shared library memory usage (including Purify overhead):
      425984 rtlib_n32.so (shared code)
       57812 rtlib_n32.so (private data)
       16384 libpurify_stubs_n32.so.pure (shared code)
         160 libpurify_stubs_n32.so.pure (private data)
      212992 libpthread.so.pure.1b7a5a8a8 (shared code)
        3680 libpthread.so.pure.1b7a5a8a8 (private data)
      475136 libm.so.pure.18462ee2e (shared code)
       95696 libm.so.pure.18462ee2e (private data)
     2883584 libc.so.1.pure.9f4e4221 (shared code)
       70640 libc.so.1.pure.9f4e4221 (private data)
  * Memory mapped usage:
       49152 mmap'd at 0x4888000
     1425408 mmap'd at 0x48c8000
      262144 mmap'd at 0x4714000
       49152 mmap'd at 0x4020000
     4259840 mmap'd at 0x42f4000
       20480 mmap'd at 0x48b0000

-- 
------------------------------------------------------
Ralph Heinkel
European Molecular Biology Laboratory (EMBL)
Meyerhofstr. 1
69012 Heidelberg
Tel. +49 6221/387 529
eMail: heinkel at embl-heidelberg.de


From tpchang at my-deja.com  Fri Jan 14 02:39:38 2000
From: tpchang at my-deja.com (tpchang at my-deja.com)
Date: Fri, 14 Jan 2000 07:39:38 GMT
Subject: Python native compiler on NT
References: <84cdo5$9u0$1@nnrp1.deja.com> <3869EA26.9AE31E6F@matavnet.hu>
Message-ID: <85mjrp$ci5$1@nnrp1.deja.com>

Thanks for the info. I tried the "Installer". However, to run
the "compiled" .exe file, say xyz.exe, you have to copy xyz.exe and
zlib.pyd to current directory before you can run it. Some how putting
xyz.exe and zlib.pyd in a directory in the PATH, and run xyz.exe from
another directory will have a error message something like:

   cannot load zlib.pyd

Do you know what's the problem?

Thanks in advance.

Ben

In article <3869EA26.9AE31E6F at matavnet.hu>,
  Arpad Kiss <sekter at matavnet.hu> wrote:
> tpchang at excite.com wrote:
> >
> > I wonder if there is a Python native compiler for NT. This will
help me
> > to sell Python to people in my company who are more familiar
with .exe
> > files than the .py files :-)
> >
> > Thanks in advance,
> >
> > Ben
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
> > --
> > http://www.python.org/mailman/listinfo/python-list
>
> Check out http://starship.python.net/crew/gmcm/install.html
>
> Arpad Kiss
>
>


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


From gmcm at hypernet.com  Fri Jan 21 11:15:46 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 21 Jan 2000 11:15:46 -0500
Subject: olemsg32 module used in win32com exchange server example?
In-Reply-To: <869u6u$qtq$1@nntp9.atl.mindspring.net>
Message-ID: <1263651147-15120752@hypernet.com>

Aahz wrote:
> Grant Edwards <grant at nowhere.> wrote:
> >In article <38863DB3.7FAEC820 at callware.com>, Ivan Van Laningham wrote:
> >>
> >>I suspect that is true, but there's a gotcha.  Unless I'm wrong, both
> >>NetScape Mail and Outlook use MAPI dll's, but they expect different
> >>functions to be found there.  So, when I want to use Outlook to retrieve
> >>mail, I can configure everything else correctly but I have to change the
> >>MAPI.DLL to the one the Outlook expects or it barfs.  Then put it back
> >>when I want to use NetScape. ...
> >
> ><Boggle>
> 
> It's even worse than that.  AFAIK, NT *still* only loads one copy of a
> DLL, so that even if you have DLLs in two different directories, it is
> impossible simultaneously to run programs that use a same-named DLL if
> they need even different versions.  

Wrong about that. Though 2 different versions of a mapi dll 
loaded at the same time might yield trouble, it's not because 
you can't load them. I do it all the time with (different) 
python15.dlls. 

- Gordon



From bruce_dodson.nospam at bigfoot.com  Wed Jan 12 23:55:57 2000
From: bruce_dodson.nospam at bigfoot.com (Bruce Dodson)
Date: Thu, 13 Jan 2000 00:55:57 -0400
Subject: Python and ArcView GIS
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca> <85hs33$4t9$2@newsserver.rrzn.uni-hannover.de>
Message-ID: <XXcf4.16787$2x3.303971@newscontent-01.sprint.ca>

Bernhard Reiter wrote in message
<85hs33$4t9$2 at newsserver.rrzn.uni-hannover.de>...
>
>Sounds interesing.
>Why do you need to embed python in ArcView?
>Would it not be enough to just use the COM objects with python's
>win32com.client?

Sorry, I forgot who I was talking to.  :-)  I didn't mean to imply that
ArcView would bestow this benefit on Python.  I meant the opposite: Without
Python, ArcView doesn't have any way to deal with COM objects.  With Python
(and win32com), it does have a way.  So your suggested alternative is
exactly what I meant in the first place!

>Thanks for sharing your work in any way, though.
>I hope you will license it as free software.


It is definitely free software.  The license will be BSD-style.

Cheers,
Bruce



From bjorn at roguewave.com  Mon Jan  3 12:24:07 2000
From: bjorn at roguewave.com (bjorn)
Date: Mon, 03 Jan 2000 10:24:07 -0700
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl> <386E7BD4.514764EA@ndh.net> <84nv0u$bu9$1@nntp9.atl.mindspring.net>
Message-ID: <3870DB37.761034EA@roguewave.com>

Aahz Maruch wrote:

> In article <386E7BD4.514764EA at ndh.net>,
> Stefan Schwarzer  <s.schwarzer at ndh.net> wrote:
> >
> >The other topics which the author on the O'Reilly site, Greg Wilson,
> >mentions might also be worth considering. For example, I never needed
> >tuples (explicitly) yet, i. e. in a sense that lists would not work.
> >(If tuples would be taken out of the language or their use deprecated)
> >tuple un/packing would become list un/packing and so on.)
>
> Anyone who claims that tuples are unnecessary has never used a tuple as
> a dict key.  That requirement necessitates differentiating between a
> mutable and immutable list -- might as well stick with tuples.
> --

So what you're saying is that dicts are broken since they can't take lists as keys
<wink>.

undefined-behavior-isn't-always-bad'ly y'rs
-- bjorn





From gerrit.holl at pobox.com  Mon Jan 31 10:16:46 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 31 Jan 2000 16:16:46 +0100
Subject: Imagemagick
In-Reply-To: <wzu2jvwdyt.fsf@sunshine.cs.uu.nl>; from piet@cs.uu.nl on Sun, Jan 30, 2000 at 08:34:34PM +0100
References: <wzzotoal2k.fsf@sunshine.cs.uu.nl> <20000129201210.A4953@stopcontact.palga.uucp> <wzu2jvwdyt.fsf@sunshine.cs.uu.nl>
Message-ID: <20000131161646.A959@stopcontact.palga.uucp>

Piet van Oostrum wrote on 949260874:
> >>>>> Gerrit Holl <gerrit.holl at pobox.com> (GH) writes:
> 
> GH> Piet van Oostrum wrote on 949163987:
> >> Does anybody have a copy of Python/Imagemagick or pymagick or a working link?
> 
> GH> Google -> PythonMagick -> I'm feeling lucky
> GH> http://starship.skyport.net/crew/zack/pymagick
> 
> But when you try to download something all links are dead alleys or invalid
> files. 

I'm sorry. I should have looked further than my nose is long.

(maybe this proverb is Dutch only, but who cares; everyone speaks
 Dutch anyway)

groeten,
Gerrit.

-- 
Please correct any bad Swahili you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From akituc at pk.highway.ne.jp  Wed Jan 26 04:19:54 2000
From: akituc at pk.highway.ne.jp (Akinori Tsuchida)
Date: Wed, 26 Jan 2000 18:19:54 +0900
Subject: Need Preprocessor Module!
References: <388E124C.B6B9BF9A@pk.highway.ne.jp> <388E3B6F.C035D5A@saiph.com>
Message-ID: <388EBC39.A05D36FE@pk.highway.ne.jp>

Thanks Skip and Pierre.

I admit M4 is powerfull tool but its syntax, grammar is nothing like the ones
we are familiar with.

If possible, I would like to complete my work with Python only.

I have started to write one but found it's not so easy task as I once
expected.

If you know any preprocessing module written in Python,
please reply to this article.

Anyway thanks.

Akinori Tsuchida.




From niliinsk at reed.edu  Mon Jan 31 13:11:10 2000
From: niliinsk at reed.edu (Noah Iliinsky)
Date: 31 Jan 2000 10:11:10 -0800
Subject: where is bbpy?
Message-ID: <3895d03e@dnews.reed.edu>

Howdy.

The Python FAQ section 7.10 ( http://www.python.org/doc/FAQ.html#7.10 )
claims that there is a Python plug-in for BBedit in the Mac distribution
called BBPy. However, there is no sign of it in the distribution or
anywhere outside of the FAQ. Where has it gone? Has anyone seen it?

Cheers, Noah
--
   Noah Iliinsky          niliinsk at reed dot edu          I use PGP.

      "If we decided that community came first, how would we use our
      tools differently?"  -Howard Rheingold, Wired 7.01 pg. 163


From rwmcgwier at home.com  Sun Jan 16 22:58:45 2000
From: rwmcgwier at home.com (Robert W. McGwier)
Date: Mon, 17 Jan 2000 03:58:45 GMT
Subject: Pardon my ignorance, data conversion
Message-ID: <38829357.ED633932@home.com>

I need to do the simple task of taking a nonnegative floating point
variable and taking its integer part and storing it in an integer.
Damned if I can find how to do it.

In "C"::


    int i;
    double x;

    i = floor(x);


Can this be so hard or am I missing the obvious?

Sorry
Bob




From composzioni at tiscalinet.it  Thu Jan 13 01:52:37 2000
From: composzioni at tiscalinet.it (Giovanni Riccardi)
Date: Thu, 13 Jan 2000 07:52:37 +0100
Subject: Some bug in IDLE??
Message-ID: <85jt7t$760$1@pegasus.tiscalinet.it>

I wrote a small script that uses input() to get
the user input,get the values, make some calculation
and return calculated value.
The problem is that it works when the script is executed directly by the
interpreter, but not in IDLE environment where after the user has entered
some values the script doesn't return to the main loop.
For example:

print "Enter the name:"
x = input("--> ")
print "Enter the age:"
y = input("--> ")

doesn't continue after the first input().

Some help??
thanx in advance
Giovanni


--
-------------------------------------------
GIOVANNI RICCARDI * TERRACINA (LT) * ITALY

EMAIL: composizioni at tiscalinet.it
  ICQ: 40670340
-------------------------------------------

*******************************************
TAKE A LOOK TO MY OWN PROJECTS ON THE WEB
*******************************************
*         INFORM ITALIAN WEB SITE         *
*(a design system for Interactive Fiction)*
*                                         *
*        http://inform.freeweb.org        *
*******************************************
*******************************************
*         c o m p o s i z i o n i         *
*        digital  art & technology        *
*                                         *
*     http://composizioni.freeweb.org     *
*******************************************





From poet at linuxports.com  Mon Jan  3 22:09:01 2000
From: poet at linuxports.com (poet at linuxports.com)
Date: Tue, 04 Jan 00 03:09:01 GMT
Subject: OT: OpenSource Python Books?
Message-ID: <mt2.0-29451-946955341@news.informatik.uni-bonn.de>

Hello,

I represent LinuxPorts.Com (http://www.linuxports.com/) and I have been
approached by a printer to possibly publish OpenSource books. To start
these books would just be paper versions of the actual documentation.
In this case, the Python Documentation and we were wondering if this
would be of interest to the Python Community.

I spoke with the printer and we both have agreed that if we do this
that we will be providing a portion of the Gross Profits back to the
individual projects. In this case the Python project.

Anyway, if this of interest to you please go to
http://www.linuxports.com/ and make your vote. It will help us
determine the demand for such a book.

Thank you for your time,

Joshua Drake
Webmaster Linux Documentation Project
Owner LinuxPorts.Com
poet.nospam at linuxports.com


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

-- 
----------- 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 parkw at better.net  Mon Jan 10 03:57:33 2000
From: parkw at better.net (William Park)
Date: Mon, 10 Jan 2000 03:57:33 -0500
Subject: New Python Books
In-Reply-To: <Pine.LNX.4.10.10001090941020.26842-100000@crazypenguins.commandprompt.com>
References: <Pine.LNX.4.10.10001090941020.26842-100000@crazypenguins.commandprompt.com>
Message-ID: <20000110035733.A13047@better.net>

On Sun, Jan 09, 2000 at 09:56:25AM -0800, python at commandprompt.com wrote:
> Hello,
> 
> I recently posted this message to comp.os.python:

<snip>

> We received about 8 responses with feedback that we much appreciated and
> would now like to solicit some more. If the people
> that read this email could just think about a few of the following
> questions and provide some feedback we would be very thankful.
> 
> We are considering publishing the current set of Python documentation as a
> single volume versus several volumes such as toExcel does. Would this be
> of interest to the Python community?

Ultimately, it depends on the price.  Difficulty with this type of books
are
    - documentation changes almost every month, so it's old when it
      comes out
    - people don't read nor need the whole online documentations
    - particularly, library doc (which takes up more than 50%) is
      reference and is not designed to be read page-by-page
Even in Linux market, I have seen 
    - LDP+Linux-HOWTOs
    - entire manpages
in the past, but I haven't seen them for some time.


> 
> One person made mention of the "Vaults of Parnassus" if we were to include
> a snapshot of this would this be of interest?
> 
> We are also considering "paying" someone to write a book about Python.
> Unfortunately, unlike traditional publishers the author would not get paid
> until the book was done, and then it would be on the basis of per book
> sold. We are aware that this is not the normal way of doing things, but we
> are not Prentice Hall and this would be the only way "currently" that we
> could pull this off. Is there anyone on this list that feels confident
> enough in their Python skills and writing ability to be able to perform
> such a task?

What type of books are you looking for?
On python subject or other topics as well?
An encyclopedia for beginners or practical introduction for programmers?

William

> 
> Last but definitely not least, no matter what we do, if we publish a
> Python book, 20% to 40% of the Gross Profits would go back to the PSA.
> 
> 
> Thanks for your time:
> 
> Joshua Drake
> Webmaster - LinuxPorts.Com
> WebMaster - LinuxDoc.org
> 
>  
>  
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From anders.eriksson at morateknikutveckling.se  Mon Jan 10 11:41:04 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Mon, 10 Jan 2000 17:41:04 +0100
Subject: starting a cgi with params
Message-ID: <iQp6OHyH9j2y64kYNrpGPcKoXKu0@4ax.com>

Hello!

I have created a Python script which takes 2 params, a filename and an
operation code. The script works well when I run it from the command
prompt. My question is:

How do I send 2 params to the script when I'm using it as an CGI
script?

At the moment I'm using sys.argv[1] and sys.argv[2] to get the params
but it doesn't work when I call it from a web page.

// Anders



From s.schwarzer at ndh.net  Mon Jan 24 20:03:17 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Tue, 25 Jan 2000 02:03:17 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse> <38866E35.F04D4ADE@ndh.net> <20000121135451.B2265@stopcontact.palga.uucp>
Message-ID: <388CF655.1FEFEC8@ndh.net>

Hi Gerrit,

Gerrit Holl schrieb:
> > Perhaps the ".." operator is something to consider (if it isn't
> > incompatible with somethings else in the language), but most of
> > the times "range" does it's job very well.
> 
> Unfortunately, not yet. It's quite hard to get a string which
> would be a..k in Perl. Maybe a nice feature for range()?

I think, for consistency, range should return a list, not a string.
So 'a'..'d' should yield [ 'a', 'b', 'c', 'd' ], not 'abcd'.
Additionally, a..d should _not_ yield [ 'a', 'b', 'c', 'd' ] IMHO;
I would expect this to evaluate the variables a and d and compute the
resulting range. Perl can do the other thing, because variables
contain $@% stuff ;-) .

> >>> string.join(map(lambda i: chr(i), range(ord('a'), ord('k'))), '')
> 'abcdefghij'
> 
> Doesn't look very well, does it?

Ok, that's true :-)  I needed it only for numbers so far and thought
of something else yet :)

Stefan


From morse at harborcom.net  Fri Jan  7 08:08:04 2000
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Fri, 07 Jan 2000 13:08:04 GMT
Subject: Getting SWIG to work with C++/Python
References: <9mad4.42353$Ke.27158@news21.bellglobal.com> <854m6a$j4o$1@nnrp1.deja.com>
Message-ID: <3875e4c8.230810858@news.oh.verio.com>

>There is an _excellent_ manual for SWIG at www.swig.org.  SWIG is a
>rather advanced tool, and you really need to look at the manual to use
>it.
>
>I use SWIG a lot, I find it relatively easy to use, but I would not have
>gotten anywhere without the manual

I agree, SWIG has a rather flat learning curve, but it is well worth the effort.
In addition to the manuals, the people on the SWIG mailig list (on Egroups) have
been very responsive.  Don't give up!


From sp00fD at yahoo.com  Fri Jan  7 15:59:36 2000
From: sp00fD at yahoo.com (sp00fD)
Date: Fri, 07 Jan 2000 20:59:36 GMT
Subject: CGI problem
Message-ID: <855k3e$agm$1@nnrp1.deja.com>

I'm creating a CGI program, and I keep getting an error from apache:

"(8)Exec format error exec of /usr/local/apache/htdocs/...
Premature end of script headers: /usr/local/apache/htdocs/..

I've put the print "Content-type: text/html\n" (with anywhere from 0 to
8 newlines after it) in there.

What am I doing wrong?


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


From zakons at objectsbydesign.com  Thu Jan  6 21:05:59 2000
From: zakons at objectsbydesign.com (Stuart Zakon)
Date: Fri, 07 Jan 2000 02:05:59 GMT
Subject: bdb: which class?
References: <B9AA1D5DA95ED211900100805F6FFF9C02F3D8F6@exbhbhmb01.us.dbisna.com> <3874C7B9.106A5730@ndh.net>
Message-ID: <38754A4D.5D8D1389@objectsbydesign.com>


Stefan Schwarzer wrote:

>
> If I understand your question: What about
>
> class where_am_i:
>     def name( self ):
>         'return the name of the class to which self belongs'
>         return self.__class__.__name__

My class is derived from class Bdb in bdp.py. This is the Generic Python
debugger class. The code frame doesn't appear to have any reference to the class
where the method is running from.

Stuart Zakon



From hathawa2 at marshall.edu  Thu Jan 20 11:43:21 2000
From: hathawa2 at marshall.edu (Mark Hathaway)
Date: Thu, 20 Jan 2000 11:43:21 -0500
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim>
Message-ID: <388765BA.3FC8@marshall.edu>

>> [Sposhua]
>>
>> I'm one of the "prefers indentation to {}" ppl, and I'd
>> just like to add: whe the heck do we need colons (:) at
>> the end of if's 'n loops anyway?  The only loss I see is
>> to these 1-liners:
>>
>> if a==b: eric=idle
>>
>> What's wrong with:
>>
>> if a==b
>>       eric=idle

> Tim Peters wrote:
>
> As Michael Hudson said, the connection between colons and blocks was adopted
> from Python's closest predecessor, ABC.  Formal usability studies showed
> that while indentation was less confusing than assorted bracketing
> constructs to newbies, the combination of indentation and an introductory
> colon was even better.  You're right that the grammar has little use for it;
> although, pragmatically, it's a great aid to semi-stupid tools (like Python
> editor modes -- not to mention all us Python programmers <wink>).
>...
> Ambiguity, and Yet Another Special Meaning for Parens, were taken as bad
> omens.  So Guido exorcised those demons.

Ambiguity is indeed an important thing, and for both humans and the
software
which interprets the code. So, what's with the "elsif:" and "else:"
colon?

I can see leaving out the "end" closure for "if", "while" & "for"
statements
to ensure they can be executed cleanly as one-liners, but is it safe to
consider
the indent as (effectively) a keyword? Isn't it less ambiguous to use a
keyword
to ensure the author's meaning isn't lost. Surely a visible keyword is
less
likely to be destroyed than spaces. The author's intent with spaces
isn't
necessarily obvious, but an "end" is forever.  :-)

if <boolean-test>:
    <statement-list>
elsif
    <statement-list>
else
    <statement-list>
end

There's the keyword "if" at the beginning, the ":" after the boolean
test,
and the other keywords to separate the statement lists. Clean and
perfectly
unambiguous what is meant by the author.

while <boolean-test>:
    <statement-list>
end

for <iteration-stuff>:
    <statement-list>
end

There's no excessive use of the ":" and the unambiguity of using
the closure "end" is clear for all to see.

Also, to counter the argument of someone who gave a wild example of
using
the "end", but not enforcing the indentation rules; enforce the
indentation
rules AND use "end". It's even clearer that way.


From niels at endea.demon.nl  Tue Jan 18 17:02:16 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Tue, 18 Jan 2000 23:02:16 +0100
Subject: [Tkinter/Win32] Message box on app start -> mouse is dead
References: <87g0vv6wlc.fsf@deneb.cygnus.argh.org>
Message-ID: <3884E2E8.F3F6A83E@endea.demon.nl>


Florian Weimer schreef:
> tk = Tk()
> Button(tk, text="Exit", command=sys.exit).pack()
> tkMessageBox.showinfo("Test app", "Click on Ok, then on Exit.\n"
>                       + "Do you see the problem?")
> tk.mainloop()
> 
> (The application doesn't respond to mouse clicks after dismissing the
> message box.)

Displaying a message box outside the main loop seems to mess up tk's
message dispatching somehow. One way to work around this is to use a
separate Tk instance and throw it away after displaying the message:
tk = Tk()
Button(tk, text="Exit", command=sys.exit).pack()
xxx = Tk()
xxx.withdraw()
tkMessageBox.showinfo("Test app", "Click on Ok, then on Exit.\n"
                      + "Do you still see the problem?", master=xxx)
xxx.destroy()
tk.mainloop()


-- 
Niels Diepeveen
Endea automatisering



From Robin.Friedrich at pdq.net  Sun Jan 30 13:58:15 2000
From: Robin.Friedrich at pdq.net (Robin Friedrich)
Date: Sun, 30 Jan 2000 18:58:15 GMT
Subject: Choosing a language
References: <01bf6b45$ff345b40$412b69ce@default>
Message-ID: <389489C7.25327549@pdq.net>

You might check out HTMLgen. See... http://starship.python.net/lib.html
-- 
Robin K. Friedrich     Houston, Texas


From marc_risney at my-deja.com  Mon Jan 10 16:25:01 2000
From: marc_risney at my-deja.com (marc_risney at my-deja.com)
Date: Mon, 10 Jan 2000 21:25:01 GMT
Subject: Zope and NT databases
References: <387956BD.CFDF17DD@bby.com.au> <3879CE9B.BE77787C@bam.com> <3879E4E4.B4BD96FD@ipro.de> <387A10CE.E8CD408F@bellatlantic.net>
Message-ID: <85dimq$pi7$1@nnrp1.deja.com>

Hmmm, yes Zope is one way to do it, ODBC Socket Server,
is actually rather elegant, here is sample code that you could use
provided that you download the ODBC Socket Server
from http://odbc.linuxbox.com/products.html

Python 1.5.2 (#1, Jan  6 2000, 15:54:56)  [GCC egcs-2.91.66 19990314
(egcs-1.1.2
  on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>from socket import *
>>>import select

>>>FORWARD = 'ntbox.domain.com'
>>>PORT = 9628
>>>s = socket(AF_INET, SOCK_STREAM)
>>>s.connect(FORWARD, PORT)

>>>conn_str = "DSN=ODBCSS;UID=;PWD=;"
>>>sql_str = "insert into Table1 values ('test1','test2')"

>>>xml_str = """<?xml version=\"1.0\"?
>\r\n<request>\r\n<connectionstring>""" + conn_str
+ "</connectionstring>\r\n<sql>" + >>>sql_str
+ "</sql>\r\n</request>\r\n"
>>>s.send(xml_str)
>>>msg = s.recv(1024)
>>>s.close()
>>>print str(msg)


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


From gerrit.holl at pobox.com  Wed Jan  5 16:24:49 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 5 Jan 2000 22:24:49 +0100
Subject: TypeError: slice index must be int ?
In-Reply-To: <20000105111525.C1172@lexgen.com>; from hz@lexgen.com on Wed, Jan 05, 2000 at 11:15:25AM -0600
References: <20000105111525.C1172@lexgen.com>
Message-ID: <20000105222449.A5064@stopcontact.palga.uucp>

Hao Zhou wrote on 947067325:
> 
> Is it possible to use long int (say 2000000000L) to 
> slice a string with gigabytes size?

No... why do you want that? I think you'd better look for a better
solution.

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From R.Brodie at rl.ac.uk  Tue Jan 25 06:10:09 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Tue, 25 Jan 2000 11:10:09 -0000
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be>
Message-ID: <86k78t$thg@newton.cc.rl.ac.uk>

"Thomas Hamelryck" <thamelry at vub.ac.be> wrote in message news:86joac$l2s$1 at mach.vub.ac.be...

> BTW, people here seem to have adopted the idea that if you don't like
> a particular feature of a language, you should abandon it (e.g., "You
> don't like indentation? Use Perl!").

Adding features to a language always seems easy but that is the best
reason for not doing so. There is the temptation to make the language
the union of all the desired features. It's all been talked around before,
and Guido has said what his feelings are. In the end, I think you've
either got to respect that decision, or fork the language.

> You can find a very nice description of this process with respect to C++
> in "The Evolution of C++ : Language Design in the Marketplace of Ideas",
> by Jim Waldo (Editor), James Waldo (Editor).

C++ has to be the classic case of a language that evolved rather than
was designed: PDP-11 assembler with OO bits tacked on. If we can
learn something from its evolution, it should be not to do the same
again.

'Oh, that way madness lies; let me shun that.' - King Lear




From cfelling at iae.nl  Mon Jan 31 14:50:00 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 31 Jan 2000 20:50:00 +0100
Subject: Namespace pollution
References: <949332864.553076@news.adtranzsig.de>
Message-ID: <874p18$2ga$1@vvs.superst.iae.nl>

Dirk-Ulrich Heise <hei at adtranzsig.de> wrote:
...
> I've got one problem with all this stuff:
> I end up with dozens of helper functions that are
> called exactly one time. They pollute the namespaces
> of the module or class they're defined in. I hate polluted
> namespaces.

> How do others cope with this problem (that does not only
> apply to Python)?

You could switch to Literate Programming and use real functions when
appropriate but otherwise rely on the inline nature of chunks like this:

===================

Some chunks of code that would have been silly functions had I not
switched to LP but are now 'inlined'.
<<do something, only needed once>>=
  pass

Enhance your coding style and explain what's going on here on a meta level
<<do something else, only needed once>>=
  pass

Let's now apply those chunks once
<<global functions>>=
  def real_work:
    <<do something, only needed once>>
    <<do something else, only needed once>>

And just to be complete call upon [[real_work]] somewhere
<<main>>=
  <<global functions>>

  real_work()

=======================

Sorry, couldn't resist to plug my oldtime love, though I've never
used it on Python (yet). Somehow it feels like overkill for such
a readable language;} But it sure is a hassel, going back to edit -
compile - run cycli.
-- 
groetjes, carel


From insyte at emt-p.org  Tue Jan 25 23:15:42 2000
From: insyte at emt-p.org (Ben Beuchler)
Date: Tue, 25 Jan 2000 22:15:42 -0600 (CST)
Subject: vi and python?
In-Reply-To: <65118AEEFF5AD3118E8300508B124877073D3C@webmail.altiris.com>
Message-ID: <Pine.LNX.4.20.0001252214240.523-100000@marvin.squad51.net>

On Tue, 25 Jan 2000, Mike Steed wrote:

> Nothing too fancy, but here are the settings I have accumulated for *.py
> files.  The ^H and ^M are actually literal control characters.
> 
> " indentation
> set shiftwidth=4 softtabstop=4 tabstop=8 noexpandtabs
> set nocindent smartindent
> set cinwords=if,else,elif,for,while,def,class,try,except,finally
> set formatoptions=croq2
> 
> " disable dedent when '#' is hit
> inoremap # X^H#
> 
> " python versions of [[ and ]]
> map ]] /^[^ \t#]\+.*:^M
> map [[ ?^[^ \t#]\+.*:^M

Sounds good.  This may sound like a stupid question, but is there a way to
have it only use these options with .py files?  I haven't done too much
yet with customizing vim...

Ben

-- 
"There is no spoon"
	-- The Matrix




From amused at webamused.com  Wed Jan 19 22:07:26 2000
From: amused at webamused.com (Joshua Macy)
Date: Thu, 20 Jan 2000 03:07:26 GMT
Subject: __name__
References: <387F4F42.B9E1DB8E@kw.igs.net>
Message-ID: <388679C5.C338B212@webamused.com>

JB wrote:
> 
> Could anyone tell me how to get a classes name:
> 
> class MyClass:
>    def className( self ):
>        return self.__name__
> 
> This doesn't fly. The docs say a class has the __name__ attribute but...
> 

Class has a __name__ attribute, but in your example, self is an
instantiated object
of type MyClass, not a class.  To get the __name__ attribute you either
have to refer to it directly
(Class.__name__), or indirectly via the __class__ attribute of the
object (object.__class__.__name__)
E.g.

>>> class MyClass:
... 	def __init__(self):
... 		print self.__class__.__name__
... 
>>> m = MyClass()
MyClass
>>> print MyClass.__name__
MyClass
>>> print m.__class__.__name__
MyClass


  Joshua


From ivanlan at callware.com  Mon Jan  3 22:26:34 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Mon, 03 Jan 2000 20:26:34 -0700
Subject: OT: OpenSource Python Books?
References: <mt2.0-29451-946955341@news.informatik.uni-bonn.de>
Message-ID: <3871686A.600E1D17@callware.com>

Hi All--

Hmmm.  Someone's not paying attention here.  What does that say re the
quality of LinuxPorts.Com's publications?

<make-mine-vegetarian-spam-and-cage-free-eggs>-ly y'rs,
Ivan

poet at linuxports.com wrote:
> 
> Hello,
> 
> I represent LinuxPorts.Com (http://www.linuxports.com/) and I have been
> approached by a printer to possibly publish OpenSource books. To start
> these books would just be paper versions of the actual documentation.
> In this case, the Python Documentation and we were wondering if this
> would be of interest to the Python Community.
> 
> I spoke with the printer and we both have agreed that if we do this
> that we will be providing a portion of the Gross Profits back to the
> individual projects. In this case the Python project.
> 
> Anyway, if this of interest to you please go to
> http://www.linuxports.com/ and make your vote. It will help us
> determine the demand for such a book.
> 
> Thank you for your time,
> 
> Joshua Drake
> Webmaster Linux Documentation Project
> Owner LinuxPorts.Com
> poet.nospam at linuxports.com
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> 
> --
> ----------- 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
> ------------------------------------------------------------
> --
> http://www.python.org/mailman/listinfo/python-list

-- 
----------------------------------------------
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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From pinard at iro.umontreal.ca  Mon Jan 17 14:16:29 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Jan 2000 14:16:29 -0500
Subject: xpot version 0.3
In-Reply-To: Bernhard Herzog's message of "16 Jan 2000 14:48:21 +0100"
References: <oq66wv5a9h.fsf@titan.progiciels-bpi.ca> <m3ln5qp1p6.fsf@greebo.nodomain.de>
Message-ID: <oqln5o4igi.fsf@titan.progiciels-bpi.ca>

Bernhard Herzog <herzog at online.de> ?crit:

> >    http://www.iro.umontreal.ca/contrib/po/xpot-0.3.tar.gz

> This should be
>      http://www.iro.umontreal.ca/contrib/po/xpot/xpot-0.3.tar.gz
> I think.

You are right, sorry.  I'm repackaging `xpot' within something I will call
`po-utils', and which will contain a few other, simpler tools.  (I hesitated
between `po-utils' and `tp-utils', because I have some tools in the works
which are related to translation without being specifically related to PO
files, but this is too soon to consider them yet.)  I'll announce 0.4 soon.

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





From mwh21 at cam.ac.uk  Sat Jan 22 16:44:37 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jan 2000 21:44:37 +0000
Subject: Python dumps core with this.. how come?
References: <000001bf651d$f05ac240$132d153f@tim>
Message-ID: <m31z79wzlm.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Interrupting Michael Hudson's fine advice to inject a nit]
> 
> [Roey Katz]
> > this brings up another concern. Given
> > the following definition:
> >
> >    class A:
> >
> >       def __getattr__( self, name ):
> >         return -1
> >       def __setattr__( self, name, val ):
> >          self.__dict__[ name ] = val
> >
> > I see that self.__dict__ must be a special variable
> > in that setting it somehow avoids a recursive call
> > to __setattr__()?
> 
> [Michael]
> > Yup. Look in Objects/classobject.c around about line 580
> > if you want the gory details.
> 
> While true, it doesn't apply to Roey's example:  he's not setting __dict__.
> He's binding a key *of* self.__dict__; self.__dict__ itself is merely
> referenced.  So Roey should really be wondering why the reference to
> self.__dict__ isn't invoking __getattr__ (&, of course, that's also due to
> special-case __dict__ magic, but in __getattr__'s implementation).

Actually I noticed this; line 580 is in instance_getattr1.

> I don't mean to be overly picky; it's just that the excruciating details are
> vital when mucking with __getattr__/__setattr__, and *any* small
> misconception about them will eventually bite hard.

Oh yes, that much is true.

[schnipp]

> Yes, pointing him to the most bizarre endcases of Python behavior will
> surely aid him in "resetting his brain" <wink>.

Sorry. I get carried away in situations like that. As you might well
know, if you remember where the bytecodehacks started...
 
> Guido gave some excellent semi-exasperated advice (similar in spirit to
> parts of Michael's advice that I've snipped) many moons ago, when the whole
> newsgroup was temporarily filled with msgs of this character:  he invited
> people to just *try* writing a Python program without using any __xxx__
> hooks.  I expect some people were amazed at how much easier their lives got!
> If he were to get that exasperated again today, I bet he'd expand his
> invitation to include avoiding map, filter, reduce and lambda too.

I'd expect so to; most of the Python code that I ever had trouble
understanding used map & co extensively. Rewriting it to use Python
loops definitely helps.

> you're-not-ready-to-use-them-in-python-before-you're-
>     happy-without-them-ly y'rs  - tim

What follows is pure opinionated speculation (or drivel, if you like).

Python makes it soooo easy to write readable code that I think it's
harder than in many languages to spot good practise. In C++ or Lisp or
Haskell or many other languages, if your logic's getting confused it's
generally obvoius (becuase the code has become nauseatingly hard to
read). Python is more subtle.

Things I find that almost uniformly improve code readability and
reliability include:

writing short functions
using Python loops
avoiding lambdas
never *ever* using "import *"
using classes where I might use a fake closure
remebering that I'm programming Python, and that using an idiom from
  another programming language is *not* necessarily a good idea

I'm gradually forming the opinion that writing *really* good code in
Python is no easier than in any other language, it's just that writing
merely very good code is much easier.

thanks-for-listening-ly y'rs
Michael



From tim_one at email.msn.com  Fri Jan 21 04:38:12 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 21 Jan 2000 04:38:12 -0500
Subject: Python Rocks!
In-Reply-To: <oqpuuwb7bg.fsf@titan.progiciels-bpi.ca>
Message-ID: <001501bf63f3$3c018ac0$ec2d153f@tim>

[Tim]
> Programs-writing-programs is a twist of mind conspicuously
> absent in the Python world.

Note that, in context, this meant Python programs transforming Python
programs into other Python programs.  The equivalent is quite common in
Lisp, but may be more common in Fortran than Python!  That whole class of
techniques just never caught on here; most Pythoneers take pride in finding
ways to avoid using even "exec".

[Fran?ois Pinard]
> ...
> Isn't there a need for compiling grammars into fast scanners
> and parsers, a bit like Flex and Bison do for C people?

Yes, but that's really a different topic.  The Python world has no good
approach to that now, paying attention to the "fast" part, and where "good"
means "enough like Flex and Bison that you don't feel you've been stranded
on some strange alien planet" <wink>.

> To be happy, I presume one would have to write small amounts
> of Python code for each "action".  Or maybe just provide the
> name of a function to call.  Yet, the result has to be Python.
>
> Or are you suggesting that such tools should directly
> generate byte-code?

Hmm -- while I don't *believe* I was suggesting anything, I'm *sure* I
wasn't suggesting that <wink>.

better-to-write-python-wrappers-for-lex-&-bison-ly y'rs  - tim





From echuck at mindspring.com  Thu Jan  6 05:13:33 2000
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Thu, 06 Jan 2000 05:13:33 -0500
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net> <N6Vc4.610$hJ1.7143@news.rdc1.tx.home.com> <slrn878b1i.k6b.kc5tja@garnet.armored.net>
Message-ID: <38746ACD.4DFA4B62@mindspring.com>

Thanks to everyone who has replied to my inquiry.

-Chuck


"Samuel A. Falvo II" wrote:

> >> Seems like many web host providers, such as MindSpring, offer CGI with
> >> Perl, but not other languages, notably Python.
> >>
> >> Does anyone know who would host a CGI Python site?
>
> Armored Internet (http://www.armored.net) supports Zope and Python CGIs.
> Zope is supported under our Unix web hosting packages, and Python CGIs are
> supported under both Unix and Windows NT web hosting packages.
>
> --
> KC5TJA/6, DM13, QRP-L #1447
> Samuel A. Falvo II
> Oceanside, CA



From scarblac-spamtrap at pino.selwerd.nl  Mon Jan 31 06:38:24 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 31 Jan 2000 11:38:24 GMT
Subject: Choosing a language
References: <01bf6b45$ff345b40$412b69ce@default> <slrn8994sf.ndn.scarblac-spamtrap@flits104-37.flits.rug.nl> <6f7l4.14957$3b6.61050@ozemail.com.au>
Message-ID: <slrn89asrd.ool.scarblac-spamtrap@flits104-37.flits.rug.nl>

Jason Stokes wrote in comp.lang.python:
> >There are two complaints about Python as a first language that I've heard:
> >1) Basically all the documentation is in English. Not a problem for you.
> 
> Python isn't the only language that has a documentation gap like that.  The
> whole programming world is incredibly English-centric, and most European
> programmers seem to be able to speak English.

Well, there are books in Dutch for many more popular languages, like
C and Perl. I haven't seen any on Python yet. A Dutch web page is being
made (http://python.nl.linux.org). 

Yes, the ones that use Python fanatically already know English. It's the
people who can't program yet who have a problem with learning a lot of
info from English text. And I want non-programmers to learn programming,
in Python :-).

I suppose I should start helping the Dutch web page project...
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
"This gubblick contains many nonsklarkish English flutzpahs, but the
 overall pluggandisp can be glorked from context"  (David Moser)


From scarblac-spamtrap at pino.selwerd.nl  Mon Jan 31 12:46:38 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 31 Jan 2000 17:46:38 GMT
Subject: list -> dict help
References: <3895BE15.A7CF569B@bam.com>
Message-ID: <slrn89bidu.p04.scarblac-spamtrap@flits104-37.flits.rug.nl>

Bill Scherer wrote in comp.lang.python:
> What's the best way for me to do this:
> 
> say I have a list:
> 
> >>> l = ['a', 'b', 'c', 1]
> 
> and I want to turn it into a dictionary that looks like this:
> 
> >>> print d
> {'a': {'b': {'c': 1}}}
> 
> where the length of list l is >= 3.
> 
> Basically, list[:-2] represents the nested keys, list[-2] is a terminal
> key, and list[-1] is the value.
> 
> It seems I need somthing to recur over list[:-2], then add the terminal
> info but I can't figure it out...

The first thing I thought of was recursion. Something cute like:

def convert(l):
    if len(l) == 1:
	   return l[0]
	else:
	   return { l[0] : convert(l[1:]) }
	   
But this copies the list every call, and recursion can use a lot of stack.
Since it's tail recursion, there must be an iterative version.

The following one works, but I don't really like it. References
are weird in cases like this. I hope it's understandable.

def convert(list):
    d = { list[0] : None } # Initialize with the first key
    lastdict = d       # Remember which dict to change and with which key,
    lastkey = list[0]  # so the 'None' can be replace in the loop.
    
    for key in list[1:-1]:
        subdict = { key : None }
	    lastdict[lastkey] = subdict
	
	    lastdict = subdict
	    lastkey = key
	
    # Add the last key
    lastdict[lastkey] = list[-1]
	
    return d
	
	
-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From aahz at netcom.com  Fri Jan 21 17:12:12 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 22:12:12 GMT
Subject: Python Rocks!
References: <200001211334.IAA28640@python.org> <86ajlq$661@news.or.intel.com>
Message-ID: <86aljs$a9$1@nntp9.atl.mindspring.net>

In article <86ajlq$661 at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>
>Yes, I think I'll let you guys wallow in your ignorance.

Huzzah!  Three cheers for the jester!
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From loewis at informatik.hu-berlin.de  Thu Jan 20 08:27:31 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Jan 2000 14:27:31 +0100
Subject: Suggest: .get() method on lists.
References: <20000118212229.A542@stopcontact.palga.uucp>
Message-ID: <p6qvh4oc1q4.fsf@informatik.hu-berlin.de>

Gerrit Holl <gerrit.holl at pobox.com> writes:

> I thought it would be useful, for example, for sys.argv. At the moment,
> I have this code:
> 
> file = sys.stdin
> if len(sys.argv) > 1:
>     file = sys.argv[1]
> 
> But this, would be much easier to read:
> file = sys.argv.get(1, sys.stdin)

I guess it wouldn't be that useful in this specific case, as
sys.argv[1] if present is a string, so you (probably) write

file = sys.stdin
if len(sys.argv) > 1:
    file = open(sys.argv[1])

right now, to open the file. I can't see how you could get the
equivalent of that code with .get.

Regards,
Martin




From tjg at avalongroup.net  Wed Jan 19 22:27:58 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Wed, 19 Jan 2000 19:27:58 -0800
Subject: Mapping Column Names to results using DBI
References: <38863BEB.BCB9E15C@exceptionalminds.com> <865enm$kkt$1@nntp4.atl.mindspring.net>
Message-ID: <388680BE.EB94E957@exceptionalminds.com>

Aahz Maruch wrote:
 
> Try "SELECT x, y, z" instead of "SELECT *".

I knew it couldn't be that simple...

I must be missing something

c = db.cursor()

c.execute('select supplierid, customerid, carrierid from brokerage')

a = c.fetchall()

now what?

a['supplierid'] and a[supplierid] both don't work...

Thought I had an epiphany there for a minute, but nor does
c.execute('select supplierid, customerid, carrierid from brokerage',
(a,b,c))

So, I will gladly accept any other enlightenment you care to provide!

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From gerrit.holl at pobox.com  Mon Jan 31 10:12:39 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 31 Jan 2000 16:12:39 +0100
Subject: [Newbie Problems] AttributeError from CGI script?
In-Reply-To: <000501bf6b79$52a62be0$482d153f@tim>; from tim_one@email.msn.com on Sun, Jan 30, 2000 at 06:25:41PM -0500
References: <20000130.14431716@mis.configured.host> <000501bf6b79$52a62be0$482d153f@tim>
Message-ID: <20000131161239.B902@stopcontact.palga.uucp>

Tim Peters wrote on 949253141:
> [David Fisher]
> > ...
> > Along with most/all of the other functions in string.py, it is
> > replaced at the end of the module by another function named replace
> > from the strop module, which is written in C.  Now why this is done
> > is a complete mystery to me.
> 
> Hysterical raisins:  originally, string.py was "it", and there was no strop
> module.  What's there today was a way to replace the important string
> functions w/ much faster C versions, without breaking any old code.  The
> string.py functions also serve as a kind of reference implementation for the
> curious, or for an initial port to a weird machine.
> 
> those-who-don't-remember-the-past-are-doomed-to-
>     be-confused-by-its-ripples-ly y'rs  - tim

Am I supposed to remember Python facts that happened when I was 5 or 6?
I'm not.

But neither I am doomed to be confused: Archives! NEWS file!
Together quite a lot, I think. I'll come back in 40 years, when
I've finished reading it all.

farewell,
Gerrit.

-- 
Please correct any bad Swahili you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From emile at fenx.com  Wed Jan  5 18:22:56 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 5 Jan 2000 15:22:56 -0800
Subject: control structures (was "Re: Sins")
Message-ID: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>

Doesn't this do that now? or is there some objection to this usage?

for i in range(3):
  try:
    elem = "... nested loops & stuff ..."
    if i < 2:
      raise "element found", elem
    "... more stuff ..."
    raise "search failed"
  except "element found", e:
    print "Found", e
  except "search failed":
    print "Not Found"
  print 'looping on i=%d' % i
print 'e-o-f'


Found ... nested loops & stuff ...
looping on i=0
Found ... nested loops & stuff ...
looping on i=1
Not Found
looping on i=2
e-o-f
Process completed successfully

maybe-I'm-missing-the-obvious?-ly y'rs

Emile van Sebille
emile at fenx.com
-------------------



Evan Simpson <evan at tokenexchange.com> wrote in message
news:<s76uc120hu870 at corp.supernews.com>...
> Steve Holden <sholden at bellatlantic.net> wrote in message
> news:387358BB.E71EC4E7 at bellatlantic.net...
> > Phil Jensen wrote:
> > >     loop until "element found" or "search failed":
> > >         ...
> > >         if blah1 == blah2:
> > >             "element found"
> > >         ...
> > >     when "element found":
> > >         ...
> > >     when "search failed":
> > >         ...
>
> I proposed a spelling for this a while back, and have even made
sporadic
> forays into the Python source to see if I can implement it myself.  It
> extends 'try', and would look like this:
>
> try:
>     ... nested loops & stuff ...
>            break "element found", elem
>     ... more stuff ...
>         break "search failed"
> continue "element found", e:
>     print "Found", e
> continue "search failed"
>     print "Not Found"
>
> > Typically one would use enumerations rather that strings to
> > represent the loop states, to combine readability with efficiency.
>
> In this syntax, the strings are treated as labels (despite being
string
> literals grammatically) and are purely static, for both efficiency and
> error-checking.  That is, "break STRING, exprlist" is a compile-time
> SyntaxError unless the string exactly matches the string in a
"continue"
> clause of an enclosing "try" and the expression list is
> assignment-compatible with the argument list of that same clause.
Once this
> matching is done, the string is discarded and the "break" converted
into an
> expression evaluation followed by a Jump.  "continue" clauses compile
into a
> Jump to the end of the "try" followed by an assignment LHS (the target
of
> the "break" Jump) and the suite body.
>
> In many ways this is quite similar to raising an exception and
catching it
> in an "except" clause.  The major difference is that everything is
totally
> resolved at compile time, and is lexically fixed.
>
> > > Whatever the loop is, its exit or exits have some "meaning"
> > > in the program, and this control structure encourages an
> > > element of Literacy by forcing them to be named.
>
> Exactly.  With no penalty in speed or (compiled) code size, in this
case.
>
> Cheers,
>
> Evan @ 4-am
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>
>





From gmcm at hypernet.com  Mon Jan  3 11:25:39 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 3 Jan 2000 11:25:39 -0500
Subject: Python on NT systems (help a Unix guy, please)
In-Reply-To: <3870CA2B.DD09F50E@bellatlantic.net>
Message-ID: <1265206654-25488231@hypernet.com>

Steve Holden writes:

> Don't suppose anyone has implemented freeze() on NT, have they? I
> only have the binary 1.5.2 on my NT box, and I don't see it there
> :-(

It's in the source distribution.

I have a compiler-less alternative for Windows:
http://starship.python.net/crew/gmcm/install.html


- Gordon



From gerrit.holl at pobox.com  Sat Jan 22 15:17:16 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 22 Jan 2000 21:17:16 +0100
Subject: 7/8 bit email (Was: Re: OT: Uitgelijnde email verzenden)
In-Reply-To: <20000122154811.A6625@stopcontact.palga.uucp>; from gerrit.holl@pobox.com on Sat, Jan 22, 2000 at 03:48:11PM +0100
References: <001101bf6445$64d51370$43c7a8c0@tentacleloco.nl> <3888B8E4.77B50FA2@kub.nl> <20000122154811.A6625@stopcontact.palga.uucp>
Message-ID: <20000122211715.A9667@stopcontact.palga.uucp>

Gerrit Holl wrote on 948552491:
[cut Dutch chatter]

*sigh*. *again* the wrong list :-(.
I have to watch out better in the future.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From Svante.Kleist at edt.ericsson.se  Thu Jan 20 02:03:53 2000
From: Svante.Kleist at edt.ericsson.se (Svante Kleist)
Date: Thu, 20 Jan 2000 08:03:53 +0100
Subject: How to differentiate Win98 from NT ?
Message-ID: <MPG.12f0d1c642257995989691@news.ericsson.se>

Dear Pythonists,

"os.name" says "nt" for both Windows NT and Windows 98.

In order to differentiate between the two,
I could search for the existence of "cmd.exe",
but that feels rather clumsy. 

Better suggestions, please?

 / Svante Kleist
/NEMESIS systemDesign, Stockholm


From tjg at avalongroup.net  Fri Jan 14 05:38:27 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Fri, 14 Jan 2000 02:38:27 -0800
Subject: Stubbing function
Message-ID: <387EFCA3.94C67567@exceptionalminds.com>

Is there a "standard" method of stubbing functions in Tkinter?

I would like to be able to use the "command=lambda:????" construct to
stub my functions as I don't like creating functions with empty bodies
just for stubbing purposes. I tried "command=lambda: pass" and that
generated syntax errors, so I know that I'm missing something basic
about how lambdas work, I just don't quite know what!

Thanks for your assistance.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From garry at sage.att.com  Fri Jan 14 10:22:18 2000
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Fri, 14 Jan 2000 15:22:18 GMT
Subject: Python Pre-processor (like php/asp)
References: <387D7349.DCF6567A@sprint.ca> <rSif4.15$jw3.712@wagner.videotron.net>
Message-ID: <387F3F2A.89CBFEE5@sage.att.com>

Patrice Girard wrote:
> 
> Yes, ASP do the job.  Just to install Pythonwin and put SCRIPT=python inside
> your .ASP and it's running!

is this limited to microsoft web servers/clients?

-- 
Garry Hodgson			Every night 
garry at sage.att.com		a child is born
Software Innovation Services	is a Holy Night.
AT&T Labs			 - Sophia Lyon Fahs


From thomas.heller at ion-tof.com  Thu Jan  6 10:07:46 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 6 Jan 2000 16:07:46 +0100
Subject: py2bat (was: How do I make a Python .bat executable file?)
References: <38722971.4B036062@mchp.siemens.de> <uembv316c.fsf@cs.uu.nl>
Message-ID: <004e01bf5857$cac3eca0$4500a8c0@thomasnotebook>

> You must be aware of the sad fact that the Windows "Command line
> interpreter" mangles some arguments, notably those with an = sign in it.
> This will be replaced by a space. Sometimes this can be avoided by quoting
> the argument, but if you need both quotes and = signs to be passed in
> arguments you are lost.
>

You should also be aware of the fact that you cannot redirect
output from a batch file. (I've heard this bug is fixed in Win2000)

> For perl there is a runperl.exe which, if copied to myscript.exe and run,
> calls perl myscript.pl with all arguments passed unchanged. It would
> probably be simple to clone this for python. I have never seen the source,
> however, and I am too lazy to write it myself.

A better solution would be to use Gordon's installer

http://starship.python.net/gmcm/

Regards

Thomas Heller




From wweston at worldnet.att.net  Fri Jan 21 13:35:55 2000
From: wweston at worldnet.att.net (wes weston)
Date: Fri, 21 Jan 2000 18:35:55 +0000
Subject: Install problem WNT4.0 & TCL
References: <3887C64F.AB67C453@vsl.com.au>
Message-ID: <3888A70B.59C333DB@worldnet.att.net>

Martin,
	I had the exact same problem. To find the fix, I did a find on *.dll on
the tcl tree. This yielded about six files. Then I located them in
winnt/system32, or was it in system. Anyhow, it turned out to be
tk80.dll. I removed tk80.dll from system32 and all worked.
			wes

Martin Kew wrote:
> 
> Hi,
> 
> I have just installed Python a WNT4.0 machine and are having problems
> running 'idle'.  When I try to use the Idle icon the machine processed
> for a while and then returned with out a message.  I then attempted to
> run Idle manually and got the following message.  If I new a bit more
> about Tcl I would try to fix it but I figure that you might be
> interested in the default install not working.
> 
> 
> C:\Python>python c:\Python\Tools\idle\Idle.py
> Traceback (innermost last):
>   File "c:\Python\Tools\idle\Idle.py", line 3, in ?
>     PyShell.main()
>   File "C:\Python\Tools\idle\PyShell.py", line 611, in main
>     root = Tk()
>   File "C:\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__
>     self.tk = _tkinter.create(screenName, baseName, className)
> TclError: Can't find a usable init.tcl in the following directories:
>     {} ./lib/tcl8.0 C:/tcl8.0/library C:/library
> 
> 
> 
> This probably means that Tcl wasn't installed properly.
> 
> C:\Python>
> 
> I also tried
> 
> 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 _tkinter
> >>> import Tkinter
> >>> Tkinter._test()
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "C:\Python\Lib\lib-tk\Tkinter.py", line 1947, in _test
>     root = Tk()
>   File "C:\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__
>     self.tk = _tkinter.create(screenName, baseName, className)
> TclError: Can't find a usable init.tcl in the following directories:
>     {} ./lib/tcl8.0 C:/tcl8.0/library C:/library
> 
> 
> 
> This probably means that Tcl wasn't installed properly.
> 
> >>>
> 
> 
> Noting that I installed using py152.exe and changed the install
> directories to c:\Python and c:\Tcl. and also included c:\Python and
> c:\Tcl\bin in an effort to fix the problem. Can anyone help ?
> 
> 
> --
> 
> Regards
> 
> Martin Kew
> 
> Vision Abell Pty Ltd
> Second Avenue, Technology Park
> Mawson Lakes SA 5095 Australia
> Tel +61 8 8300 4544
> Fax +61 8 8349 7420
> email Martin.Kew at vsl.com.au
> http://www.vsl.com.au
>


From skip at mojam.com  Tue Jan 25 10:00:26 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 25 Jan 2000 09:00:26 -0600 (CST)
Subject: anydbm - a simple question
In-Reply-To: <388CE28F.D3A100AB@inka.de>
References: <20000121110715.A184793@vislab.epa.gov>
	<3diu0lbz8e.fsf@amarok.cnri.reston.va.us>
	<20000124112320.A440571@vislab.epa.gov>
	<388CE28F.D3A100AB@inka.de>
Message-ID: <14477.47754.65118.177514@beluga.mojam.com>

    Michael> Same thing on my system (SuSE 6.3 Linux). The database is
    Michael> created (default is dbhash in my case). I can re-open it with
    Michael> module dbhash but not with anydbm (same error as mentioned
    Michael> above). The whichdb module fails on determining the DB type.

This comes up from time-to-time.  It is a fixed problem.  You can either
grab the CVS version of the entire shootin' match or visit

    http://www.musi-cal.com/~skip/Python/

and browse into the Lib directory and grab whichdb.py.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From reio-ta at MailAndNews.com  Tue Jan  4 18:57:54 2000
From: reio-ta at MailAndNews.com (jerry smith)
Date: Tue, 4 Jan 2000 18:57:54 -0500
Subject: what if i dont want a 32 bit int?
Message-ID: <38752609@MailAndNews.com>

what if all i want is really just an 8 bit or 16 bit number?
so if i do this

i=0x80
i=~i
print '%x' %i
ffffff7f

that is too many leading f's so i have to have an anding mask
of  i & ff? is there any alternative? to select how much space i
want to dedicate for a python object to be stored as?
also a special thanks goes out to Eugene for his help from before :)

------------------------------------------------------------
 Get your FREE web-based e-mail and newsgroup access at:
                http://MailAndNews.com

 Create a new mailbox, or access your existing IMAP4 or
 POP3 mailbox from anywhere with just a web browser.
------------------------------------------------------------



From nobody at nowhere.nohow  Sun Jan 23 19:44:10 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 24 Jan 2000 00:44:10 GMT
Subject: text-screen
References: <86cckj$2bgf$1@black.news.nacamar.net>
Message-ID: <slrn88n82q.cbl.nobody@isis.visi.com>

In article <86cckj$2bgf$1 at black.news.nacamar.net>, Roland Kruggel wrote:

>i'm a new a new python user. i am develop my programs in informix and change
>now to python. i thing python is bette. i write programms text-screen
>oriented. wat for modules i need to programm? who can help?
>many thanks
>roland...

There a curses module, and there is also a module that uses the "newt"
text widget library that runs on top of the slang screen manipulation
library.  That module (can't remember the name) is in included in
RedHat Linux distributions, but not in the standard Python dist.

-- 
Grant Edwards                   grante             Yow!  Not SENSUOUS... only
                                  at               "FROLICSOME"... and in
                               visi.com            need of DENTAL WORK... in
                                                   PAIN!!!


From digitome at iol.ie  Mon Jan 17 13:49:13 2000
From: digitome at iol.ie (Sean Mc Grath)
Date: Mon, 17 Jan 2000 18:49:13 GMT
Subject: using threads with python and winnt
References: <1264075593-4560755@hypernet.com>
Message-ID: <38835f4c.1380925@news.iol.ie>

>Kreiner Karl or Charly or somebody wrote:
>
>>             does anyone have expiriences with using threads and winnt ?
[...]
<gmcm at hypernet.com> wrote:
>Threads work dandy on NT
[...]

Does anybody know how good/bad/indifferent the threading stuff works
on Windows 95/98? I am considering using threads for an app
under development but I'm a bit worried (read "terrified") about
the issues that might pop up in '95 and '98.

regards,
Sean McGrath
 


From michael.stroeder at inka.de  Mon Jan 24 18:38:55 2000
From: michael.stroeder at inka.de (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Tue, 25 Jan 2000 00:38:55 +0100
Subject: anydbm - a simple question
References: <20000121110715.A184793@vislab.epa.gov> <3diu0lbz8e.fsf@amarok.cnri.reston.va.us> <20000124112320.A440571@vislab.epa.gov>
Message-ID: <388CE28F.D3A100AB@inka.de>

Randall Hopper wrote:
> 
> It created the database.dir and database.dat files in the write phase,
> but the read phase fails on the open.
> 
> Thinking it might be extension related I tried changing the file-to-open
> from "database" to "database.dat".  But no dice:
> 
>    anydbm.error: db type could not be determined

Same thing on my system (SuSE 6.3 Linux). The database is created
(default is dbhash in my case). I can re-open it with module dbhash
but not with anydbm (same error as mentioned above). The whichdb
module fails on determining the DB type.

Ciao, Michael.


From michael.stroeder at inka.de  Mon Jan 10 12:58:01 2000
From: michael.stroeder at inka.de (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 10 Jan 2000 18:58:01 +0100
Subject: starting a cgi with params
References: <iQp6OHyH9j2y64kYNrpGPcKoXKu0@4ax.com>
Message-ID: <387A1DA9.77EB7011@inka.de>

Anders M Eriksson wrote:
> 
> How do I send 2 params to the script when I'm using it as an CGI
> script?

Read about module cgi in the docs of your friendly Python
distribution.

Ciao, Michael.


From tseaver at starbase.neosoft.com  Fri Jan 28 22:26:15 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 28 Jan 2000 21:26:15 -0600
Subject: Case Sensitivity and Learnability
Message-ID: <B4191059635E81B1.8AFDD386EFACF4A3.6FF161EA0151B5EB@lp.airnews.net>

Thinking about the notion of case sensitivity as a barrier to CP4E, as
advocated quite glibly at IPC8, I typed up this module on the plane to
explore the consequences of allowing/encouraging users to enter case-
mangled code, especially in non-interactive uses (I think that distinc-
tion may be important, because readability is almost insignificatn in
an interactive session).

# begin randomCase.py

""" Scramble the case of standard input.

    This module attempts to rebut the argument that case insensitive code
    intrinsically more "usable" for J. Random Novice than case sensitive
    code.  The flaw in this argument is that it confuses "typability" with
    "usability" -- while a case-insensitive language may be less daunting
    for the fledgling programmer to type, code which actually _uses_ case
    insensitivity is MUCH harder to read.  Precisely as with natural
    language, learning to read code is essential to learning to write code;
    code which is harder to read frustrates this process.

    I would therefore argue that addressing the "typability" issue by
    modifying the language is the Wrong Thing -- the place to fix this is in
    the tools, through the equivalent of an incremental spell checker.
"""
import sys
import string
import whrandom

def identity( c ):
    return c

flippers = {}

for c in string.uppercase:
    flippers[ c ] = string.lower( c )

for c in string.lowercase:
    flippers[ c ] = string.upper( c )

def flipCase( c ):
    return flippers.get( c, c )

#
#   This set of transformation functions perturbs the stream pretty severely.
#
HEAVY_TRANSFORMS = ( identity
                   , identity
                   , identity
                   , identity
                   , identity
                   , identity
                   , string.upper
                   , string.upper
                   , string.upper
                   , string.lower
                   , string.lower
                   , string.lower
                   )

#
#   Perhaps this is more reasonable -- one alpha character in eight flips its
#   case?
#
SAMPLE_TRANSFORMS = ( identity
                    , identity
                    , identity
                    , identity
                    , identity
                    , identity
                    , identity
                    , flipCase
                    )

def randomCase( istream=sys.stdin
              , ostream=sys.stdout
              , transforms=SAMPLE_TRANSFORMS
              ):
    """
        Replace random characters in a istream using the functions supplied
        in transforms.
    """
    for line in istream.readlines():
        mangled = []
        for c in line:
            mangled.append( whrandom.choice( transforms )( c ) )
        ostream.write( string.join( mangled, '' ) )

if __name__ == '__main__':
    randomCase()

# end randomcase.py

Now observe the effect of running it on itself:

""" Scramble thE case of standard inpuT.

    tHis moduLe attempts to rebut the argumeNt that case insensiTive code
    intrinsically more "usable" for J. Random Novice than case sensITive
    code.  The flaw in this Argument iS THAt it confuSes "typabilIty" with
    "usAbIlity" -- while a case-inSensitive language may be less dauntIng
    For the fledgLing programmEr tO tyPe, cOde which actuALly _Uses_ case
    insEnsitivItY is MUCH hArdER TO Read.  PreciseLy as wIth natural
    Language, learning to read COde is essential to lEarNing to write code;
    code which is harder to reaD fRUstrates thiS proceSs.

    I would tHerefore argue thaT addrEssing thE "typAbility" isSue by
    MoDiFying The laNguage Is the WRoNg Thing -- thE place to fIx this is in
    the tools, thrOugh the equiValent of a spell checker.
"""
import sys
import string
impoRT whrandom

Def identity( c ):
    return c

flipPerS = {}

for C iN sTring.uppercase:
    flippers[ c ] = strIng.lower( c )

For c in string.lOwercaSe:
    flippErs[ c ] = string.upper( c )

def flipCaSe( c ):
    reTUrn flippers.get( c, c )

#
#   This seT of transforMation fUnctions perturbs the strEam pRetty severeLy.
#
HEAVY_TRaNSFoRmS = ( iDeNTiTy
                   , identiTy
                   , identiTy
                   , identity
                   , IdentiTy
                   , idenTitY
                   , STrIng.uPper
                   , striNg.upper
                   , stRing.upper
                   , string.LoWer
                   , strinG.lower
                   , strinG.lower
                   )

#
#   Perhaps this IS more reasonaBle -- oNe alpha chaRactER in eight flips its
#   casE?
#
SAMPLE_TRANSFORMS = ( identity
                    , identity
                    , identity
                    , identity
                    , identity
                    , ideNtity
                    , IdentiTy
                    , flipCasE
                    )

def randomCase( iStreaM=syS.stdin
              , ostream=syS.stdout
              , tRanSfOrms=SAMPLE_TRaNSFOrMS
              ):
    """
        ReplAce random charaCterS in a istream Using ThE functionS SuPpliED
        in Transforms.
    """
    For line in istrEam.readLines():
        mAnGled = []
        for c in liNe:
            maNgled.Append( wHrANdom.choice( transfoRms )( c ) )
        ostream.write( string.join( mangled, '' ) )

if __name__ == '__main__':
    randomCase()
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From gerrit.holl at pobox.com  Mon Jan 24 09:30:25 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 24 Jan 2000 15:30:25 +0100
Subject: If you can only buy *one* Python book...
In-Reply-To: <86hkc6$mh4@newton.cc.rl.ac.uk>; from R.Brodie@rl.ac.uk on Mon, Jan 24, 2000 at 01:33:56PM -0000
References: <N%rg4.4360$3b6.20735@ozemail.com.au> <20000116182502.A1106@better.net> <t6tg4.305$iL4.4241@news.rdc1.tx.home.com> <20000117204759.B1291@stopcontact.palga.uucp> <86hkc6$mh4@newton.cc.rl.ac.uk>
Message-ID: <20000124153025.A2566@stopcontact.palga.uucp>

Richard Brodie wrote on 948717236:
> > > The 'Python Essential Reference' is a good addition to the online
> > > documentation as well.
> >
> > Disagreed. I think the online reference is far the best and most
> > complete there is.
> 
> That really depends on how much you like online documentation,
> and/or online development. I often find the presence of a computer
> a distraction when coding, and consumes to much real estate on
> the desk. It's pretty much personal taste - if your first reaction on
> seeing HTML docs  is to look for the PDF/PS version to print then
> PER should be your second stop after 'Learning Python'.

The online reference is printed as well.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From jhauser at ifm.uni-kiel.de  Mon Jan  3 12:19:32 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 03 Jan 2000 18:19:32 +0100
Subject: Considering using python for something, and need a bit of info :)
References: <0.3903b60a.25a2245f@aol.com>
Message-ID: <874scvqdi3.fsf@ifm.uni-kiel.de>

As a common reply in these days take a look a zope :-). Although it's
a web thing somebody ( Evan Simpson <evan at 4-am.com>) has written a
product to it, so that a restricted part of python can be used for
scripting the web application server (Zope), thus exposing a reduced
part of the complete python interpreter which is running.

He used bytecodehacks. This is all very involved, but perhaps a start.


Here is a link for PythonMethods:
http://www.zope.org/Members/4am/PythonMethod

And another one for bytecodehacks.
http://starship.python.net/crew/mwh/
ftp://starship.python.net/pub/crew/mwh/bytecodehacks-doc-$(VERSION).tar.gz

HTH __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 martindvorak at my-deja.com  Fri Jan 28 17:16:36 2000
From: martindvorak at my-deja.com (martindvorak at my-deja.com)
Date: Fri, 28 Jan 2000 22:16:36 GMT
Subject: existence of a variable
Message-ID: <86t4fu$s6u$1@nnrp1.deja.com>

Hi,

What is the best way to determine existence
of a variable in Python (I need to check
if a variable is defined and if so,
delete it)?

Thanks for your opinions.

Martin


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


From robin at jessikat.demon.co.uk  Mon Jan 31 11:34:36 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 31 Jan 2000 16:34:36 +0000
Subject: shelve implementation
Message-ID: <nuVkmAAcmbl4EwGe@jessikat.demon.co.uk>

can someone explain why shelve uses StringIO as an intermediate rather
than doing the pickles direct

ie shelve uses
        def __getitem__(self, key):
                f = StringIO(self.dict[key])
                return Unpickler(f).load()

        def __setitem__(self, key, value):
                f = StringIO()
                p = Pickler(f)
                p.dump(value)
                self.dict[key] = f.getvalue()


whereas I would have done
        def __getitem__(self, key):
                return loads(self.dict[key])

        def __setitem__(self, key, value):
                self.dict[key] = dumps(value)

-- 
Robin Becker


From fdrake at acm.org  Thu Jan  6 13:17:40 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 6 Jan 2000 13:17:40 -0500 (EST)
Subject: no return values for __init__ ??
In-Reply-To: <3874CFE5.765A40EA@mdlink.de>
References: <3874CFE5.765A40EA@mdlink.de>
Message-ID: <14452.56388.442949.799702@weyr.cnri.reston.va.us>

Helge Hess writes:
 > recently I was heavily wondering why __init__ does not allow return
 > values ! Is there any special reason not to do this ?

Helge,
  Yes:  It doesn't make sense.  The job of the __init__() method is to 
initialize the instance which has already been created.  There is no
way for a Python program to receive the return value.
  To implement the singleton pattern you appear to be interested in,
try this:

        _C_singleton = None

        class _C:
            ...

        def C():
            global _C_singleton
            if _C_singleton is None:
                _C_singleton = _C()
            return _C_singleton

 >   c1 = C()
 >   c2 = C()
 >   'if c1 is c2' will be true

  This will still be true, and the code will more rationally reflect
what's going on.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From tim_one at email.msn.com  Mon Jan 24 20:27:04 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jan 2000 20:27:04 -0500
Subject: License Question: Embedding Python
In-Reply-To: <388c274e@news.access.net.au>
Message-ID: <001201bf66d3$4a5cedc0$a42d153f@tim>

[Mark Klink]
> I want to embed Python in a commercial application (binary
> only) ...
> The Python License requires that the copyright and permission
> notices in the original code appears in supporting
> documentation.  Thus, the documentation will include the
> following text:
> "Permission to use, copy, modify, and distribute this software
> and its documentation for any purpose and without fee is
> hereby granted, provided that the above copyright notice
> appear in all copies and that both that copyright notice and
> this permission notice appear in supporting documentation ..."
>
> Could you help me understand how I can satisfy the spirit of
> the Python License requirements without comprising the
> commercial nature of the product or confusing the copyright of
> the overall application where we do not give permission to
> copy, modified and distribute the overall application?

I suspect Usenet is the worst place on earth to seek legal advice <0.5
wink>.

There's certainly no GPL-ish "copyleft" *intent* in the Python license (see
http://www.python.org/doc/Copyright.html for confirmation).  At worst,
what's wrong with stuffing the quoted text into the docs, indented, and
preceded by a paragraph explicitly saying that what follows applies only to
the Python subsystem?  If that worries you too, you had better see an actual
lawyer <wink>.  The true intent of the license appears to be to prevent CWI
and CNRI from getting sued "no matter what".

the-number-of-lawsuits-involving-python's-license-so-far:0-ly
    y'rs  - tim





From mhammond at skippinet.com.au  Tue Jan  4 19:37:42 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 05 Jan 2000 00:37:42 GMT
Subject: pythonwin installation
References: <218A08577EF@nds-3.azg.nl>
Message-ID: <qnwc4.9760$lv5.46540@news-server.bigpond.net.au>

This generally means you need a COM update.  If you visit my starship
pages at http://starship.python.net/crew/mhammond you should find some
links to the Microsoft site where you can get the COM updates...

Mark.
"P.J.W.S. Vrijlandt" <P.J.W.S.VRIJLANDT at INT.azg.nl> wrote in message
news:218A08577EF at nds-3.azg.nl...
> Hi all,
>
> When installing pythonwin (from win32all-125) I first saw this
> message about 3 times. But pythonwin did install and I can start it.
>
> It seems I can't do anything like using running makepy &c.
> This is what happens:
>
> PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit
>  (Intel)] on win32 Copyright 1991-1995 Stichting Mathematisch
> Centrum, Amsterdam Portions Copyright 1994-1999 Mark Hammond
> (MHammond at skippinet.com.au)
> >>> import pythoncom
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
> ImportError: DLL load failed: Een apparaat dat op het systeem is
> aangesloten, werkt niet.
> >>>
>
> Since Dutch is your 2nd language (as in another thread), you all
> understand that the DLL does not load because "a device connected to
> the system, does not work". In the configuration screen everything
> seems fine.
>
> Any ideas?
>
> I'm running Win95 osr 2 on a pentium with 32 mb. Does pythonwin have
> any configuration requirements?
> I used all the default installation options.
> Python and Idle/Tkinter work fine.
>
> TIA,
>
> Patrick Vrijlandt
>




From gerrit.holl at pobox.com  Wed Jan  5 05:02:37 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 5 Jan 2000 11:02:37 +0100
Subject: using python to press keys
In-Reply-To: <3872C5EA.D004E77D@trojanslair.zzn.com>; from themantis@trojanslair.zzn.com on Tue, Jan 04, 2000 at 08:17:47PM -0800
References: <3872C5EA.D004E77D@trojanslair.zzn.com>
Message-ID: <20000105110237.A2374@stopcontact.palga.uucp>

Black Mantis wrote on 947013467:
> Can you use python to "press down" keys? If so, how would you do that?

Under Unix, use the curses module. See the curses HOWTO on www.python.org.
Onder Windows, there is some other module. My signature should state
'clearly' why I don't know about that.

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From karl-sve at dsv.su.se  Tue Jan 11 11:08:08 2000
From: karl-sve at dsv.su.se (Karl Svensson)
Date: Tue, 11 Jan 2000 17:08:08 +0100
Subject: Filenames of cgi-uploaded files
In-Reply-To: <Pine.WNT.4.21.0001101735310.249-100000@wks_3.e-shopwork.com>
References: <Pine.WNT.4.21.0001101735310.249-100000@wks_3.e-shopwork.com>
Message-ID: <Pine.LNX.4.10.10001111440010.8926-100000@l137.dsv.su.se>

Hello!

On Mon, 10 Jan 2000, Sposhua wrote:
> Uploading a file from a form - I need to know the filename (to ignore all
> non-gif/jpg uploads) and filesize. Can't find anything in the tutorials on
> theis :-( Then I guess I upload the pic using
> 
> form = cgi.FieldStorage
                         ^
dont forget the (), should be cgi.FieldStorage()...

> fileitem = form['picture']
> if fileitem.file:
> 	pic_upload = fileitem.file.read()
> 
> (adapted from the tutorial, though it doesn't make much sense to me -
> expecially the fileitem.file bit)

When in doubt, consult the source... :)

In the docstring of the FieldStorage class in cgi.py is:

"... it has the following attributes:
...
filename: the filename, if specified ...
...
file: the file(-like) object from which you can read the data;
   None if the data stored is a simple string ..."

So

form = cgi.FieldStorage()
fileitem = form['picture']
# if image is gif or jpg
if fileitem.file and ((string.find('.gif',fileitem.filename) != -1) or \
(string.find('.jpg',fileitem.filename) != -1)):
    # read file data to memory
    imagefile_data = fileitem.file.read()
    # to get filesize
    imagefile_length = len(imagefile_data)

might work.
I haven't tested it, so I'm not sure...
The string.find bits could surely be more elegantly written with regex,
but I don't know how, I'm quite new to Python myself. Please tell me of
any errors.

HTH

/Kalle Svensson, just another guy...



From aahz at netcom.com  Sun Jan  9 10:56:29 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Jan 2000 15:56:29 GMT
Subject: fastest way to merge lists.
References: <000201bf5a6c$1a133f20$f62d153f@tim>
Message-ID: <85ab3d$h4q$1@nntp3.atl.mindspring.net>

In article <000201bf5a6c$1a133f20$f62d153f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>Can't beat it, asymptotically speaking -- it's a linear-time algorithm.
>Practically speaking, Python's dicts are highly optimized too.  You can save
>a little typing via e.g.
>
>for s in list1, list2, list3, list4, list5:
>    for x in s:
>        d[x] = 1

In practice, I know that the integer "1" is a cached, unique object, but
that's an implementation detail that I prefer not to rely on.  I
therefore almost always would use "d[x] = None"; that's particularly
true if I *might* want to attach data to it later, in addition to making
it clear that I'm interested only in the dictionary keys.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From rwgk at my-deja.com  Wed Jan 12 05:02:23 2000
From: rwgk at my-deja.com (rwgk at my-deja.com)
Date: Wed, 12 Jan 2000 10:02:23 GMT
Subject: Perl speed vs. Python speed
Message-ID: <85hjfa$mlp$1@nnrp1.deja.com>

I just wrote a tiny Python program to do some text processing.
When I ran it I was very surprised how long it took to finish.
My first guess is that Perl would be a lot faster, maybe even a
factor 10, but I did not do a direct comparison.

Could someone with more experience comment on Perl speed vs.
Python speed?

Thanks!
        Ralf


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


From tim_one at email.msn.com  Thu Jan  6 21:29:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 6 Jan 2000 21:29:15 -0500
Subject: SETL links (was RE: Python not a Very High-Level Language?)
In-Reply-To: <3874CE06.9A434C20@lanl.gov>
Message-ID: <000901bf58b6$fe5f16c0$932d153f@tim>

[William B. Clodius]
> ftp://cs.nyu.edu/pub/languages/setl2/
>
> appeears to be the most likely source

[and later]
> Another version of SETL, for UNIX systems only, is available at
>
> http://birch.eecs.lehigh.edu/~bacon/download-setl.html
>
> See also
>
> http://birch.eecs.lehigh.edu/~bacon/
>
> and
>
> http://www.CS.NMSU.Edu/~complog/sets/implem.html

Thanks, William!  In addition, the following variation for Windows appears
to have been used (at least) to develop mathematics courseware in the mid
90's:

http://csis03.muc.edu/isetlw/isetlw.htm

can't-keep-a-good-idea-down?-ly y'rs  - tim





From fdrake at acm.org  Fri Jan 28 10:36:02 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 28 Jan 2000 10:36:02 -0500 (EST)
Subject: Python online docs are great!
In-Reply-To: <461182F0B9B7D111B4000008C7282ADC22B0C5@nynt04.umny.scripps.com>
References: <461182F0B9B7D111B4000008C7282ADC22B0C5@nynt04.umny.scripps.com>
Message-ID: <14481.46946.329807.372735@weyr.cnri.reston.va.us>

Kolosov, Victor writes:
 > How do we persuade him to do the same to Zope documentation which really
 > needs it.

  I'm not familliar enough with Zope to do much there, and changing
*that* would be fairly difficult.
  Seriously, what I'm doing often is not *creating* documentation from 
scratch but integrating contributed material, providing some editorial 
consistency, and supporting the conversions to usable output formats.
When I create new content it's often in response to specific questions 
or messages stating that something in the documentation is missing,
misleading, or wrong.  To get good documentation for Zope, someone
needs to start by writing bad documentation for Zope, and then handle
requests for improvements and integration of contributed material.
It's not a full-time job, but can be very disruptive of things which
are.  ;)  And the initial bad documentation can be time-consuming to
create as well; I was lucky someone beat me to that!  (Ok, it wasn't
all bad; Guido's original docs were some of the best I'd seen for free 
software at the time, and still would compare very well.)


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From eddy at vortigen.demon.co.uk  Mon Jan 17 20:07:15 2000
From: eddy at vortigen.demon.co.uk (Edward Welbourne)
Date: Tue, 18 Jan 2000 01:07:15 +0000 (GMT)
Subject: Suggestions for python 2
In-Reply-To: <388380E0.36AB1146@maxtal.com.au> (message from skaller on Tue,
	18 Jan 2000 07:51:45 +1100)
References: <m129zpL-00006JC@vortigen.demon.co.uk>
	            <388328F8.85147F85@maxtal.com.au> <200001171837.SAA20866@lsls4p.lsl> <388380E0.36AB1146@maxtal.com.au>
Message-ID: <m12AN75-00007lC@vortigen.demon.co.uk>

> It hadn't occured to me you wouldn't know the arguments.
Heh.  Well, only one of us is a functional programmer.

> I always know the positional arguments (not necessarily the
> optional ones) -- there's no other way to call a function :-)

In python, some functions take `as many arguments as the user choses to
supply'.  I may wish to take such a function, bind in an initial
sub-sequence of those arguments, yet still allow the result's callers to
supply arbitrarily many following.

An implementation of currie which can't cope with *args isn't pythonic.

	Eddy.
--
Liberty's a glorious feast.



From TheBlueWizard at pressroom.com  Thu Jan 20 20:17:53 2000
From: TheBlueWizard at pressroom.com (The Blue Wizard)
Date: Fri, 21 Jan 2000 01:17:53 GMT
Subject: Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us> <200001171159.LAA17397@lsls4p.lsl> <200001171403.JAA10668@eric.cnri.reston.va.us>  <m12AizP-00006lC@vortigen.demon.co.uk>  <200001190040.TAA18277@eric.cnri.reston.va.us>
Message-ID: <01bf63ad$2a940920$565008d1@TheBlueWizard>

Guido van Rossum <guido at CNRI.Reston.VA.US> wrote in article
<200001190040.TAA18277 at eric.cnri.reston.va.us>...
> Eddy,
> 
> Have you read my (very old) metaclasses article?
> 
> It seems to me that where you write
> 
> > Body Sun:
> >     """The Sun: the star at the heart of the solar system."""
> >     type = 'G2 V'
> >     magnitude = 4.79
> >     aliases = 'Sol',
> 
> I would write
> 
> class Sun(Body):
>     """The Sun: the star at the heart of the solar system."""
>     type = 'G2 V'
>     magnitude = 4.79
>     aliases = 'Sol',
> 
> I.e. only the syntax at the top is slightly different, the rest is the
> same.  The metaclass is defined in a similar way too; where you define
> __call__(*bases, **what), I define __init__(name, bases, what).
> 
> There are some details that are different, but in general it seems
> that your proposal is of similar power.  Have you considered this?
> 
> --Guido van Rossum (home page: http://www.python.org/~guido/)

Um...out of curiosity, why create a *class* for specific objects (e.g. Sun)
rather than
making instances?  But I can see the syntactical niceties here, almost as
if it is a
mix of class and instance stuff.  Something to think about.

Cheers,

The Blue Wizard



From boud at rempt.xs4all.nl  Sat Jan  1 17:30:12 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 1 Jan 2000 22:30:12 GMT
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com>
Message-ID: <84lv5k$rqv$1@news1.xs4all.nl>

randy_shaffer at my-deja.com wrote:
> Hello,

> The softswitch application lives on Unix, as does the testing harness
> implementation at this time.  The developers who are my customers will
> be using the test harness via their NT workstations, which means that
> the test harness gui will be served up by a soft Xserver (Exceed).

If using a Windows X server is no problem, then you've got an enormous
choice. PyQt/PyKDE, WxWindows, PyGTK, PyGnome, FxPy, PyFLTK... There are
a lot of others, e.g. All of them are free,  quite good, quite easy to
use and very nice looking. PyQT, WxPython and some other also appear to
work on Windows - take a look at

  http://www.vex.net/parnassus/apyllo.py/808292924.247038364
	
for more options ;-). Be prepared to spend a few days evaluating them,
if you're taking it serious.

PyQt has the disadvantage that you need to spend money if you want to
run it on Windows - PyKDE and PyGnome obviously don't run on Windows at
all. WxPython depends on a lot of other stuff - WxWindows, and GTK or
Motif on Unix, and it can be tough to install everything; more difficult
still to demand from users that they go the same route. I don't really
like the widgets offered by Fox, on which FxPy is based - they look a
bit old-fashioned. 

I'm quite fond of the clean object model of PyQt, with the signals and
slots for _really_ easy message/event based development, but for some
applications the Gnome canvas might be a godsent (caveat - I've only
read about that canvas, never used it myself, but it sounds good).

Going over to JPython might be another route - I'd say, try it out.

-- 

Boudewijn Rempt  | http://denden.conlang.org


From ovazquez at sprint.ca  Thu Jan 13 01:40:09 2000
From: ovazquez at sprint.ca (Orlando Vazquez)
Date: Wed, 12 Jan 2000 22:40:09 -0800
Subject: Python Pre-processor (like php/asp)
Message-ID: <387D7349.DCF6567A@sprint.ca>

Hi I'm looking for a Python equivalent of php or asp. I like their basic
ideas but I dislike the languages. Is there any module or set of modules
for this purpose? (I've seen one on the snippets site but it's kind of
bare)

return "Thanks" * 100

--
Orlando Vazquez


From PClaerhout at CREO.BE  Tue Jan 11 10:11:14 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Tue, 11 Jan 2000 16:11:14 +0100
Subject: Sys.argv
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA7048@MSGEURO1>

Hello,

I have an application I created using Gordon McMillan's installer
and I made an exe file of it. What I want to do know, is to catch
the files the user drops on the icon of the application.

If the app is not started this is no problem and can be done by
accessing sys.argv. I now want to catch the same thing when the
app is already running. Any ideas on how to do that??

Kind regards,


Pieter

Pieter Claerhout - PClaerhout at creo.be
Response Center - Applications Support

Creo Europe - Excelsiorlaan 21 - 1930 Zaventem - Belgium
Tel: +32 (2) 711 14 00 - Fax: +32 (2) 720 96 71




From robin at jessikat.demon.co.uk  Wed Jan 26 19:02:05 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 27 Jan 2000 00:02:05 +0000
Subject: test
Message-ID: <qdLXKDA9r4j4Ewkt@jessikat.demon.co.uk>

just testing
-- 
Robin Becker


From gchiaramonte at ibl.bm  Mon Jan 31 10:32:32 2000
From: gchiaramonte at ibl.bm (Gene Chiaramonte)
Date: Mon, 31 Jan 2000 11:32:32 -0400
Subject: Choosing a language
In-Reply-To: <5650A1190E4FD111BC7E0000F8034D26A0F2A3@huina.oceanic.com>
Message-ID: <LPBBLPGHGGJMCPPOKONLEEJGCCAA.gchiaramonte@ibl.bm>

The only trouble you will have if you start with python is when you try to
learn other languages. I think you will find them cumbersome. You will
always be thinking, "I could do this with 5 lines of Python code, jez!"


> -----Original Message-----
> From: python-list-admin at python.org
> [mailto:python-list-admin at python.org]On Behalf Of Doug Stanfield
> Sent: Sunday, January 30, 2000 2:48 PM
> To: 'Patrick Cooper'; python-list at python.org
> Subject: RE: Choosing a language
>
>
> Python is an excellent first language (or second or fourteenth for that
> matter). I found putting together some cgi programs for the first
> time easy
> enough.  I might have found doing cgi programming as my first Python
> programs very frustrating though.
>
> I'd suggest if you aren't familiar with the programming process start by
> working through as many tutorials as you can. Follow the links on
> the Python
> home page <http://www.python.org/>, --> 'Introductions to Python'
> <http://www.python.org/doc/Intros.html> 'Introductions to Python
> programming
> for non-programmers' where there are three currently listed.  I
> consistently
> hear good things about Alan Gaulds tutorial for the complete beginner.
>
> I find the greatest strength of Python to be the interactive interpreter.
> In essence it is a sandbox to play in to try things out.  There
> is no other
> way to learn to program than to write programs, but writing them can be
> frustrating if you can't get into the middle of the execution and find out
> what those commands you wrote are really doing.  The interpreter
> is the tool
> for that.  Almost without fail if I can't make a program work I'll try a
> sequence of commands in the interpreter that show me the problem.  While
> you're learning to program, this instant feedback can make all the
> difference in speeding you to comprehension.
>
> Trying to jump right into programming cgi puts a couple of layers between
> you and the results.  I think the only effective way to do it as
> a beginner
> would be to have the interpreter available to try out and verify your
> programming ideas before you try to call them using a web browser.
>
> When you do get started learning, another great resource is the
> Python Tutor
> mailing list.  Check out
<http://www.python.org/psa/MailingLists.html#tutor>
to subscribe.  There you can get good direction on the how to program
questions as well as the Python questions.

I'd suggest you spend two weeks to a month absorbing all you can before you
start trying to do the cgi stuff.  When you do, a good resource is
<http://starbase.neosoft.com/~claird/comp.lang.python/web_python.html> which
among other things contains a link to 'Writing CGI Programs in Python'
<http://www.devshed.com/Server_Side/Python/CGI/> which answers most of your
'since I want to focus on CGI, what would be the best resources to check
out' question.

Good luck,
-Doug-

> -----Original Message-----
> From: Patrick Cooper [mailto:future at dotstar.net]
> Sent: Sunday, January 30, 2000 7:20 AM
> To: python-list at python.org
> Subject: Choosing a language
>
>
> I want to start learning computer programming and from what I
> have found so
> far I have chosen to start with Python.  I am going to be
> focusing on CGI
> first but eventually want to work with desktop applicates.
> Python seems to
> be the most versitle and easiest to learn.  Before I go in
> any more depth
> does anyone think I should start with a different language?  I want to
> eventually learn C, but think Python is the best place to start.
>
> Also, since I want to focus on CGI, what would be the best
> resources to
> check out?  Anything besides what's listed in  the CGI Tools
> section of the
> Python web site?
>
>
>
>   -----------== Posted via Newsfeeds.Com, Uncensored Usenet
> News ==----------
>    http://www.newsfeeds.com       The Largest Usenet Servers
> in the World!
> ------== Over 73,000 Newsgroups - Including  Dedicated
> Binaries Servers ==-----
> --
> http://www.python.org/mailman/listinfo/python-list
>

--
http://www.python.org/mailman/listinfo/python-list




From grant at nowhere.  Thu Jan 20 14:05:25 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 20 Jan 2000 19:05:25 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <slrn886oku.2ac.grant@grante.comtrol.com> <38871EA1.8AF33147@sage.att.com> <slrn88ebbe.67b.grant@grante.comtrol.com> <867hui$6to$1@nntp3.atl.mindspring.net>
Message-ID: <slrn88en3l.7aj.grant@grante.comtrol.com>

In article <867hui$6to$1 at nntp3.atl.mindspring.net>, Aahz Maruch wrote:
>
>>Then he complained that it was too much work supporting
>>POP/IMAP, and he wanted everybody to use Outlook.  People
>>resisted (particularly the Unix users, most of whom left
>>shortly thereafter).  The fight went all the way to the
>>president of the company, and the decision was made that
>>everybody use Outlook.  (Why is management always so big on
>>monotony and so down on individuality?)
>
>Everything you've said has made sense except for one thing: why
>are you still working there?  I know jobs are relatively easy
>to find in Minneapolis, particularly for techies.

The design projects I'm working are pretty interesting.  I've
been given considerable lattitude in the product design area --
even to the point of using Gnu tools under Linux an open source
RTOS instead of something like Nucleus or (god forbid) WinCE on
the embedded project I'm working on.  There are even paid-for
commercial development tools and an RTOS license sitting unused
on the shelf.  So, I can put up with the Outlook issue (as long
as I get to coplain about it once in a while).

The network operation in general is way better here than my
previous employer.  We've got a T1 to a good ISP and a firewall
that is pretty much invisible when you're inside looking out.
My previous employer had a fireway that was really difficult to
live with.  And, that firewall was in Austin, Texas at the
other end of a network held together with 56k leased lines. On
top of that, they used cc:Mail with a badly broken
ccMail<-->SMTP gateway runnin on an IBM mainframe in
Marshalltown, Iowa.  Us Unix users just sort of lived "under
the radar" except when we needed to send e-mail with a ccMail
user: a one-way trip through the e-mail gateway took somewhere
between 10 minutes and 2 days.

Besides, I use an outside e-mail account for 90% of my mail
anyway (I probably only get 3 or 4 e-mails a week that I
currently have to deal with in Outlook).  We all need windmills
with which to tilt...

-- 
Grant Edwards                   grante             Yow!  YOW!! I am having
                                  at               FUN!!
                               visi.com            


From NoSpam at linkamericacorp.com  Thu Jan  6 20:17:06 2000
From: NoSpam at linkamericacorp.com (-james)
Date: Thu, 6 Jan 2000 19:17:06 -0600
Subject: Now What? :o)
Message-ID: <n2bd4.64$%Ph.18874529@news.randori.com>

My Y2K resolution is to get out of the Microsoft only rutt that I'm in and
get LINUX running on an extra box I've got, and learn a new non-microsoft
language. I've looked around, and I really like what I see in python.  I've
got a bunch of newbie questions and I appologise in advance if you've gone
over these before:

I'll be working with the language on NT and LINUX boxes; Are there any
issues with writing code to run on both platforms that I should be aware of
?

I'm almost through the online tutorial, and I'm hooked. I don't see any info
on coding GUI  forms though. Can anyone recommend a website, sample code, or
a good book that covers how to do this? Are GUI forms a platform dependant
issue? I'm coming from a world of drag and drop and, from what I gather,
this done programatically in python.

I saw some conflicting articles on Amazon.com about source code On said
python source must always be a text file; another one mentioned that python
source code can kinda be compiled to byte code (forgive me if I'm off on
this) and ran using a Java virtual machine. Is this true? Or are text files
my only choice? How about pickle & unpickle? Where do they  fit in to this
picture?

I would greatly appreciate *ANY*  words of wisdom, advice, or encouragement
you wish to graciously bestow upon me regarding these question or regarding
questions you know I should be asking but did not.

Thanks In advance for you help and kindess,


-james

(To email me directly simply replace NoSpam in my address with jbragg)





From s.schwarzer at ndh.net  Thu Jan  6 14:24:39 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 06 Jan 2000 20:24:39 +0100
Subject: Naming conventions
Message-ID: <3874EBF7.8317091D@ndh.net>

Hi,

are there common naming conventions for Python (functions,
classes, variables ...)?

>From the code I have seen, I suppose the following naming styles:

variable   this_is_a_variable
class      ThisIsAClass (or This_is_a_class?)
function   this_is_a_funtion
module     ThisIsAModule, thisisamodule (or what else?)

Have I forgotten something; do you have more remarks or hints?

Stefan


From gcutrupi at ams.greenpeace.org  Thu Jan  6 06:47:54 2000
From: gcutrupi at ams.greenpeace.org (Gilberto Cutrupi)
Date: Thu, 06 Jan 2000 12:47:54 +0100
Subject: JOB VACANCY - sysadmin for Greenpeace.org
Message-ID: <387480E8.6CDAF942@ams.greenpeace.org>

The environmental organisation Greenpeace International is looking for a
new system administrator for our external website
 (http://www.greenpeace.org. We are committed to use open source
software therefore the new system administrator will work on our Linux
 server and will help in the implementation of a content management
system for this platform.

 The candidate's requirements are:
 - 2-3 years Linux systems administration experience
 - Apache administration
 - Perl/cgi-bin programming skills
 - Python programming skills
 - SQL Database experience
 - PHP experience is an advantage

 - Full job description available under request (e-mail to
gcutrupi at ams.greenpeace.org).
 - Deadline for applications: January 25th 2000
 - The location is Amsterdam, The Netherlands

 Contact for final applications: lhatton at ams.greenpeace.org

Regards,
Gilberto Cutrupi



From bitbucket at isomedia.com  Thu Jan  6 20:19:55 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Fri, 07 Jan 2000 01:19:55 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com>  <005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <slrn87ad9u.hp9.neelk@brick.cswv.com>
Message-ID: <38753ef4.441952444@news.isomedia.com>

On 7 Jan 2000 00:37:29 GMT, neelk at brick.cswv.com (Neel Krishnaswami)
wrote:

>Darrell <darrell at dorb.com> wrote:
>> From: "Skip Montanaro"
>> >
>> > I find nothing wrong with it.  Some people object to using try/except for
>> > anything but handling errors however.
>> 
>> Don't know if this if fair. But exceptions are slow when raised.
>
>Random question: why?
>
>When I tried timing a dummy loop with 100,000 exceptions versus 100K
>function calls, I found that the exceptions took roughly 3 times
>longer. (On CPython 1.5.2)
>

Random postulate - could it have to do with the creation of the
exception object?  _I'm_ sure not qualified to know, test, or examine
this answer, so it's up to the next person...

      -Eugene
import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From fredrik at pythonware.com  Fri Jan 21 03:48:41 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 09:48:41 +0100
Subject: regex question
References: <roy-835F11.21172620012000@netnews.nyu.edu>
Message-ID: <01eb01bf63ec$532be300$f29b12c2@secret.pythonware.com>

Roy Smith <roy at popmail.med.nyu.edu> wrote:
> Let's say I've got a pattern "[0-9]", i.e. a 1-digit number.  I want to 
> search a string for that pattern, and only return a match if it's found 
> preceeded by either a non-digit or the beginning of the string, and 
> likewise followed by either a non-digit or the end of the string.  Is 
> there a single compact regex I can write to cover that?
> 
> For example:
> 
> pat = re.compile (pattern)
> x = pat.search ("1")
> y = pat.search ("12")
> z = pat.search ("the number 1, don't you know?")
> 
> x and z should be MatchObjects, but y should be None.
> 
> The best I can come up with would be something like:
> 
> (^[0-9][^0-9])|([^0-9][0-9][^0-9])|([^0-9][0-9]$)|(^[0-9]$)
> 
> which is fairly ugly.

umm.  speaking as the "official 1.6 regex implementor", I'd
go for a simple expression, and filter out any hits that are
not 12 digits.

haven't tried it, but I have a gut feeling that the resulting
code will be faster than any possible alternative...

(I'm happy to be proven wrong, of course)

</F>




From warlock at eskimo.com  Fri Jan 21 14:22:03 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Fri, 21 Jan 2000 11:22:03 -0800
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com>  <865lsm$s3p@news.or.intel.com>
Message-ID: <slrn88hcer.404.warlock@raven.myth>

On Wed, 19 Jan 2000 16:43:43 -0800, 
 tye4, in the persona of <tye4 at yahoo.com>,
 brought forth the following words...:

>
>Ivan Van Laningham <ivanlan at callware.com> wrote in message >
>
>> Don't code that way to start with.  It's dumb.  Treat yourself to a good
>> editor.
>>
>Better yet. Fix the dumb language.

I like it as it is, if you don't, why are you using it?

>
>
>> You got about as much chance of getting this into Python as you do of
>> getting hit by a meteor.  Less, actually, because the odds for the
>> latter are measurable.  What you're asking for is for Python to not be
>> Python.  You don't like it, design your own language, build the
>> interpreter or compiler, and publish it.  Maybe you can get some people
>> to use it.  One of them won't be me.
>>
>> You're saying to people who know and love Python, "Python is a great
>> language except that it's Python.  Let's fix it."  We don't think it's
>> broken.  Your insistence on swimming upstream on this issue is not only
>> wrong-headed but boring.
>>
>Python is good but could be better.
>Everything has flaws and nothing is perfect. If we can't fix it then its a
>different
>matter. But in this case the problem can be fixed. It just so happens that
>there
>are a lot of bull-headed close-minded people than I reckoned there would be.
>I'm pretty certain now that this problem will never be fixed.


because so few people think it's a 'problem'. Frankly, to me.
it's a solution...

>
>I suggest you go to other newsgroups and let them know how brilliant Python
>is
>because it has an invisible block terminator.
>
>Since we are so lazy let's just skip the ending anything and just use
>indentation.
>
>Math:
>Before:
>    (2 + 3) * 5
>
>After Python rule:
>    (2 + 3
>            * 5
>
>C:
>Before:
>    if (x == 5)
>    {
>        printf("blah\n");
>    printf("blah blah");
>
>HTML:
>Before:
>    <h1> Python indentation: <i>Why it is such a pain</i>
>   </h1>
>    Blah blah blah
>
>After:
>    <h1> Python indentation
>            <i> Why it is such a pain
>    Blah blah blah
>
>Go ahead and convince these guys to switch do it the "Python Way".
>And Good luck. You will need it.
>
>-tye4

You are comparing apples and oranges. HTML is not a programming
language, it is a (seriously warped) markup language, by
definition, it doesn't even have any indentation rules. (just
look at some of the code written by some of the clicky pointy
webpage "builders" for examples of that...)
 Again, you think it's a problem, others don't. Deal with it. 


-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From a.eyre at optichrome.com  Mon Jan 31 08:45:15 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 13:45:15 -0000
Subject: IOError: (0,"Error") when closing "os.popen"ed file?
In-Reply-To: <m3n1pmmlse.fsf@devel.no.rwd.dinet>
Message-ID: <000701bf6bf1$672bb030$3acbd9c2@peridot.optichrome.com>

> After hacking for a while, I managed to work around the problem by
> doing:
> 
>     time.sleep(0.1)
>     f.close()
> 
> But this is highly unsatisfactory..... Is there anyone that can give me 
> a better suggestion?

I expect the OS hasn't finished with the file before it receives the
close().

Try replacing "time.sleep(0.1)" with "f.flush()". IIRC flush() blocks
untils the file operation has completed.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From smigowsky at dspace.de  Wed Jan 19 02:01:59 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Wed, 19 Jan 2000 08:01:59 +0100
Subject: WIN32 : how to list the current processes by their executable
	 name?
Message-ID: <326536345498D311B3BC00105A39802A074589@newsintern.dspace.de>

Hi, 

stating from the mails of Mark Hammond and Robin Becker :

>In article <862936$g33$1 at nnrp1.deja.com>, MHammond at skippinet.com.au
>writes
>>In article <CxKdRDAwDEh4Ewzj at jessikat.demon.co.uk>,
>>  Robin Becker <robin at jessikat.demon.co.uk> wrote:
>>
>>> this fails for me with win95.
>>
>>Unfortunately, win32pdh only works on NT/2000.  I dont know how to do
>>the same thing on Win95/98...
>>
>>Mark.
>Well there's the famous demos of PS and Kill which are claimed to work;
>certainly they use different methods depending on the system. Kill
>doesn't always quite do it on Win95.
>-- 
>Robin Becker

Will probably answer your question

Stefan




From skaller at maxtal.com.au  Mon Jan 17 15:51:45 2000
From: skaller at maxtal.com.au (skaller)
Date: Tue, 18 Jan 2000 07:51:45 +1100
Subject: [PSA MEMBERS] Re: [Types-sig] Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk>
	            <388328F8.85147F85@maxtal.com.au> <200001171837.SAA20866@lsls4p.lsl>
Message-ID: <388380E0.36AB1146@maxtal.com.au>

Edward Welbourne wrote:
> 
> >>   * certain tools the functional programmers crave, notably a truly
> >>     faithful implementation of currie()
> >
> >       not required: python lambdas subsume what can be done with
> >       currying.
> 
> Unfortunately, unless I'm missing something, this is only true when you
> know the call-signature of the function you are packaging.

> Can you show me a piece of valid python (using lambdas or otherwise)
> which achieves this effect ?

It hadn't occured to me you wouldn't know the arguments.
I always know the positional arguments (not necessarily the
optional ones) -- there's no other way to call a function :-)

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia voice: 61-2-9660-0850
homepage: http://www.maxtal.com.au/~skaller
download: ftp://ftp.cs.usyd.edu/au/jskaller



From jhefferon at my-deja.com  Wed Jan 26 15:59:05 2000
From: jhefferon at my-deja.com (jhefferon at my-deja.com)
Date: Wed, 26 Jan 2000 20:59:05 GMT
Subject: CGI: keeping mySQL awake between hits?
Message-ID: <86nn6m$tkg$1@nnrp1.deja.com>

I have CGI scripts that want to write to a database.  If I have that
every hit causes Python to wake up, and then mySQL to wake up, so that
the script can write to the database (and then they both go back to
sleep) then the response is very slow.

I understand that I can put Python into Apache as a module, so that
Python will always be awake.  I want to also keep mySQL up (maybe
a half-dozen clients).  I believe that I can do this with Perl's
Apache::DBI, but I'd much to prefer to go with Python.

Can I?  (I thought that this is a Python issue because Python needs
to talk to the open connections -- isn't that why Apache::DBI is a
Perl module?  I looked a bit in the SIG database archives and did a
few deja queries but I didn't see any discussion of this; sorry if I
missed a recent one.)

Thanks,  Jim jim at joshua.smcvt.edu


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


From ivanlan at callware.com  Tue Jan 18 21:22:01 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 18 Jan 2000 19:22:01 -0700
Subject: Pardon my ignorance, data conversion
References: <38829357.ED633932@home.com> <85u6cs$85d$1@nntp9.atl.mindspring.net> <oqpuv04ixm.fsf@titan.progiciels-bpi.ca> <7e8h4.3861$Ef6.1020706@news.shore.net>
Message-ID: <38851FC9.E5CBC0E9@callware.com>

Hi All--

"Michael P. Reilly" wrote:
> 

[bobbitt]

> Floor does funny things to negative numbers.
> 

I disagree.  They're not funny, only unexpected if you've been working
with C or some other language that doesn't floor() or ceil() the same
way Python does.  It's merely a different perspective on what's right.

If you rely on the C-like behaviour, then you can devise your own
function that checks the sign of the operands, does an abs(), does the
floor() or ceil() on the positive numbers, and returns the negative of
that result when appropriate.

Personally, I've only found one and only one instance where I required
the C-like behaviour.

<an-n-of-one-is-not-a-groundswell>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From bowman at montana.com  Sun Jan  2 19:31:25 2000
From: bowman at montana.com (bowman)
Date: Mon, 03 Jan 2000 00:31:25 GMT
Subject: New To Python
References: <1cRb4.10160$zp.149897@news3.mia>
Message-ID: <x5Sb4.4196$S%2.21007@newsfeed.slurp.net>

Gary N. Potts <gnpotts at email.uophx.edu> wrote:

>however, I have not been able
>to find a decent instructional guide on how to actually use the program
>itself.  Is there any good tutorials for that purpose out there and if so,

What do you mean by 'how to use the program?' You have two choices; type
python at the prompt, and you will be in an interactive mode, indicated by
the default '>>>' prompt. You can then enter simple expressions, function
definitions, and so forth to experiment with the language.

Other than that, you edit a Python program, and wither run it as
> python foo.py 
from your prompt, or with something like
#! /usr/local/bin/python
as the first line in the file, depending on your platform and setup.

You can use your favorite text editor to do this, or the idle utiity included
with the distro.

Other than that, Python doesn't 'do' anything. It is a programming language;
you write the program, and it interprets and executes it. 







From jstok at bluedog.apana.org.au  Sun Jan 16 17:55:41 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Sun, 16 Jan 2000 22:55:41 GMT
Subject: If you can only buy *one* Python book...
Message-ID: <N%rg4.4360$3b6.20735@ozemail.com.au>

Which one should you get?  I quite like the look of the New Riders
Python Reference, but is it comprehensive enough?  I already know Perl,
and I actually prefer learning from reference works.  Suggestions?

-- 
Jason Stokes: jstok at bluedog.apana.org.au


From pinard at iro.umontreal.ca  Tue Jan 18 21:06:09 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 18 Jan 2000 21:06:09 -0500
Subject: Python Rocks!
In-Reply-To: "tye4"'s message of "Tue, 18 Jan 2000 12:50:55 -0800"
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com>
Message-ID: <oqpuuystm6.fsf@titan.progiciels-bpi.ca>

"tye4" <tye4 at yahoo.com> writes:

> I don't need to design a language to detect flaws in it.  This problem
> needs to be fixed.  [...] (I don't want to type 4 spaces for every line.)

My humble feeling is that you need your editor fixed, and that you report
the problem you have as being one of Python language design.  We should
be much wiser, and keep things in better proportion.  Let me suggest that
you pick some better editor, and just enjoy Python like most of us do! :-)

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





From thomas at xs4all.net  Mon Jan 17 09:20:49 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 17 Jan 2000 15:20:49 +0100
Subject: conflict between my package and os standard module
In-Reply-To: <08Eg4.25980$zU5.258198@wbnws01.ne.mediaone.net>; from benwittner@mediaone.net on Mon, Jan 17, 2000 at 12:43:40PM +0000
References: <Dkcg4.19933$zU5.216301@wbnws01.ne.mediaone.net> <38823BF3.40D3AB96@easics.be> <20000117095108.C21652@xs4all.nl> <08Eg4.25980$zU5.258198@wbnws01.ne.mediaone.net>
Message-ID: <20000117152049.D21652@xs4all.nl>

On Mon, Jan 17, 2000 at 12:43:40PM +0000, Ben Wittner wrote:

> Actually, though, my problem is the reverse; in my_proj/os.py, I don't have
> access to the standard os services because when I say
> import os
> in my_proj/os.py, I get my_proj/os.py instead of the standard os.

This is actually also solved by renaming your os.py to my_proj_os.py ;)

-- 
Thomas Wouters <thomas at xs4all.net>

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



From aaron_watters at my-deja.com  Wed Jan  5 12:39:43 2000
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Wed, 05 Jan 2000 17:39:43 GMT
Subject: Pseudo-SQL: searching for Parser/Rewriter
References: <16600378@oberon.noris.de> <87vh5c35xh.fsf@baal.winnegan.de> <16600381@oberon.noris.de>
Message-ID: <84vvks$6ae$1@nnrp1.deja.com>

In article <16600381 at oberon.noris.de>,
  "hartmut Goebel" <hartmut at oberon.noris.de> wrote:
> Siggy Brentrup (bsb at winnegan.de) schrieb:
>
>  >As soon as the search engine on www.python.org is up again, look out
>  >for Gadfly by Aaron Watters, it's an SQL database engine written in
>  >Python.
>
> As far as I understood, Gadfly is it's own DBMS. I want to use
postgresql for
> speed reasons.

ahem.  It might be interesting to try it out and at least
see if postgressql is really faster than gadfly for what you
are doing.  It's easy to try anyway, please do.
   -- Aaron Watters http://www.chordate.com

===
...and now if you'll excuse me it's time for my
daughter to give me another Donkey Kong tutorial...


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


From michelorengo at netscape.com  Wed Jan 19 15:59:25 2000
From: michelorengo at netscape.com (Michel Orengo)
Date: Wed, 19 Jan 2000 20:59:25 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com> <slrn88c42t.53m.grant@grante.comtrol.com> <slrn88c8uk.5a8.grant@grante.comtrol.com>
Message-ID: <38862692.B59C43D3@netscape.com>


Grant Edwards wrote:

> I'm assuming that stuff like the message body, sender's e-mail
> address, etc. are all contained in the MAPIOBJECT object.
>
> How do I find out what type of object that is?
>

For the body, I believe you can try msg.text
For the Sender address, try msg.Sender.address

The following table summarizes each object's properties and methods.


Object  Available since version
Properties
Methods
AddressEntries collection  1.1  Application, Class, Count, Filter, Item, Parent,
RawTable1, Session  Add, Delete, GetFirst, GetLast, GetNext, GetPrevious, Sort
AddressEntry  1.0.a  Address, Application, Class, DisplayType, Fields, ID,
Manager, MAPIOBJECT1, Members, Name, Parent, Session, Type  Delete, Details,
GetFreeBusy, IsSameAs, Update
AddressEntryFilter  1.1  Address, Application, Class, Fields, Name, Not, Or,
Parent, Session  IsSameAs
AddressList  1.1  AddressEntries, Application, Class, Fields, ID, Index,
IsReadOnly, Name, Parent, Session  IsSameAs
AddressLists collection  1.1  Application, Class, Count, Item, Parent, Session
(none)
AppointmentItem  1.2  AllDayEvent, Application, Attachments, BusyStatus,
Categories, Class, Conversation, ConversationIndex, ConversationTopic,
DeliveryReceipt, Duration, Encrypted, EndTime, Fields, FolderID, ID, Importance,
IsRecurring, Location, MAPIOBJECT1, MeetingResponseStatus, MeetingStatus,
Organizer, Parent, ReadReceipt, Recipients, ReminderMinutesBeforeStart,
ReminderSet, ReplyTime, ResponseRequested, Sender, Sensitivity, Sent, Session,
Signed, Size, StartTime, StoreID, Subject, Submitted, Text, TimeCreated,
TimeExpired, TimeLastModified, TimeReceived, TimeSent, Type, Unread
ClearRecurrencePattern, CopyTo, Delete, GetRecurrencePattern, IsSameAs, MoveTo,
Options, Respond, Send, Update
Attachment  1.0.a  Application, Class, Fields, Index, MAPIOBJECT1, Name, Parent,
Position, Session, Source, Type  Delete, IsSameAs, ReadFromFile, WriteToFile
Attachments collection  1.0.a  Application, Class, Count, Item, Parent, Session
Add, Delete
Field  1.0.a  Application, Class, ID, Index, Name, Parent, Session, Type, Value
Delete, ReadFromFile, WriteToFile
Fields collection  1.0.a  Application, Class, Count, Item, Parent, Session  Add,
Delete, SetNamespace
Folder  1.0.a  Application, Class, Fields, FolderID, Folders, HiddenMessages, ID,
MAPIOBJECT1, Messages, Name, Parent, Session, StoreID  CopyTo, Delete, IsSameAs,
MoveTo, Update
Folders collection  1.0.a  Application, Class, Count, Item, Parent, RawTable1,
Session  Add, Delete, GetFirst, GetLast, GetNext, GetPrevious, Sort
GroupHeader  1.1  Application, Class, Count, Level, Name, Parent, Session, Unread
(none)
InfoStore  1.0.a  Application, Class, Fields, ID, Index, MAPIOBJECT1, Name,
Parent, ProviderName, RootFolder, Session  IsSameAs
InfoStores collection  1.0.a  Application, Class, Count, Item, Parent, Session
(none)
MeetingItem  1.2  Application, Attachments, Categories, Class, Conversation,
ConversationIndex, ConversationTopic, DeliveryReceipt, Encrypted, Fields,
FolderID, ID, Importance, MAPIOBJECT1, MeetingType, Parent, ReadReceipt,
Recipients, Sender, Sensitivity, Sent, Session, Signed, Size, StoreID, Subject,
Submitted, Text, TimeCreated, TimeExpired, TimeLastModified, TimeReceived,
TimeSent, Type, Unread  CopyTo, Delete, Forward, GetAssociatedAppointment,
IsSameAs, MoveTo, Options, Reply, ReplyAll, Respond, Send, Update
Message  1.0.a  Application, Attachments, Categories, Class, Conversation,
ConversationIndex, ConversationTopic, DeliveryReceipt, Encrypted, Fields,
FolderID, ID, Importance, MAPIOBJECT1, Parent, ReadReceipt, Recipients, Sender,
Sensitivity, Sent, Session, Signed, Size, StoreID, Subject, Submitted, Text,
TimeCreated, TimeExpired, TimeLastModified, TimeReceived, TimeSent, Type, Unread
CopyTo, Delete, Forward, IsSameAs, MoveTo, Options, Reply, ReplyAll, Send, Update
MessageFilter  1.1  Application, Class, Conversation, Fields, Importance, Not, Or,
Parent, Recipients, Sender, Sent, Session, Size, Subject, Text, TimeFirst,
TimeLast, Type, Unread  IsSameAs
Messages collection  1.0.a  Application, Class, Count, Filter, Item, Parent,
RawTable1, Session  Add, Delete, GetFirst, GetLast, GetNext, GetPrevious, Sort
Recipient  1.0.a  Address, AddressEntry, AmbiguousNames, Application, Class,
DisplayType, ID, Index, MeetingResponseStatus, Name, Parent, Session, Type
Delete, GetFreeBusy, IsSameAs, Resolve
Recipients collection  1.0.a  Application, Class, Count, Item, RawTable1, Parent,
Resolved, Session  Add, AddMultiple, Delete, GetFirstUnresolved, GetFreeBusy,
GetNextUnresolved, Resolve
RecurrencePattern  1.2  Application, Class, DayOfMonth, DayOfWeekMask, Duration,
EndTime, Instance, Interval, MonthOfYear, NoEndDate, Occurrences, Parent,
PatternEndDate, PatternStartDate, RecurrenceType, Session, StartTime  (none)
Session  1.0.a  AddressLists, Application, Class, CurrentUser, Inbox, InfoStores,
MAPIOBJECT1, Name, OperatingSystem, Outbox, OutOfOffice, OutOfOfficeText, Parent,
Session, Version  AddressBook, CompareIDs, CreateConversationIndex, DeliverNow,
GetAddressEntry, GetAddressList, GetArticle, GetDefaultFolder, GetFolder,
GetInfoStore, GetMessage, GetOption, Logoff, Logon, SetLocaleIDs, SetOption




From skip at mojam.com  Thu Jan  6 10:15:08 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 6 Jan 2000 09:15:08 -0600 (CST)
Subject: control structures (was "Re: Sins")
In-Reply-To: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
Message-ID: <14452.45436.641890.197509@beluga.mojam.com>

    Emile> Doesn't this do that now? or is there some objection to this usage?

    [ ... snip try/except used to implement state machine ... ]

I find nothing wrong with it.  Some people object to using try/except for
anything but handling errors however.

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  Mon Jan 24 21:12:24 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 24 Jan 2000 21:12:24 -0500
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: Stefan Schwarzer's message of "Tue, 25 Jan 2000 02:09:24 +0100"
References: <388670D2.2FA842F6@starmedia.net> <Pine.SOL.3.91.1000120001532.29235B-100000@sunse> <tIeGOHk6cMDxz41yCwn96Iy45Dgx@4ax.com> <388CF7C4.D6247285@ndh.net>
Message-ID: <oqln5ex5kn.fsf@titan.progiciels-bpi.ca>

Stefan Schwarzer <s.schwarzer at ndh.net> writes:

> The only programming languages I (personally) know to pay attention
> to amount of whitespace are Fortran, Assembler and Python ;-) .

For most of the remaining languages, any good programmer has to pay
attention to the amount of whitespace him/herself, because nicely presented
programs are a good thing.  Any serious programmer has to take care anyway.
This being given, compilers should really validate and diagnose wrong or
inconsistent indentation, in my opinion.  It is a pity to see such amounts
of programmer-added redundancy to be thrown away and wasted by compilers.

I believed in the above for years, and unsuccessfully tried to convince the
GNU project to amend their C compiler accordingly.  So, when I re-discovered
Python, not so long ago, it was a real and deep pleasure for me to see this
particular aspect of programming, correct indentation, so well integrated
in the language.

To be honest, I just do not understand the big fuss some people do about
whitespace in Python.  They are all given a very nice gift, and it makes
their misery.  I shudder to think what their sources would look like, if
they really consider that indentation is a curse.  I like the idea that
Python programs have readability on their side from the start, by design.

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





From tim_one at email.msn.com  Fri Jan  7 22:24:07 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 7 Jan 2000 22:24:07 -0500
Subject: lazy (?) evaluation, request for advice
In-Reply-To: <6D8A17398E28D3119F860090274DD7DB4B3DB1@pces.cadlab.it>
Message-ID: <000301bf5987$d3447800$342d153f@tim>

[Alex Martelli]
> Here we go again.
> ...

> ... [arguments of the form "I'm implementing my language
>      using Python, therefore it has Python semantics"]

The imposition of lazy semantics onto unattributed names cannot be explained
by reference to the Python language, without detailing your implementation.
No Python programmer using your language could account for their program's
runtime behavior, since such behavior is not provided by Python directly,
and there is (as you had determined already in your original post) no way
even to *trick* it into such behavior:  you have to emulate (or otherwise
interfere with) key parts of Python's runtime evaluation yourself to get the
behavior you want.  So names in your language do not mean what they mean to
a Python programmer -- they have different behavior.  That's about all there
is to it!

You don't want Python's builtin behavior, and that's OK by me -- don't
*call* it Python anyway and I'll have no argument (although I'll still have
advice you dislike ...).

"eval" etc are irrelevant; those just happen to be what you're using (or may
use) to implement your variant semantics.

It is not the same as library etc API design.  The lowest level of what the
user sees works in ways that Python does not, in areas where Python does not
provide hooks to change the way it works.

> ...
> If nothing changes except for speed,

I count speed as a crucial aspect of behavior here.  So do you (you've said
so often enough <wink>).  Lazy evaluation has purely semantic consequences
too in a language with side-effects (changes in when-- or even if --things
get called can have very visible consequences).

> how can what-used-to-be-Python's semantics suddenly become
> "my own private and badly designed language"'s...?

While I have suggested several times that it would be better to let your
users write in straight Python (yes, with an ordinary "import this, that"
module UI), I've said very little about your design.  I certainly didn't
call it "badly designed".  If anyone asked <wink>, I would *not* say it's
badly designed.  I merely think it (as I said before) falls into a classic
little-language trap (providing ungeneralizable magic catering to its first
envisioned applications).  It's a classic trap *because* it looks so
attractive at first.

> ...
> So, I'll have a textfield where (as one of the different
> optional ways to express what it is they're looking for)
> users can enter a Python expression (an expression in the
> language which you loudly claim is not Python but rather
> "my own specially purpose-designed language") -- if they do
> that then clic a certain pushbutton, the expression gets
> (perhaps computed, then) evaluated as above.

I agree with Tres that it would be nice if this GUI showed (or offered to
show) a translation into Python -- as he said, "learn by example".  When
they get to writing complex functions, to do things you haven't anticipated,
they'll need to learn how to get things done for themselves.

Obviously, the harder it is *to* supply a straight Python equivalent, the
more dubious your original claim that "I'd much rather let the user write
the deal-selection function in Python" (you do realize we're still arguing
about that single sentence -- indeed, the last two words of it <0.7 wink>?).

>> From what you've said so far, it doesn't *look* like anything
>> fancier than a regexp replace is needed; e.g.,

> Looks to me like that could be slow if the vocabulary is
> pretty large.  E.g., hundreds of possible words, of which
> maybe 4 to 6 will typically be used in one expressions.

Again, it's impossible to be truly helpful here before you define your
language fully:  if you're not defining your own language, how is it that
all your questions are about language translation and implementation
techniques <0.1 wink>?  The most appropriate techniques depend on details of
the source language, and c.l.py hasn't seen a definition. "What's the best
way to sort?" -- same thing.

Anyway, that you may have hundreds of special keywords wasn't mentioned
before.  Yes, the Python/Perl flavor of backtracking regexp engine will bog
down given hundreds of alternatives; but if you're going to do the
translation once and then run the function thousands (hundreds of
thousands?) of times, does that matter?  Time it and see!  The regexp engine
runs at C speed, so it's not obvious that speed is an issue here.

> ...
> I have no intention of somehow forbidding users from having
> string literals in the expression they can enter, although
> offhand I'm not sure what use they could have for them in
> such an expression.

Then you have a harder parsing problem.  I'd suggest looking at these next:

1) What (the current CVS version of) pyclbr.py (std library) does
   to skip over string literals rapidly.  This is ad hoc regexp
   trickery.

2) tokenize.py (std library).  High-level parsing of "Python-like"
   syntax.  Delivers a pure token stream, not a parse tree.  But
   that shouldn't matter unless you intend to introduce e.g.
   new block constructs.  No canned way to convert back to source,
   but very easy to write that.

3) Marc-Andre Lemburg's mxTextTools pkg (Starship), provided you
   can tolerate shipping with an extension module (i.e., the
   search engine in that is coded in C).  This will be the
   fastest; a non-backtracking recognition DFA can be built that
   recognizes keywords in one pass no matter how many there are;
   ditto for skipping over strings; etc.

4) John Aycock's SPARK system (see DejaNews for a current URL).
   This is wonderful for rapid prototyping of translators, but
   not swift compared to the alternatives.

...

>> Didn't work, eh?

> Apparently not.

Ya -- I picked that up.  Thanks for the confirmation <wink>.

offense-is-where-you-look-for-it-ly y'rs  - tim





From grant at nowhere.  Mon Jan 17 15:49:24 2000
From: grant at nowhere. (Grant Edwards)
Date: Mon, 17 Jan 2000 20:49:24 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <85vs9o$sbh$1@nntp9.atl.mindspring.net>
Message-ID: <slrn88702k.2fm.grant@grante.comtrol.com>

In article <85vs9o$sbh$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:
>In article <slrn886kml.284.grant at grante.comtrol.com>,
>Grant Edwards <grant at nowhere.> wrote:
>>
>>I just found the http server page on Exchange server that allows access
>>to e-mails via a web browser.  What I'd really like is to write a
>>Python program that just sucks the mail out of the inbox and delivers
>>it locally.
>
>No can do unless you implement a JavaScript engine.  :-(  (Trust me, I'd
>*love* to use Lynx for my mail.)

I don't think that you actually need JavaScript -- it's just to
fancy-up the user-interface (it does things like checking the
form field values for sanity before sumitting the form).  

You have to _pretend_ that you understand JavaScript, so that
the server will send you the page, but It looks like if you
submit the forms with the right requests you can retrieve
e-mail.  You do have to support cookies...

I've found jrush's webbot.py, and that looks like a good
starting point.  

I'm going to figure out how to get my mail without running
Outlook if it takes hundres of years and costs thousands of
lives...

-- 
Grant Edwards                   grante             Yow!  I want you to
                                  at               MEMORIZE the collected
                               visi.com            poems of EDNA ST VINCENT
                                                   MILLAY... BACKWARDS!!


From robin at jessikat.demon.co.uk  Wed Jan  5 12:12:37 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 5 Jan 2000 17:12:37 +0000
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
In-Reply-To: <20000105091821.D9441@vislab.epa.gov>
References: <841d0c$j0n$1@violet.singnet.com.sg>
 <387f6d37.25522789@news.demon.co.uk> <19991230065516.A778039@vislab.epa.gov>
 <W8TkoEATH4a4Ewpu@jessikat.demon.co.uk> <387378c0.4549271@news.demon.co.uk>
 <xc1BqAAkwpb4EwWW@jessikat.demon.co.uk> <20000105082703.A9441@vislab.epa.gov>
 <uBCWdBAkv0c4Ewdy@jessikat.demon.co.uk> <20000105091821.D9441@vislab.epa.gov>
Message-ID: <vSamZDAFu3c4EwM7@jessikat.demon.co.uk>

In message <20000105091821.D9441 at vislab.epa.gov>, Randall Hopper
<aa8vb at yahoo.com> writes
>Robin Becker:
> |Randall Hopper:
> |>Robin Becker:
> |> |how arithmetic errors are handled using built in IEEE exceptions and
> |> |how does one gracefully crash a python script? (eg an abend/abort).
> |>
> |>raise ArithmeticError, "Oops"
> |>raise ZeroDivisionError, "Ouch"
> |>
> |>?
> |>
> |yes that's python, but what's the best way to do it in an existing C
> |library?
>
>And in the function that returns control to Python, do this:
>
>    PyErr_Format( PyExc_ArithmeticError, "Oops\n" );
>    return NULL;
>
>The first statement can be invoked elsewhere if desired.
>
OK then I have to catch my own C exceptions to pass back to python. How
about file descriptors. Must I have explicit control over these or must
I create something equivalent.
-- 
Robin Becker



From da at ski.org  Mon Jan  3 18:16:44 2000
From: da at ski.org (David Ascher)
Date: Mon, 3 Jan 2000 15:16:44 -0800
Subject: Call for Participation -- O'Reilly Open Source Software Convention 
Message-ID: <018701bf5640$99d8a130$c355cfc0@ski.org>

Please find below the call for participation for the O'Reilly Open Source
Software Convention.

Note that the deadline for paper and tutorial submissions is February 1,
2000.

-- David Ascher, on behalf of the Python track program committee.

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


O'Reilly Open Source Software Convention
July 17-20, 2000
Monterey, California

Call for Participation
======================

O'Reilly & Associates is pleased to announce the 2nd annual Open Source
Software Convention. This is the largest cross-technology gathering of the
open source community. It features 13 concurrent tracks focused on Linux,
Perl, Apache, Python, Sendmail, Tcl/Tk, and Open Source Business Strategies.
Last year's event attracted over 1700 attendees. The Perl Conference 4.0 is
a 'conference-within-a-conference' at this event.

The Open Source Convention is the central gathering place for the open
source community to exchange ideas, techniques, and to advance the open
source software. The Convention is a four-day event designed for
programmers, developers and technical staff involved in or interested in
open source technology and its applications. The Convention will be held at
the Monterey Conference Center in Monterey California, July 17-20, 2000.

The Convention includes two days of intensely focused tutorials aimed mainly
at intermediate and advanced users of open source technology. These
tutorials are designed to provide concrete knowledge that leads directly to
better application of open source tools. Two days of multi-tracked
conference sessions focus on the cutting edge of open source technology and
feature talks, demonstrations, and panel debates.

Technically sophisticated individuals who are actively working with open
source technology make all tutorial and conference presentations.
Presentations by marketing staff or with a marketing focus will not be
accepted.

Important Dates
===============

     Proposals due: February 1, 2000
     Notification to conference and tutorial presenters: March 1, 2000
     Tutorial presentations due: May 1, 2000
     Conference presentations due: June 1, 2000

Participation Opportunities
===========================

Individuals and companies interested in making technical presentations at
the Convention are invited to submit proposals to the Convention organizers
following the guidelines below. Proposals will be considered in three
classes: tutorial programs, conference presentations, and refereed papers.
The program committees for the appropriate Convention track will evaluate
all proposals.


Topics
======

A wide variety of topics will be considered for the following tracks:

     Linux
     Apache
     Python
     Open Source Business Strategies
     Tcl/Tk
     Other open source technologies

This year the Convention organizers are focusing on the open source
technologies enabling secure, high-volume e-commerce. Technical talks in any
of the above tracks that focus on e-commerce will be particularly welcomed.

Presenters interested in Perl are directed to review the separate call for
participation for the Perl Conference 4.0.


Tutorial Presentations
======================

The first two days of the conference are devoted to tutorial presentations.

Tutorial proposals must include:
     Tutorial Title
     Tutorials length (3 or 6 hours)
     Target audience including any prerequisites for taking advantage of the
material
     What attendees will learn--the knowledge or skill take-away attendees
will have at the completion of the tutorial
     Tutorial outline--a short bullet list or description of the course
itself and the material it covers
     Speaker name
     Speaker biography
     Complete speaker contact information; if applicable, include
administrative support contact information


Conference Presentations
========================

Conference presentations and participation will beconsidered for these types
of talks:

     Case studies showing an application of open source technology, the
problems that were solved and the lessons learned

     Focused technology discussions covering a specific open source
technology and showing a variety of implementation or application options

     New product/technology demonstrations--live demonstrations of released
or near released open source software

     Panel participation--speakers to participate in one of the various
panel discussions that are part of the conference


Conference proposals must include:
==================================

     Type of talk--technology, new product, case study, or panel

     Title of talk or demonstration

     Requested talk length: 30, 45, 60, or 90 minutes

     Abstract of talk, maximum 250 words

     Speaker name

     Speaker biography

     Complete speaker contact information; if applicable, include
administrative support contact information

Submitting Your Proposal
========================

Please send all proposals to:
oscon00proposals at oreilly.com
(Note: use this address to submit proposals to the Open Source Convention
only; see the instructions for the call for proposals for the Perl
Conference 4.0 for submission to the Perl tracks of the Convention).

Submit all proposals via electronic mail in plain text with no attachments.
Submit one proposal per email.

The subject line of your email must follow this format:
Track: Your Last Name: Proposal Type (tutorial, presentation): proposal
title
For example: Python: Smith: Tutorial: Advanced Python Programming. You will
receive an automatic confirmation of receipt for each proposal.

Hard copy confirmation may be sent via mail or fax to the address below.
Address all questions regarding proposals and proposal status to the
Convention Director.

     Joseph McIntyre
     Director of Conferences
     O'Reilly & Associates
     101 Morris St., Sebastopol, CA 95472
     joseph at oreilly.com
     Phone: (707) 829-0515 x372
     Fax: (707) 823-9746





From skip at mojam.com  Wed Jan 19 15:22:07 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 19 Jan 2000 14:22:07 -0600 (CST)
Subject: Newbie Q: Zope with static content?
In-Reply-To: <Pine.SO4.4.02.10001180010450.27409-100000@ub.d.umn.edu>
References: <Pine.SO4.4.02.10001180010450.27409-100000@ub.d.umn.edu>
Message-ID: <14470.7407.813617.817100@beluga.mojam.com>

    Aaron> I hope this is an appropriate newsgroup to discuss this -- I read
    Aaron> this newsgroup offen, but do not subscribe to the Zope mailing
    Aaron> list(s).

Probably better at zope at zope.org.

    Aaron> Is there anyway Zope can read a file straight off the disk,

Not that I'm aware of.  It really likes to stick to its database.  On the
other hand, perhaps some enterprising Zoper has written a product to do this 
by now.

    Aaron> presumably within it's own directory structure? (I've Zope
    Aaron> installed at /var/zope) I could run an Apache process just to
    Aaron> server the webcam, but I would like to avoid doing that.  The
    Aaron> other option I see is to have the jpeg uploaded to the Zope FTP
    Aaron> server.  I'm having problems getting xtp to work.

ftp to the Zope Server is what I would recommend.  It does work.  

    Aaron> Also, I think it'd be handy, for other reasons, to know how one
    Aaron> can get Zope running zserver to access static content saved as
    Aaron> flat files on the filesystem.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From prestonlanders at my-deja.com  Tue Jan 11 11:36:05 2000
From: prestonlanders at my-deja.com (Preston Landers)
Date: Tue, 11 Jan 2000 16:36:05 GMT
Subject: Newbie simple question - can't find answer
References: <85fhha$6js$1@nnrp1.deja.com>
Message-ID: <85fm5k$af3$1@nnrp1.deja.com>

In article <85fhha$6js$1 at nnrp1.deja.com>,
  eleisonme at juno.com wrote:

> I do not seem to be able to use any of the math or cmath functions
> either in a script or on the python command line.
>
> For example if I type:
> >>>import math
> >>>exp (4)

> I get a traceback error:
> NameError: exp (or any other math function that I try to use).

Greetings,

import by itself doesn't make a modules names visible in your current
namespace.  In other words, you have to put the module name in front of
the function you want to call.  This is so you can access identically
named things (functions) in different modules.

Basically, you want this:

>>> import math
>>> math.exp(4)
54.5981500331

There *is* a way to make all of a modules' names visible in the current
namespace.  However, it's generally not recommend (for the collision
reaons mentioned above.)

>>> from math import *
>>> exp(4)
54.5981500331

Read the tutorial section on namespaces for further clarification, or
feel free to ask here.

yours,

---Preston


--
|| Preston Landers <prestonlanders at my-deja.com> ||


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


From stadt at cs.utwente.nl  Wed Jan  5 06:29:02 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Wed, 05 Jan 2000 12:29:02 +0100
Subject: downloading a file
References: <sXFfOL1U=LX0FKvyzRFVu0hsS4cw@4ax.com>
Message-ID: <38732AFE.69DB881B@cs.utwente.nl>

Anders M Eriksson wrote:

With a name like that you should be working for a certain
Telecommunications company :-)

> 
> Hello!
> 
> I need a script which will log the number of times a specifik file has
> been downloaded.
> 
> I have created a script that will do the logging but I don't know how
> to perform the downloading.  I want it to behave like when you
> download a file thru the web browser. The user just select the link
> and gets the standard downloading question by the web browser.
> 
> I have tried to use httplib but I don't know what kind of file to
> 'Accept'
> 
> // Anders

What about this:

import urllib, string
url = 'http://the_url_of_the_file_to_download'
data = urllib.urlopen(url).read()
#find the format of the file (usually the stuff after the rightmost dot):
suffix = url[string.rfind(url, '.') + 1:]
#Let the browser know what's coming:
if suffix == 'ps': apptype = 'postscript'
elif suffix == 'pdf': apptype = 'pdf'
elif suffix == 'zip': apptype = 'x-zip-compressed'
elif suffix == 'doc': apptype = 'msword'
elif suffix == 'gz': apptype = 'x-gzip'
elif suffix == 'Z': apptype = 'x-compress'
else apptype = 'unknown'
print 'Content-type: application/' + apptype + '\n'
#now simply print the data to the browser:
print data


Richard.


From mstenber at cc.Helsinki.FI  Tue Jan 18 04:00:35 2000
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 18 Jan 2000 11:00:35 +0200
Subject: Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us> <m12AMt3-00007rC@vortigen.demon.co.uk>
Message-ID: <al8u2kbycss.fsf@sirppi.helsinki.fi>

Edward Welbourne <eddy at vortigen.demon.co.uk> writes:
> OK, some examples would be a good idea, and Evan has shown me some ways
> of saying tunneling clearer ... Evan has at least one further change he
> wants here but I'd rather leave it to him to describe ... if only for
> the hope that he'll express why this is a useful idea - albeit he spells
> it different - and he expresses it better than I do.
> I only know how to design and implement stuff.
> Computers understand my better than folk do.
>
> 
> Three new kinds of argument come into being (with my given spelling):
>   * safe tunnel: after both * and **
>   * keyword-only: after * but not **
>   * positional-only: after ** but not *

Ugh.. *puke* this looks like ugly hack from hell designed to counter
Python's original debatable deficiency in namescape/scoping department.
All examples I've seen to date have just provided ways of countering
current scoping, and I'd rather see Python's scope system changing
(possibly something along the lines mr. Skaller of Viper fame proposed)
than language getting more and more cryptic with very little gain to show
for it. 

Question is: "Why"?

Admittedly, there is some slight convenience in whole proposal, but
additional syntax isn't good idea, and I'd hate to try to describe features
like that to general public. And AFAIK Python's mostly based on KISS
principle.

<snip>

> You want to let your function see some value you can compute, but you
> want it to be able to take arbitrarily many positional args; or you need
> to be sure your caller can't mess with one of your args.

Arguably, this might be desirable; but, language extensions like ** sound
downright ugly. 

>   def filename(*fragments, **, root=os.getcwd()):

Why not just

def filename(*fragments, root=os.getcwd()): ?

It's not allowed right now, but looks pretty clear to me, and it's logical
extension to current form.

<snip>

> --
> Recipient-list truncated to python-list,

-- 
Finger mstenber at sirppi.helsinki.fi for my PGP public key:
pub  1024/5DAC7D21 1994/04/20 Markus Stenberg <fingon at iki.fi>
Key fingerprint = 25 BA 5B 57 F1 77 0C F0  2A D0 E6 3A 2E 81 6F D3


From robin at jessikat.demon.co.uk  Mon Jan 10 13:10:34 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 10 Jan 2000 18:10:34 +0000
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu>
 <slrn87fol9.15t.kc5tja@garnet.armored.net>
 <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk>
 <slrn87imad.thc.wtanksle@hawking.armored.net>
 <slrn87j11s.vc0.kc5tja@garnet.armored.net>
 <Y$u7SGAdYae4EwJv@jessikat.demon.co.uk> <387A04AC.886DD88C@bellatlantic.net>
Message-ID: <T2FFKFAaCie4EwJn@jessikat.demon.co.uk>

In article <387A04AC.886DD88C at bellatlantic.net>, Steve Holden
<sholden at bellatlantic.net> writes
>In Samuel Falvo's defence, and without wishing to start a flame war,
>might I point out that he specifically mentioned his news server didn't
>show him the original article.  It might, therefore, be appropriate to
>suggest that he misunderstood your points.  To read his post as an
>attack
>shows a thinness of skin which won't get you by in the average newsgroup
>without daily psychological trauma.  Pythonistas are, in my limited
>experience, rather more polite than most.  With the exception of one
>particular group member, who can only be described as a complete
><wink>er.  But Tim appears to poke fun with a dry sense of humour
>which, as a Yorkshireman, reminds me of home.  [You had COM? You
>were lucky.  When I were a lad we had to pull objects out of the
>labraries with our bare hands and shout at them to find out how they
>worked.
>Etc.]
>
>I didn't read his post as attacking you, just expressing his (apparently
>well-founded) opinions on Corba and COM.  The nicer parts about COM are
>the ways it allows you to discover the interfaces of the objects you
>want
>to control.  The fact that it works so well with Python lends some
>credibility to the assertion that it's one of Microsoft's better ideas.
>That and, I would suggest, ODBC, which while not perfect has become a
>de facto standard primarily due to its ubefulness and ubiquity.
>
>regards
> Steve
>
>Robin Becker wrote:
>> 
>> In article <slrn87j11s.vc0.kc5tja at garnet.armored.net>, Samuel A. Falvo
>> II <kc5tja at garnet.armored.net> writes
>> ...
>> 
>> Don't get me wrong, but to attack me for saying I treat COM as too
>> object oriented and then attack me for being an OO neophyte seems a bit
>> silly. I'm from the fortran era, I treat COM objects as exactly that;
>> blobs with interfaces, that doesn't mean I don't appreciate nice OO
>> languages like Python. The reason I like Python is that it allows me to
>> "just do stuff".
>>

I'm sorry if I sounded too defensive. Perhaps I should regard Sam's post
as being enthusiastic rather than aggressive. Telling us we're in the
gutter of standard OO won't endear though.
-- 
Robin Becker


From chris.stromberger at mmt.bellhowll.com.cut_here.gov.org.com.edu  Wed Jan 26 10:59:41 2000
From: chris.stromberger at mmt.bellhowll.com.cut_here.gov.org.com.edu (Chris Stromberger)
Date: Wed, 26 Jan 2000 09:59:41 -0600
Subject: Where is python15_d.dll?
Message-ID: <8B06D3D65EAA8FF6.B5390C8B3C382830.3DAD830FA4CDC9B9@lp.airnews.net>

Newbie attempt to embed python in C++, calling Py_Initialize() and
PyRun_SimpleFile().  Compiles and links, but cannot find
"python15_d.dll" when it runs.  If I build the release version (using
MSVC 6), it runs (uses python15.dll, which is present) but crashes.
Where can I get the debug dll?

TIA.


From gerrit.holl at pobox.com  Wed Jan 19 01:50:01 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 19 Jan 2000 07:50:01 +0100
Subject: More problems....
In-Reply-To: <2E7h4.230$Pe6.2431@newsfeed.slurp.net>; from ScrewMe@mypad.com on Tue, Jan 18, 2000 at 06:42:35PM -0600
References: <2E7h4.230$Pe6.2431@newsfeed.slurp.net>
Message-ID: <20000119075001.A1479@stopcontact.palga.uucp>

TechSpood wrote on 948217355:
> I have been reading tutorials for about 2 weeks now, and just now decided to
> put that knowledge to work, but I am having a huge problem...I will write a
> simple program (in the IDLE enviroment) then when I run it, it will run it
> til it gets to the first input statement, at which time it will stop reading
> the program.  I don't get a syntax error  or any other kind of error.  After
> I went back and read and re-read the tutorials i got very frusterated after
> not being able to find absolutly any errors in my code... I even went so far
> as to copy the code excactly as it was in a tutorial and put it in the
> compiler.... same damn problem.  Is there some option I need to turn on or
> some option that I might of accidently turned on that I wasn't supposed to?

Well, maybe idle doesn't have any STDIN?
What happens if you run your program from a Terminal (Windows people call
it DOS box)?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From moeller at networksplus.net  Sun Jan 23 02:36:48 2000
From: moeller at networksplus.net (Derek Moeller)
Date: Sun, 23 Jan 2000 01:36:48 -0600
Subject: TCL_GetFile removed
Message-ID: <20000123013648.B27888@networksplus.net>

I've been compiling 1.5.2 of Python, and it stops compiling when it gets
to the TCL bit. There are undefined ref's to TCL_GetFile.

I searched Deja, and found that that feature (TCL_GetFile) had been
removed somewhere in the 8.* line.

Which TCL version is the latest safe way to go?

-- 
Derek Moeller
  moeller at networksplus.net
  http://dophnic.yi.org



From someone at deneb.cygnus.argh.org  Tue Jan 18 13:52:47 2000
From: someone at deneb.cygnus.argh.org (Florian Weimer)
Date: 18 Jan 2000 19:52:47 +0100
Subject: [Tkinter/Win32] Message box on app start -> mouse is dead
Message-ID: <87g0vv6wlc.fsf@deneb.cygnus.argh.org>

This code doesn't work on Windows (X11 is fine, though):

from Tkinter import *
import tkMessageBox
import sys

tk = Tk()
Button(tk, text="Exit", command=sys.exit).pack()
tkMessageBox.showinfo("Test app", "Click on Ok, then on Exit.\n"
                      + "Do you see the problem?")
tk.mainloop()

(The application doesn't respond to mouse clicks after dismissing the
message box.)

Any suggestions?  I think the problem is caused by the call to `showinfo'
before the Tk message loop is entered.  But my startup code needs to
display some information.


From marut at my-deja.com  Thu Jan 13 16:31:50 2000
From: marut at my-deja.com (marut at my-deja.com)
Date: Thu, 13 Jan 2000 21:31:50 GMT
Subject: MySQLdb, blobs, and inserting
References: <85itgj$mlc$1@nnrp1.deja.com> <85k0fd$hv2$1@news1.xs4all.nl>
Message-ID: <85lg86$jtk$1@nnrp1.deja.com>

In article <85k0fd$hv2$1 at news1.xs4all.nl>,
  boud at rempt.xs4all.nl wrote:
> marut at my-deja.com wrote:
>
> >>>> c.execute(r"INSERT INTO items (story) VALUES (%s)" % file )
> > Traceback (innermost last):
> > File "<stdin>", line 1, in ?
> > File "MySQLdb.py", line 153, in execute
> > File "MySQLdb.py", line 189, in _query
> > _mysql.OperationalError: (1064, 'You have an error in your SQL
syntax
> > near \'<!DOCTYPE HTML PUBLIC "-//W3C
> > //DTD HTML 4.0 Transitional//EN" "http://www.w3.org\' at line 1')
>
> I haven't tried it - but wouldn't
>
> ...." % file.read() )
>
> work? I mean, are you trying to insert a file object or the contents
> of the file? What does the mysql log file show? It's quite possible
> that that will enlighten you, since it shows the completed SQL
> statement.
> --
>
> Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt
>

I think the problem is quoting. How can I quote/escape all of the
special characters in my BLOB?


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


From tim_one at email.msn.com  Thu Jan 13 03:48:51 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 13 Jan 2000 03:48:51 -0500
Subject: Perl speed vs. Python speed
In-Reply-To: <1264425395-448550@hypernet.com>
Message-ID: <000701bf5da3$04582080$d62d153f@tim>

[Tom Culliton, to /F]
> ...
> Ask Tim and he'll recommend using small simple regular
> expressions 9 times out of 10.  At least thats how he's
> responded every time I've asked about problems with some
> excessively clever regular expression here.

Well, there *is* a bit of bias in the stimulus/response here:  by the time
someone writes a regexp so hairy that they feel compelled to post it to the
world in despair of ever getting it to work, the advice to simplify isn't
really hard to come up with <wink>.

[Gordon McMillan]
> Fredrik is emphasizing size of the string, not complexity
> of the expression.

This is putting words in /F's mouth -- although I suspect they were the
words he intended to put there himself.  The whole "speed problem" with
regex vs re is that the latter has higher fixed overhead per call, but is
(often) much faster once it gets going.

> [and here Gordon rudely repeats what I just wrote <wink>]

> Also, a long pattern is not necessarily a complex pattern. A
> humongous set of alternatives is no problem. The problem is
> when people expect to be able to do more than one thing at a
> time with a pattern.
>
> Such as match an entire Python triple quoted string.
>
> Which Tim does.

Hmm.  I'm inclined to call a single string one thing -- unless you pay
exaggerated attention to the three quotes <wink>.  For sure, I don't
hesitate to write extremely involved regexps for tokenization, but those are
always of the form

    alt1 | alt2 | alt3 | ... | altn

where the alt_i are mutually exclusive and so can be developed & reasoned
about in isolation (your "humongous set of alternatives", albeit that n is
rarely more than a dozen).  And I move heaven, earth and three acres of hell
to keep any possibility of backtracking out of each alt_i!  The regexp is
thus doubly defanged, and can't cause conceptual, speed or memory problems
regardless of string size or whether it matches or fails.

The whole trick to outsmarting an NFA engine lies in making sure none pf the
NFA "let's guess what they meant by this ambiguous pattern" machinery ever
gets triggered -- once it does, death eventually follows.

well-really-your-user's-but-it's-easy-to-get-more-
    of-them<wink>-ly y'rs  - tim





From tismer at tismer.com  Thu Jan 20 15:35:27 2000
From: tismer at tismer.com (Christian Tismer)
Date: Thu, 20 Jan 2000 21:35:27 +0100
Subject: Stackless Python 1.0 + Continuations 0.6
Message-ID: <3887718F.82E1B327@tismer.com>

ANNOUNCING:

               Stackless Python 1.0
           A Python Implementation That
             Does Not Use The C Stack

               * plus the real toy *

              Continuation Module 0.6
        Continuations as First Class Objects

What is it?
A plugin-replacement for core Python.
It should run any program which runs under Python 1.5.2 .
But it does not need space on the C stack.

Why did I write it?
Stackless Python was never written before (afaik), since it
was said to be impossible without major rewrites of core Python.
I have proven the controverse: It is easy to write, just hard to think.
About 3 times harder was finally the continuation module.
The whole project took about 6 man months where 80 percent of
the time was thinking and trying. The rest was coding and to
become a reference counting champion :-)

Recent changes:
Version 1.0 has been optimized like hell and is now
3-5 percent faster than Standard Python.

Continuation module is in version 0.6, very stable,
and it allows to save a program's "future" at any
time, in a portable way. Continuations are callable
Python objects with a very small footprint.

Who needs it?
Since the continuations are done, this is no more
only useful for C programmers who want to try certain new
ideas. Everybody who is interested to develop his
own generators, coroutines and tiny threads is
invited to check it out.

Status of the final 1.0:
Pystone works correctly and is 5% faster than standard
Python. Version 0.3 was 10 percent slower.
Continuations work with PythonWin and Idle.

The overall result is now better than expected,
and I'm happy to call this *FINAL*
(until the next version of course:)

Downloadable files can be found at
http://www.tismer.com/research/stackless/

Some older documentation:
http://www.tismer.com/research/stackless/stackless.htm

Some better documentation can be found in my IPC8 paper:
http://www.tismer.com/research/stackless/spc_final.zip
or be read directly as HTML
http://www.tismer.com/research/stackless/spcpaper.htm

Source code and a VC++6.0 build for Windows (340K):
http://www.tismer.com/research/stackless/spc_win32.zip

cheers - chris

== 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home

<P><A HREF="http://www.tismer.com/research/stackless/stackless.htm">
Stackless Python 1.0</A>
 - a version of Python 1.5.2 that does not need space on the C
stack.  (20-Jan-00)



From Jean-Daniel.Aussel at bull.net  Fri Jan 21 12:57:49 2000
From: Jean-Daniel.Aussel at bull.net (Jean-Daniel Aussel)
Date: Fri, 21 Jan 2000 18:57:49 +0100
Subject: Changing the passed arguments value after returning from a module
References: <002e01bf641b$5a5f8fd0$3acbd9c2@peridot.optichrome.com>
Message-ID: <86a66k$8h01@tosca.frcl.bull.fr>

thanks for the tips.
I tried the code you sent that does not however succeeds:

    /* N.B.: Not tested */
    static PyObject* sct_tt( PyObject* zself, PyObject* args)
    {
        PyObject* o;
        if (!PyArg_ParseTuple(args, "O", o))
             return NULL;
        ((PyIntObject*) o)->ob_ival = 101;
        Py_INCREF(Py_None);
        return Py_None;
    }

I am currently investigating code generated with SWIG1.1, wich looks like:

static PyObject *_wrap_fact(PyObject *self, PyObject *args) {
    PyObject * _resultobj;
    int  _result;
    int * _arg0;
    char * _argc0 = 0;

    if(!PyArg_ParseTuple(args,"s:fact",&_arg0))
        return NULL;
    if (_argc0) {
        if (SWIG_GetPtr(_argc0,(void **) &_arg0,"_int_p")) {
            PyErr_SetString(PyExc_TypeError,"Type error in argument 1 of
fact. Expected _int_p.");
        return NULL;
        }
    }
    _result = (int )fact(_arg0);
    _resultobj = Py_BuildValue("i",_result);
    return _resultobj;
}

I had no time to test though, since the weekend is coming. More on this next
week...

Regards,

/jda

Adrian Eyre wrote in message
<002e01bf641b$5a5f8fd0$3acbd9c2 at peridot.optichrome.com>...
>> >>> import mymodule
>> >>> i = (0,)
>> >>> apply(mymodule.test, i)
>> >>> print i
>> 101
>
>Slight correction here. Probably obvious, but...
>
>>>> import mymodule
>>>> i = (0,)
>>>> apply(mymodule.test, i)
>>>> print i
>(101,)
>
>--------------------------------------------
>Adrian Eyre <a.eyre at optichrome.com>
>http://www.optichrome.com
>
>
>




From boud at rempt.xs4all.nl  Thu Jan 13 02:56:29 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 13 Jan 2000 07:56:29 GMT
Subject: MySQLdb, blobs, and inserting
References: <85itgj$mlc$1@nnrp1.deja.com>
Message-ID: <85k0fd$hv2$1@news1.xs4all.nl>

marut at my-deja.com wrote:

>>>> c.execute(r"INSERT INTO items (story) VALUES (%s)" % file )
> Traceback (innermost last):
> File "<stdin>", line 1, in ?
> File "MySQLdb.py", line 153, in execute
> File "MySQLdb.py", line 189, in _query
> _mysql.OperationalError: (1064, 'You have an error in your SQL syntax
> near \'<!DOCTYPE HTML PUBLIC "-//W3C
> //DTD HTML 4.0 Transitional//EN" "http://www.w3.org\' at line 1')

I haven't tried it - but wouldn't 

...." % file.read() )

work? I mean, are you trying to insert a file object or the contents
of the file? What does the mysql log file show? It's quite possible
that that will enlighten you, since it shows the completed SQL
statement.
-- 

Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt


From bjorn at roguewave.com  Wed Jan 19 21:32:54 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jan 2000 19:32:54 -0700
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com> <slrn88cnb6.4i6.talon@coursen.ifu.net> <oqzou1v5rj.fsf@titan.progiciels-bpi.ca>
Message-ID: <388673D6.192A1907@roguewave.com>

> P.S. - I begin to think he is trolling us, and rather successfully! :-)
> OK, enough said.  Let's switch to something more productive...

but, but, but,... I haven't posted to this thread yet.... ;-)

-- bjorn




From aa8vb at yahoo.com  Mon Jan 24 11:41:42 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 24 Jan 2000 11:41:42 -0500
Subject: Executing X-Windows app
In-Reply-To: <3888A53C.7B4CE735@earthlink.net>
References: <3888A53C.7B4CE735@earthlink.net>
Message-ID: <20000124114142.A358575@vislab.epa.gov>

Jamey Cribbs:
 |
 |Hi, I hope this doesn't end up being a stupid question.  I am trying to
 |start a n X app from within a Python script.
 |
 |If I use os.system, my system freezes until I CTRL-C.  If I run os.system
 |from interactive Python, it works ok.  I have tried popen, but it does
 |the same thin g.
 |
 |What am I missing?  I am sure it is something very basic.  I have looked
 |throug ht the ref manual, but cannot seem to find the answer.  By the
 |way, I am running on top of RH Linux 6.1 with Python 1.52.  Any help
 |would be greatly appreciated.

Two things: 

   1) the X app needs to know which X display to use, and 
   2) the app needs to have permission to write to that X display.

1) So first, which display. The X display to be used is typically
   identified by the $DISPLAY environment variable (or the -display
   command-line option, as an override).

   Make sure the display is being set correctly.  For example:

        >>> os.system( "xclock -display ralph:0.0" )
   OR:
        >>> os.system( "env DISPLAY=ralph:0.0 xclock" )

2) Second, make sure this app has "permission" to write to that display.
   If it's running as a user which has a cookie for that display, and it is
   running on a machine where that cookie resides, then it does.  Or, if
   the (insecure) xhost utility has been used to open up access to the
   display to the app's host, then it does.

   For more info on magic cookies, see xauth(1); the cookies typically
   live in ~/.Xauthority.  For more info on xhost, see xhost(1).

   If your hosts are isolated from the world by a firewall, then for
   testing you can just run:
    
        xhost +

   on the X display host to open display access wide up to everyone and
   his brother, including your app.  This is very insecure!, so I wouldn't
   use it long term.

So why does the command line work?  When you start Python from the
command-line, $DISPLAY is already set, so Python (and it's subprocesses,
like your X app).  Just inherit that setting.  Also, your X terminal shell
has permission to write to your display (usually via a magic cookie), and
python and it's children inherit that too (assuming the children run as the
same user).

-- 
Randall Hopper
aa8vb at yahoo.com



From PClaerhout at CREO.BE  Fri Jan 21 06:59:58 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Fri, 21 Jan 2000 12:59:58 +0100
Subject: Screen depth
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA707B@MSGEURO1>

I'm using wxPython...

Pieter

-----Original Message-----
From: Fredrik Lundh [mailto:fredrik at pythonware.com]
Sent: Friday, January 21, 2000 1:17 PM
To: Pieter Claerhout
Cc: Python Mailing List
Subject: Re: Screen depth


Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> I'm working on Windows NT and 95/98.

using Tkinter?  wxPython?  gtk?  PythonWin?

for Tkinter, the answer is here:
http://www.pythonware.com/library/tkinter/introduction/basic-widget-methods.htm
look for winfo_depth under "window related information"

(or you can use an undocumented feature in PIL:
Image.core.display_mode() returns a tuple with mode
and size).

</F>



From grauf at rfa.org  Fri Jan 28 13:04:29 2000
From: grauf at rfa.org (Federico Grau)
Date: Fri, 28 Jan 2000 13:04:29 -0500
Subject: syslog module and thread safety question
Message-ID: <3891DA2D.AB5BBE79@rfa.org>

Hi folks,

I am building a daemon with python and will be incorporating threads
into it.  I have a parent object initializing itself (amongst other
things opening a connection to syslog) and looking for files in a
directory.  When it finds a file it will be spawning a child thread and
running a function to convert the file to another type.  My
question/concern is if I can simply pass the syslog object I initialized
in the parent to the child thread for it (and its "brothers") to use
also?

thanks,
Federico Grau ( donfede )


From aahz at netcom.com  Tue Jan  4 12:58:11 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 4 Jan 2000 17:58:11 GMT
Subject: exception arg
References: <slrn8749le.d0.mn@banger.iol.it>
Message-ID: <84tcbj$s29$1@nntp4.atl.mindspring.net>

In article <slrn8749le.d0.mn at banger.iol.it>,
Mirko Nasato <mn at altern.org> wrote:
>
>>>> 1/0
>Traceback (innermost last):
>  File "<stdin>", line 1, in ?
>ZeroDivisionError: integer division or modulo
>>>> try:
>...     1/0
>... except ZeroDivisionError, arg:
>...     if arg == "integer division or modulo":
>...             print "This is what I expect."
>...     else:
>...             print 'Unexpected. "arg" is "%s".' % arg
>... 
>Unexpected. "arg" is "integer division or modulo".

Try plain

   print arg

or

  print type(arg)


Hint: you're getting confused by the %s forced string conversion.  After
you've done the previous two steps, try

   if str(arg) == "integer division or modulo":
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From anders.eriksson at morateknikutveckling.se  Wed Jan 19 07:34:33 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Wed, 19 Jan 2000 13:34:33 +0100
Subject: Zope and Apache
Message-ID: <ta6FOAJep64FzHytVh77WRvo3KA=@4ax.com>

Hello!

I have been reading a bit about Zope and some questions has come to
mind:

On a Solaris system must I be root to install zope?

Is it possible to configure zope so it will run with a preexisting
Apache webserver?

Must I be root to do this?

// Anders
---
English isn't my first or second language, 
so errors or anything that you find offending is not there on 
purpose it's just due to the translation.


From ttnttn at toad.scripps  Fri Jan  7 22:51:53 2000
From: ttnttn at toad.scripps (Thien-Thi Nguyen)
Date: 07 Jan 2000 19:51:53 -0800
Subject: Getting SWIG to work with C++/Python
References: <9mad4.42353$Ke.27158@news21.bellglobal.com> <yPmd4.37$ed.1161540647@news.euroconnect.net>
Message-ID: <yqtogax8bl2.fsf@toad.scripps>

"Brian Dam Pedersen" <brian.pedersen at mail.danbbs.dk> writes:

> SWIG is very picky with its C++ syntax. For example it chokes on the
> following code from VC++

please find below work in progress towards being able to automatically
extract swig-compatible interface specs from C++ header files.  it is
not useful yet, having only attained rudimentary syntax analysis.  still
to go are:

- semantic analysis: (me L-paren R-paren) => default-ctor
- semantic mapping: ctors => __init__, operator+= => __add__, etc
- generation of swig/Foo.cc when ../Foo.h has inlined definitions
- hooks for class-specific tweaks and/or overrides
- documentation and an emacs mode ;->

however, i'm hoping that by posting it here others may hack on it to make
things better more quickly than i can by myself.  if so, please send code
improvements to ttnttn at scripps.edu.

swig author d.beazley says SWIG1.3 will be itself swiggable, in which case
some of this code will be replaced by calls to swig.

happy hacking!

thi, who hopes to never have to actually write python by hand <shudder>


---------------------
;;; swigpp.el --- make swig interface files from annotated C++ headers
;;; ID: $Id: swigpp.el,v 1.16 2000/01/05 16:59:11 ttnttn Exp $
;;; SOURCE: $Source: /home/ttnttn/repo/mead/swig/swigpp.el,v $

;;; Commentary:

;; This file provides the command `swigpp-process-file' and its batch variant,
;; `batch-swigpp-process-file'.  Args are IN-FILE-NAME (usually "../Foo.h"),
;; OUT-FILE-NAME (usually "Foo.j") and MODULE-NAME (usually "Foo").  See the
;; GNUmakefile in this directory for `batch-swigpp-process-file' usage.
;;
;; The input file is searched for class declarations prefixed with "//!wrap!"
;; directives.  For each such class, the public members are processed to form
;; a swig-compatible class declaration in the output file.  Class members
;; prefixed w/ "//!nowrap!" are ignored, as are all text between "//!nowrap!+"
;; and "//!nowrap!-".
;;
;; In the output, ctors and dtors are grouped first, followed by operators,
;; followed by other class members.

;; Caveat: As of $Revision: 1.16 $, the swig-compatible processing is not yet
;; implemented; i.e., the text of the class members is used directly, w/o
;; further analysis.  TODO TODO TODO.

;;; Code:

(require 'cl)				; sanity

;;;---------------------------------------------------------------------------
;;; Variables

(defvar swigpp-debug nil
  "Non-nil means swigpp.el writes debugging info to the output file.")

(defvar swigpp-wrap-one "//!wrap!"
  "Directive to wrap the next class or member.")
(defvar swigpp-nowrap-one "//!nowrap!"
  "Directive to not wrap the next class or member.")
(defvar swigpp-nowrap-start "//!nowrap!+"
  "Directive to start not wrapping successive following members.")
(defvar swigpp-nowrap-end "//!nowrap!-"
  "Directive to stop not wrapping successive members.")

;;;---------------------------------------------------------------------------
;;; Signature analysis support functions

(defun sa-point-syntax-string ()
  (char-to-string (char-syntax (char-after (point)))))

(defun sa-tokenize (b e)
  (save-excursion
    (goto-char b)
    (let ((me (cons 'w (intern (get 'swigpp 'name))))
	  (state (sa-point-syntax-string))
	  tokens snekot)
      ;; first go one direction
      (while (< (point) e)
	(if (or (looking-at "//.*") (looking-at "public:"))
	    (goto-char (match-end 0))
	  (let ((token (cons state
			     (buffer-substring
			      (point)
			      (+ (point) (skip-syntax-forward state e))))))
	    (unless (string-match "[-> ]" state)	;;; ignore whitespace
	      (push (if (string= "w" (car token))
			(cons (intern (car token)) (intern (cdr token)))
		      token)
		    snekot))))
	(setq state (sa-point-syntax-string)))
      ;; remove cruft
      (when (equal '("." . ";") (car snekot))
	(setq snekot (cdr snekot)))
      (when (equal '(w . const) (car snekot))
	(setq snekot (cdr snekot)))
      ;; now go the other direction
      (while snekot
	(let ((one (car snekot))
	      (two (cadr snekot)))
	  (cond ((and (string= "." (car one))
		      (equal '(w . operator) two))
		 (push (cons 'operator (cdr one)) tokens)
		 (setq snekot (cddr snekot)))
		((equal '("(" . "(") one)
		 (push 'L-paren tokens)
		 (setq snekot (cdr snekot)))
		((equal '(")" . ")") one)
		 (push 'R-paren tokens)
		 (setq snekot (cdr snekot)))
		((equal '("." . ",") one)
		 (push 'comma tokens)
		 (setq snekot (cdr snekot)))
		((and (equal '("." . "&") one)
		      (equal me two))
		 (push 'me-ref tokens)
		 (setq snekot (cddr snekot)))
		((equal me one)
		 (push 'me tokens)
		 (setq snekot (cdr snekot)))
		((equal '(w . const) one)
		 (push 'const tokens)
		 (setq snekot (cdr snekot)))
		(t
		 (push one tokens)
		 (setq snekot (cdr snekot))))))
      tokens)))

;;;---------------------------------------------------------------------------
;;; Support functions

(defun swigpp-ibuf () (get 'swigpp 'ibuf))
(defun swigpp-obuf () (get 'swigpp 'obuf))

(defun swigpp-intangiate-region (b e)
  (put-text-property b e 'swigpp-intangible t))

(defun swigpp-intangible-p (p)
  (get-text-property p 'swigpp-intangible))

(defun swigpp-insert (&rest s)
  (with-current-buffer (swigpp-obuf)
    (apply 'insert s)))

(defun swigpp-append (b e &optional if-0 comment)
  (when if-0 (swigpp-insert "#if 0" (if comment "" "\n")))
  (when comment (swigpp-insert "  // " comment "\n"))
  (append-to-buffer (swigpp-obuf) b e)
  (when if-0 (swigpp-insert "#endif\n")))

(defun swigpp-next-member ()
  "Advance point past newline following end of member.  Return region.
If a the ending regexp is not found, return nil."
  (let ((v (let ((beg (point)) debug-point)
	     (when (re-search-forward ";.*\n" (point-max) t)
	       (if (swigpp-intangible-p (1- (point)))
		   (progn
		     (when swigpp-debug (setq debug-point (1- (point))))
		     (goto-char (next-single-property-change
				 (1- (point))
				 'swigpp-intangible))
		     (when swigpp-debug
		       (message "moved from %d to %d" debug-point (point)))
		     (swigpp-next-member))
		 (list beg (point)))))))
    (when swigpp-debug
      (if (not v)
	  (message "<<<>>>>")
	(message "<<<")
	(message (apply 'buffer-substring v))
	(message ">>>")))
    v))

(defun swigpp-mask-non-public (name)
  (let (end)
    (goto-char (point-max))
    (while (progn (setq end (point))
		  (re-search-backward
		   "^\\(\\(public\\)\\|\\(private\\)\\|\\(protected\\)\\):"
		   (point-min) t))
      (let ((md (match-data)))
	(when (or (nth 6 md) (nth 8 md)) ; private or protected
	  (goto-char (match-beginning 0))
	  (when swigpp-debug
	    (swigpp-append (point) end t "non-public"))
	  (swigpp-intangiate-region (point) end))))))

(defun swigpp-mask-nowrap-region (name)
  (goto-char (point-min))
  (while (search-forward swigpp-nowrap-start (point-max) t)
    (let ((beg (progn (beginning-of-line) (point))))
      (search-forward swigpp-nowrap-end)
      (end-of-line) (forward-char 1)
      (when swigpp-debug
	(swigpp-append beg (point) t "nowrap region"))
      (swigpp-intangiate-region beg (point)))))

(defun swigpp-mask-single-nowrap (name)
  (goto-char (point-min))
  (let ((single-re (concat (regexp-quote swigpp-nowrap-one) "$")))
    (while (re-search-forward single-re (point-max) t)
      (let ((beg (progn (beginning-of-line) (point)))
	    (end (cadr (swigpp-next-member))))
	(when swigpp-debug
	  (swigpp-append beg end t "single nowrap"))
	(swigpp-intangiate-region beg end)))))

(defun swigpp-simple-re-snarf (re)
  (goto-char (point-min))
  (let (member match-p)
    (while (setq member (swigpp-next-member)
		 match-p (string-match re (apply 'buffer-substring member)))
      (swigpp-insert (format "  // %S\n" (apply 'sa-tokenize member)))
      (apply 'swigpp-append member)
      (apply 'swigpp-intangiate-region member))))

(defun swigpp-snarf-ctors-dtors (name)
  (message "Snarfing ctors dtors...")
  (swigpp-insert "\n  // ctors and dtors\n")
  (swigpp-simple-re-snarf (concat "^\\s-*~*\\<" name "\\>\\s-*(")))

(defun swigpp-snarf-operators (name)
  (message "Snarfing operators...")
  (swigpp-insert "\n  // operators\n")
  (swigpp-simple-re-snarf "\\<operator\\>"))

(defun swigpp-maybe-snarf-member (name b e)
  (when swigpp-debug
    (message "swigpp-maybe-snarf-member:\n%s" (buffer-substring b e)))
  (swigpp-append b e))

(defun swigpp-process-class (name)
  (swigpp-insert "class " name " {\npublic:\n")
  (let ((procs '(swigpp-mask-non-public
		 swigpp-mask-nowrap-region
		 swigpp-mask-single-nowrap
		 swigpp-snarf-ctors-dtors
		 swigpp-snarf-operators
		 ;; Add new procs here.
		 )))
    (while procs
      (funcall (car procs) name)
      (setq procs (cdr procs))))
  (message "Snarfing other things...")
  (swigpp-insert "\n  // other things\n")
  (goto-char (point-min))
  (let (member)
    (while (setq member (swigpp-next-member))
      (apply 'swigpp-maybe-snarf-member name member)))
  (swigpp-insert "}; // " name "\n\n"))

;;;---------------------------------------------------------------------------
;;; Entry points

(defun* swigpp-process-file (inf outf module-name)
  (let (c++-mode-hook find-file-hooks hs-minor-mode-hook)
    (put 'swigpp 'obuf (find-file outf))
    (erase-buffer)
    (insert "// Written " (format-time-string "%c") "\n")
    (insert "// Do not edit -- autogenerated file.\n\n")
    (insert "%module " module-name "\n")
    (insert "%{\n#include \"" inf "\"\n%}\n\n")
    (put 'swigpp 'ibuf (find-file inf))
    (modify-syntax-entry ?_ "w")
    (goto-char (point-min))
    (let ((class-id-re (concat (regexp-quote swigpp-wrap-one)
			       ".*\nclass\\s-+\\([a-zA-Z0-9_]+\\)\\s-*"))
	  (classes nil))
      (while (re-search-forward class-id-re (point-max) t)
	(push (list (match-string 1)
		    (1+ (point))
		    (progn (forward-sexp 1) (1- (point))))
	      classes))
      (unless classes
	(message "No classes found, not writing %s" outf)
	(return-from swigpp-process-file))
      (setq classes (nreverse classes))
      (message "Found classes: %s" (mapconcat 'car classes " "))
      (while classes
	(widen)
	(apply 'narrow-to-region (cdar classes))
	(let (buffer-read-only)
	  (put 'swigpp 'name (caar classes))
	  (swigpp-process-class (caar classes)))
	(setq classes (cdr classes))))
    (widen)
    (with-current-buffer (swigpp-obuf)
      ;; kludge
      (goto-char (point-min))
      (while (search-forward "{\npublic:\n" (point-max) t)
	(when (search-forward "\n\npublic:" (point-max) t)
	  (replace-match "")))
      (save-buffer))
    (kill-buffer (swigpp-obuf))
    (set-buffer (swigpp-ibuf))
    (set-buffer-modified-p nil)
    (kill-buffer (swigpp-ibuf))
    (setplist 'swigpp nil)))

(defun batch-swigpp-process-file ()
  (apply 'swigpp-process-file command-line-args-left))

;;;---------------------------------------------------------------------------
;;; Load-time actions

(provide 'swigpp)

;;; swigpp.el ends here


From tiddlerdeja at my-deja.com  Mon Jan  3 14:37:29 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Mon, 03 Jan 2000 19:37:29 GMT
Subject: class assignment copy
Message-ID: <84qtpn$lr1$1@nnrp1.deja.com>

Sorry for the lame question.

If I have a class Foo:

class Foo:
    def __init__(self):
        self.bar = 1


If I create one instance of the class:

x = Foo()

How do I make a copy of x?

y = x

Just makes y refer to x, so if I change bar on y the value is also
changed on x.

I realise this is a lame question, I just couldn't find any python
material on this.

Do I need to create some copy constructor on the class?


Any help appreciated.


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


From jam at quark.emich.edu  Mon Jan  3 19:25:28 2000
From: jam at quark.emich.edu (Jeff)
Date: Mon, 3 Jan 2000 19:25:28 -0500
Subject: CGI Script using Python
In-Reply-To: <D1bc4.40$hJ1.1161@news.rdc1.tx.home.com>; from pbible@home.com on Tue, Jan 04, 2000 at 12:20:51AM +0000
References: <D1bc4.40$hJ1.1161@news.rdc1.tx.home.com>
Message-ID: <20000103192528.A19258@quark.emich.edu>

On Tue, Jan 04, 2000 at 12:20:51AM +0000, Paul E. Bible wrote:
> Hi there,
> 
> I'm currently developing a CGI script using Python under Apache Web Server.
> I've written a simple script, placed it in the /cgi-bin subdirectory and
> perform a chmod 755 on it.  Here is the script:
> 
> -----------------------------------------------------------
> #!/usr/local/bin/python
> import os, string, sys
> from cgi import *
> 
> def Main():
>     print "Content-type: text/html\n\n"
>     print "<HTML>"
>     print "<HEAD>"
>     print "This is a test"
>     print "</HEAD>"
>     print "<BODY></BODY>"
>     print "</HTML>"
>     sys.exit(0)
> 
> Main()

change the above to:

print "Content-Type: text/html"
print
print "<HTML>"
# continue with the rest of the script. notice extra 'print' above

hope that helps.

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From anders.eriksson at morateknikutveckling.se  Wed Jan 12 07:45:18 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Wed, 12 Jan 2000 13:45:18 +0100
Subject: converting tuple to a list
Message-ID: <tHZ8OL+ftbNGsmad91aonRmEC1nh@4ax.com>

Hello!

Is this the 'best' way to convert a tuple to a list?

t = (1,2,3,4,5)
l = []

for i in range(len(t)):
	l.append(t[i])

// Anders



From dheise at debitel.net  Fri Jan 28 20:32:30 2000
From: dheise at debitel.net (Dirk Heise)
Date: 28 Jan 2000 20:32:30 MEST
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <949076889.194238@news.adtranzsig.de> <20000128180823.A1261@stopcontact.palga.uucp>
Message-ID: <01bf69c5$fe225580$LocalHost@dreadzone>

You're right; i was annoyed today when i looked at
a large C++ project with real production code in it.
But as an option of the parser, it would
make sense, so the build manager could reject all
code the developer didn't test with that option 
switched on. Of course, it's only important for
big team efforts.

Dirk

Gerrit Holl <gerrit.holl at pobox.com> schrieb im Beitrag <20000128180823.A1261 at stopcontact.palga.uucp>...
[...]
> I disagree. If you're writing a quick script, you want to do it
> quick. Maybe it's better for large programs, but for small ones, there
> will be onle more people who'll think Python sucks.



From greg at cosc.canterbury.ac.nz  Mon Jan 31 22:39:15 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Tue, 01 Feb 2000 16:39:15 +1300
Subject: Tuples
References: <20000126221606.A962@stopcontact.palga.uucp>
Message-ID: <38965563.E2EB8BC0@cosc.canterbury.ac.nz>

Gerrit Holl wrote:
> 
> With tuples, some things aren't really Pythonic. In some cases, even
> the Perl-like TMTOWTDI thing is true!
> a[1,2] = 'f'
> is equivalent to:
> a=[(1,2)] = 'f'

Yes -- for exactly the same reason that

    a = 42

is equivalent to

   a = (42)


> And:
> "aaa %s bbb" % 'foo'
> is equivalent to:
> "aaa %s bbb" % ('foo',)

Here, Guido evidently considered that the one-argument
case was sufficiently common, and the general way of
writing it sufficiently ugly, that a special case was
justified. I'm inclined to agree with him -- in this
case, it does improve readability.

-- 
Greg Ewing, Computer Science Dept,
+--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg at cosc.canterbury.ac.nz	   +--------------------------------------+


From mwh21 at cam.ac.uk  Wed Jan 12 14:21:49 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 12 Jan 2000 19:21:49 +0000
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <386E7BD4.514764EA@ndh.net> <84nv0u$bu9$1@nntp9.atl.mindspring.net> <3870DB37.761034EA@roguewave.com> <75B5A3A6EB0EDE04.FCE8CED8D9B414D5.95006D8FA0ECEB31@lp.airnews.net> <slrn872iqi.gie.neelk@brick.cswv.com>
Message-ID: <m3so03qenm.fsf@atrus.jesus.cam.ac.uk>

neelk at brick.cswv.com (Neel Krishnaswami) writes:

> Tres Seaver <tseaver at aftershock.neosoft.com> wrote:
> > 
> > What reasonable use can you propose for an associative container
> > which allows mutable keys?  You might as well revert to a list of
> > (key,value) pairs, since you're going to do a linear search on each
> > lookup anyway if the keys are allowed to change.
> 
> You can compute the hash function using the object's identity, rather
> than its value. IOW, hash on the object's address. Then you can mutate
> the contents of the object as needed and still access the hash. In
> fact, I believe this is the default behavior for Common Lisp hash
> tables (though of course you can create hash tables with different
> hash functions).

But then you have problems along these lines:

>>> d={}
>>> d[id("!2")] = 1
>>> d[id("!2")]
Traceback (innermost last):
  File "<stdin>", line 1, in ?
KeyError: 135419160

(I had to be a bit sneaky to make sure the string didn't get interned,
which possibly just hilights the problem more...)

Object equality is something that there is no right answer to; I don't
think Common Lisp has all of eq, eql, equal and equalp for the sheer
fun of it.

Using object identity would make using lists as keys (which is where
this started) almost impossible, too.

Cheers,
Michael


From ryuch at my-deja.com  Fri Jan  7 19:47:02 2000
From: ryuch at my-deja.com (ryuch at my-deja.com)
Date: Sat, 08 Jan 2000 00:47:02 GMT
Subject: timeout for urlopen
Message-ID: <8561e2$kah$1@nnrp1.deja.com>

Hi, python gurus!

How can I set the timeout for urlopen?
If it fails to connect a web site, it takes too long time
to detect the site is down. How can I set the timeout to
5 second or 10.

-cheol


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


From robin at jessikat.demon.co.uk  Thu Jan  6 14:16:09 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 6 Jan 2000 19:16:09 +0000
Subject: bsddb item size?
Message-ID: <mrE03AA5nOd4Ew2R@jessikat.demon.co.uk>

Anybody know if there are item size limitations with bsddb?
-- 
Robin Becker


From michel at digicool.com  Wed Jan 12 12:15:31 2000
From: michel at digicool.com (Michel Pelletier)
Date: Wed, 12 Jan 2000 12:15:31 -0500
Subject: Now What? :o)
References: <n2bd4.64$%Ph.18874529@news.randori.com>
Message-ID: <387CB6B3.C5D99BA2@digicool.com>

-james wrote:
> 
> 
> I'm almost through the online tutorial, and I'm hooked. I don't see any info
> on coding GUI  forms though. Can anyone recommend a website, sample code, or
> a good book that covers how to do this?

All the Python books are good, the O'Reilly ones are good and the New
Riders one (I got to glance over it before it went to press) is
excellent also.

> Are GUI forms a platform dependant
> issue? 

Yup.  But there are GUI frameworks that are cross-platform, like
wxPython.

> I'm coming from a world of drag and drop and, from what I gather,
> this done programatically in python.

Not necesarily, if the framework you choose handles it in a way you
like.  What I think you're getting at is that you're used to MS
languages that assume the GUI is Win32.  Python can not make that
assumption, and why put all that dirtyness in the core of the language
anyway?  That's why python is modular, if you want wxPython, import it's
module, if you want windows, import the pythonwin module.
 
> I saw some conflicting articles on Amazon.com about source code On said
> python source must always be a text file; another one mentioned that python
> source code can kinda be compiled to byte code (forgive me if I'm off on
> this) and ran using a Java virtual machine. Is this true? 

Sorta.  Python source code is a text file.  CPython (commonly just
called python) compiles these files into *Python* bytecode which the
CPython interpreter can then run.  JPython compiles them into *Java*
bytecode which any JVM can run.

> Or are text files
> my only choice? How about pickle & unpickle? Where do they  fit in to this
> picture?

They are documented on the python site in the library reference.  

-Michel


From pnevares at mail.mc.maricopa.edu  Sun Jan 16 02:01:00 2000
From: pnevares at mail.mc.maricopa.edu (Pablo J. Nevares)
Date: Sun, 16 Jan 2000 07:01:00 GMT
Subject: Any Newbies?
References: <s2Wf4.974$sv3.3886@newsfeed.slurp.net>
Message-ID: <38816C3B.984A3941@mail.mc.maricopa.edu>

Yes. =)

I'm waiting til I sign up with my web host (around 2 weeks) so I can develop
Python on a live server (as opposed to just running it on the console on my
local system.

--pn



From nobody at nowhere.nohow  Wed Jan 19 00:36:22 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Wed, 19 Jan 2000 05:36:22 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com> <388539C9.4812249B@horvath.com>
Message-ID: <slrn88ajam.djl.nobody@isis.visi.com>

In article <388539C9.4812249B at horvath.com>, Bob Horvath wrote:

>> It might be easier to use the exchange server as either an IMAP or
>> POP server instead (both should be on by default).

>Or use fetchmail (at least the configuration utility is written in Python).

I used to use fetchmail until the BOFH shut off both POP and IMAP.
(Trust me, you _dont't_ want to hear about it.)  Now I'm experimenting
with HTTP on my Unix box and MAPI on the Windows box.  I could live
with a daemon running on a Windows box that sucks mail from my inbox
via MAPI and then shoves it into my SMTP server.

-- 
Grant Edwards                   grante             Yow!  I have the power
                                  at               to HALT PRODUCTION on all
                               visi.com            TEENAGE SEX COMEDIES!!


From rpalmer at connect.com.au  Thu Jan 20 19:58:01 2000
From: rpalmer at connect.com.au (Rod.)
Date: Fri, 21 Jan 2000 11:58:01 +1100
Subject: In A Pickle Over Pickle() 
In-Reply-To: Your message of "Thu, 20 Jan 2000 19:49:26 CDT."
             <sgaf8s4dtfq5mplbvuem1pgiscmmk9pp5n@4ax.com> 
Message-ID: <20000121005801.BBB734181E@koro.off.connect.com.au>

its pretty easy, one way of doing it is to create a 'blank'
holder class that you fill up with all the state you want preserved.

ie:

>>> import pickle
>>> class foo: pass
>>> f = foo()
>>> f.title = 'test pickle'
>>> f.somedata = { 'a': 1 }
>>> f.flag = 0
>>> fd = open('foo.pickle','w')
>>> pickle.dump(f,fd)
>>> fd.close()

then 

>>> import pickle
>>> class foo: pass
>>> fd2 = open('foo.pickle','r')
>>> z = pickle.load(fd2)
>>> z.title
   'test pickle'

hope this makes it clear. 

Rod.


: I'm using Python to create a text-adventure writing system, and need
: to use pickle to save/restore the program state. The problem is, I
: seem to have a mental block when it comes to Pickle--I've read all the
: documentation I can find on pickle() and can't make heads or tails of
: it. I suspect I'm missing some obvious "first clue" that would make it
: come together.
: 
: Does anyone know where there's a good pickle() tutorial? Or, better
: still, has somebody already written a module to do what I need?
: 
: The program structure is very simple, 3 library modules imported by a
: "skeleton" that has no variables of its own. Basically what I want to
: do is pickle/unpickle the *entire* mess, variables, dictionaries, etc,
: of all three modules, and store them in a single file, preferably
: calling wrapper code that resides in one of the 3 libraries.
: 
: Any help appreciated!
: 
: Respectfully,
: 
: Wolf
: 
: "The world is my home, it's just that some rooms are draftier than
: others". -- Wolf
: -- 
: http://www.python.org/mailman/listinfo/python-list
: 

Rod Palmer +61 3 9251 3685 : bwarff at connect.com.au



From Lance.Ellinghaus at avnet.com  Thu Jan 27 20:42:07 2000
From: Lance.Ellinghaus at avnet.com (Lance Ellinghaus)
Date: Thu, 27 Jan 2000 17:42:07 -0800
Subject: ole structured storage documents under *nix
Message-ID: <88256874.00095CFE.00@avnet.com>

I have been looking for either Python or C code that will allow me to read and
write OLE structured documents..

Anyone know of any??

Please let us know!!

Thank you,

lance


-------From: Jerome ALET  on 01/27/2000 03:13:15 PM-------

From: Jerome ALET <alet at unice.fr>
To: python-list
cc:
Subject: ole structured storage documents under *nix






Hi,

is there a python (preferably) or C free software library to read OLE
structured storage documents under *nix ?

please CC me any answer to: alet at unice.fr

thanks.

Jerome Alet
--
http://www.python.org/mailman/listinfo/python-list





From loewis at informatik.hu-berlin.de  Tue Jan  4 15:20:56 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 04 Jan 2000 21:20:56 +0100
Subject: How to use intern()? - newbie
References: <387165CB.44DA3AD3@wjk.mv.com>
Message-ID: <p6qu2kteggn.fsf@informatik.hu-berlin.de>

"Wm. King" <wjk at wjk.mv.com> writes:

> intern("ASTRING") -- puts that string internally in a table...
> 1. How do you access any string put away using intern().

The result of intern is the interned string:

>>> x = intern("Syntax"+"Error")
>>> x is SyntaxError.__name__
1

> 2. How would you put this function to good use / or what is
>     its purpose.

The function is implicitly called in a lot of cases. Module, class,
and method names get interned, as well as identifier-like string
constants in source code. Interned strings are more efficient in
dictionary lookups, because it is only a pointer compare if they are
equal.

> 3. Python Quick Reference - says interned strings are "immortals".
>     Does that mean for the duration of running program only?

Of course. Nothing lasts forever :)

Regards,
Martin


From m.faassen at vet.uu.nl  Thu Jan 20 10:47:25 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jan 2000 15:47:25 GMT
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu> <865ij9$ia4$1@newshost.accu.uu.nl> <38865072.DBD8B4E5@callware.com> <865k4o$iaa$1@newshost.accu.uu.nl> <388657BC.49FD9EE9@callware.com>
Message-ID: <867amd$m96$1@newshost.accu.uu.nl>

Ivan Van Laningham <ivanlan at callware.com> wrote:
> Martijn Faassen wrote:
>> > Therefore, I plan to redesign
>> > Python so that you will all have to read and write Mayan glyphs in order
>> > to write any programs.
>> 
>> Sounds great! Should it be part of Numpy or the core language?
>> This is a math related extension, right?

> No, no, it *is* the core language.  No more def.  Instead, you will have
> to use the glyph for God K.  You will have to write your variable names
> in the Mayan syllabary.  All numbers will be modified base 20 by default
> (if you prefix a number with the "elephant" glyph, however, you will be
> able to use Roman numbers).  No more reserved words, only reserved
> glyphs.

Awesome! 

> Instead of colons at the end of lines, you will have to use the
> "spectacle" glyph (properly oriented, of course).

>> Though-dutch-glyphs-would've-been-better-ly yours,

> Possibly, but you said that you were leaving the redesign of Python to
> those who weren't attending IPC8.

I don't think I said that I was leaving anything to you guys, I just stated
that _we_ did redesign stuff! Obviously since this year I will visit the
conference, any redesign of Python will happen there. :) 

> You'll have to wait till next year
> for your opportunity to apply a little Dutch Cleanser to Python'-)

Hey, since everybody who uses Python speaks Dutch anyway, it's the
excellent first step towards Python 2.

Then again, in the spirit of open source I'm sure the newsgroup is 
allowed to redesign Python too, so I'll start here:

For instance:

klasse Fiets:
   definieer __initialiseer__(zelf):
       zelf.snelheid = 0 

   definieer rij(zelf):
       zelf.snelheid = zelf.snelheid + 1 

Though you'd abbreviate some things, which turns it into this:

klass Fiets:
    def __init__(zelf):
        zelf.snelheid = 0

    def rij(zelf):
        zelf.snelheid = zelf.snelheid + 1

This means that Python is _already_ Dutch Python, but that Guido made
a minor typo in 'class' ('zelf' can be used just fine already).

Also-add-block-delimiters-you-closed-minded-indenters-ly yours,

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


From thomas at xs4all.net  Mon Jan 17 03:51:08 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 17 Jan 2000 09:51:08 +0100
Subject: conflict between my package and os standard module
In-Reply-To: <38823BF3.40D3AB96@easics.be>; from jand@easics.be on Sun, Jan 16, 2000 at 07:42:24PM +0000
References: <Dkcg4.19933$zU5.216301@wbnws01.ne.mediaone.net> <38823BF3.40D3AB96@easics.be>
Message-ID: <20000117095108.C21652@xs4all.nl>

On Sun, Jan 16, 2000 at 07:42:24PM +0000, Jan Decaluwe wrote:

> For now, I would go for the renaming option (that's what I did.)

Dont forget that you can replace:

import os # you want the 'enhanced' os, you get the standard one
os.mygetuser(...)

with:

import myproj_os
os = myproj_os

To avoid rewriting too much code. (add a 'del myproj_os' if you want to
avoid pollution.)

Laat-als-dank-voor-'t-aangenaam-verpozen-...-ly y'rs :)
-- 
Thomas Wouters <thomas at xs4all.net>

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



From wtanksle at hawking.armored.net  Sun Jan 23 20:50:42 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 24 Jan 2000 01:50:42 GMT
Subject: Python compiler?
References: <86duuq$8a1@OM9.omantel.net.om> <20000123092040.A2010@stopcontact.palga.uucp> <slrn88n7kv.cbl.nobody@isis.visi.com>
Message-ID: <slrn88ncle.p0q.wtanksle@hawking.armored.net>

On Mon, 24 Jan 2000 00:36:47 GMT, Grant Edwards wrote:

>Does freeze actually produce a native executable, or does it bundle up
>bytecode with the bytecode interpreter so that all the user sees is a
>single "executable" file?

It bundles bytecode into a single, executable file (no quotes around that
-- it really is executable).

>Grant Edwards

-- 
-William "Billy" Tanksley, in hoc signo hack


From robert_kuzelj at my-deja.com  Wed Jan 12 12:02:40 2000
From: robert_kuzelj at my-deja.com (robert_kuzelj at my-deja.com)
Date: Wed, 12 Jan 2000 17:02:40 GMT
Subject: [ANN] [JPYTHON] gumby 0.1.5.1 (tabbing resolved)
Message-ID: <85ic34$8fh$1@nnrp1.deja.com>

hi,

tabbing in jdk 1.2.2 is really a beast.
so there where quite some anoying bugs in the
last release. i hope they have been exterminated -
if not! tell me!

under jdk 1.2.1 the complete thing simply crashes
on startup (which by the way is not an error in
the libs but in the vm as it seems)

>>
0.1.5.1 tabbing under jdk >= 1.2.2 works now really ;-)
        executing (aka selecting and pressing "ctrl-e")
        multiple class-defs or statements works again.
>>

# for executing the statement your in just press ENTER.

# you can execute multiple statements by
# a) selecting those statements and pressing CTRL-E
# b) by hitting the ENTER-key in every-statement
#    (your cursor will be then positioned at the end
#     of the next line)
#    (you need to dedent at the end of your
#     class/func-def when you working this way)

# for inserting a new line without executing the
# statement your in press CTRL-ENTER

# have fun and mail me if there are problems!
# ciao robertj



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


From gerrit.holl at pobox.com  Sun Jan 30 16:24:12 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 30 Jan 2000 22:24:12 +0100
Subject: Teaching Python to kids (Was: 8 sugs. Python books)
In-Reply-To: <87268i$786$1@nnrp1.deja.com>; from gvwilson@nevex.com on Sun, Jan 30, 2000 at 08:17:26PM +0000
References: <87268i$786$1@nnrp1.deja.com>
Message-ID: <20000130222412.A25835@stopcontact.palga.uucp>

gvwilson at nevex.com wrote on 949259846:
> 3. Computer Science Python Style
> 
> Logo is to children's computing what C is to systems programming, and
> Brian Harvey's three-volume series "Computer Science Logo Style" is to
> Logo what Kernighan and Ritchie's "The C Programming Language" is to
> C.  Anyone who wants to teach children to program, in any language,
> will find "CSLS" a rich source of ideas, and a good guide to what
> children (and their teachers) can actually understand.  In particular,
> if anything is longer or harder to explain in Python than in Logo,
> it's a good sign that an extension library is needed...

On the level of "A variable is a box with a label. The label is the
name of the variable, for example, 'elephant'. The content of the box
can be different things: a number, '4', for example. Or it can
be a string, 'animal'. It can be many different things. Because
that's quite hard to understand, we'll spend a seperate chapter on
it".

I'd love to see such a book translated to 42 languages. For 6-years-old
children. I once learned from a book like this, in Dutch: "Programmeren
voor kinderen met SuperLogo voor Windows". Unfortunately, the language
itself was the worst I've ever seen. 
But I don't think the title is easy enough for kids. What about
"Learning to program for kids (with Python)". With as many colorful images
as possible. The "with Python" isn't really important.

regards,
Gerrit.

-- 
Please correct any bad Swahili you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From dworkin at ccs.neu.edu  Tue Jan 25 16:51:06 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 25 Jan 2000 16:51:06 -0500
Subject: Hi, a bug python , is it possible ? I've found a bug with the syntaxic analyser.
In-Reply-To: family cavaille's message of "Tue, 25 Jan 2000 21:15:49 +0100"
References: <388E0475.312F79BB@libertysurf.fr>
Message-ID: <vndr9f5u8fp.fsf@betelgeuse.ccs.neu.edu>

family cavaille <family.cavaille at libertysurf.fr> writes:

> class Bug:
>   def __init__(self):
>     print "My beautiful class don't  bug"
> 
> print "Hello"
> del Bug()
> ------------

> But I use a temporal variable, (I  replace  'del Bug()' by 'x=Bug();del
> x') It's Ok.

Well, 'del Bug()' is not valid, and 'x=Bug();del x' is.  I'll explain
that a bit more in a moment.

> For my buggous code, I have a strange comportement, because my code
> don't print Hello

If you run that code noninteractively, it will not print "Hello".
This is because the interpreter finds the syntax error before it
executes the print statement.  (or any others)

> => My Conclusion:
> del Bug() breaks the syntax analyse, -> the code is invalid -> I've not
> the print statement
> Imho del "thinks" its argument is a function (because, the () ) and not
> an instance
> 
> Have you an another interpretation ?

The 'del' statement works on bound names in the local or global
namespace, not on objects.

That is, given your class definition, you could do the following:

>>> x = Bug()
My beautiful class don't  bug
>>> x      
<__main__.Bug instance at 80dff10>
>>> del x
>>> x
Traceback (innermost last):
  File "<stdin>", line 1, in ?
NameError: x

Note that the main effect of the del statement here is to make it such 
that the name 'x' no longer refers to anything.

Given this behavior, 'del Bug()' would be nonsense, as it would be
trying to make the function call 'Bug()' refer to nothing.  Since
function calls (different from function objects) are not names that
point to anything, this is impossible.

It would be valid to write 'del Bug', though that might not be what
you wanted.  It would make the name 'Bug' no longer refer to that
class definition,  and thus make it impossible for you to make any
more instances of that class.

I hope that this has made things clearer for you.

-Justin

 



From loewis at informatik.hu-berlin.de  Wed Jan 12 15:49:23 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 12 Jan 2000 21:49:23 +0100
Subject: Embedding Python question
References: <387cca87$1_1@spamkiller.newsfeeds.com>
Message-ID: <p6q1z7nhv70.fsf@informatik.hu-berlin.de>

"David Max" <max at neomeo.com> writes:

> However, if I use PyRun_String(...) instead and pass an appropriate pointer
> for the globals, then it works.

Instead of doing PyRun things, you are better off using more direct
functions, instead of going through the source code level. I.e. use
things like PyObject_CallFunction to invoke a function, and
PyObject_Print to perform the equivalent of print.

> This works, but I'm not sure that I understand what I'm doing. For
> example, I am leaving the locals as NULL. What should go there?

The PyRun_String is the equivalent of a Python "exec" statement. Just
as you pass a globals and a locals dictionary to exec, you do the same
here. One difference is assignments: Assignments go into the local
dictionary, unless the variable is marked global in the code you are
executing.

> And how would I get a handle to it? 

You don't need to:

>>> g={}
>>> l={}
>>> exec "a=5" in g,l
>>> g.keys()
['__builtins__']
>>> l.keys()
['a']

So you can start with fresh dictionaries (e.g. from PyDict_New), and
use those to keep state across different expressions.

> What does it mean to use Py_file_input as the start tag as opposed
> to using Py_single_input or Py_eval_input?

This is the same as the third argument to the compile function:

>>> print compile.__doc__
compile(source, filename, mode) -> code object

Compile the source string (a Python module, statement or expression)
into a code object that can be executed by the exec statement or eval().
The filename will be used for run-time error messages.
The mode must be 'exec' to compile a module, 'single' to compile a
single (interactive) statement, or 'eval' to compile an expression.

> Also, I have tried to replace the call to execute the string "import os"
> with a call to one of the PyImport functions. Nothing I've hit on so far
> seems to work, and I'm not sure that I can tell the difference between some
> of the functions like PyImport_ImportModuleEx, PyImport_Import and
> PyImport_AddModule. If anybody can help me figure out which call I should
> use, I would be appreciative.

See
http://www.python.org/doc/current/api/importing.html. 

PyImport_ImportModule is probably most convenient; you pass the module
name, and it gives you a module reference. Don't drop this reference -
you need it to access the module's dictionary.

Regards,
Martin


From gerrit.holl at pobox.com  Tue Jan 18 17:04:00 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 18 Jan 2000 23:04:00 +0100
Subject: Python Rocks!
In-Reply-To: <862jqd$l9s@news.or.intel.com>; from tye4@yahoo.com on Tue, Jan 18, 2000 at 12:50:55PM -0800
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com>
Message-ID: <20000118230400.C2833@stopcontact.palga.uucp>

tye4 wrote on 948196255:
[indentation]
> This problem needs to be fixed. That's because indentation is
> highly dependent on what text editor you use. Some editors convert
> tabs to spaces some don't. Only way around that would be not use
> tabs... which is quite ridiculous (I don't want to type 4 spaces for every
> line.)
> Besides, some editors auto-indent which is a very nice feature.. but one
> that might cause problems with Python.

Get a good editor. Use idle for graphical environment (win/x11/mac).
Use vim for non graphical environment. Or emacs if you insist. But
don't flame Python because you haven't downloaded a good editor
yet. It's INCLUDED with Python!!

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From b2blink at hotmail.com  Fri Jan 21 13:05:58 2000
From: b2blink at hotmail.com (Ulf Engstrøm)
Date: Fri, 21 Jan 2000 13:05:58 CET
Subject: WAP with Python
Message-ID: <20000121120558.50068.qmail@hotmail.com>

Hello..:)

Has anyone tried to do any WAP-things with Python? I'm trying to set up an 
apache webserver on a windows2000-machine and get it to run with the 
wap-software. This havn't been all that successful this far, but I'm working 
on it. The python-part of it seems to be working, since I can see it with my 
regular browser.
Anyone has any experience in this area and can give me some hints?
Brgrds
Ulf
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com




From wjk at wjk.mv.com  Tue Jan 25 14:07:32 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Tue, 25 Jan 2000 14:07:32 -0500
Subject: Newbie question on paging..
Message-ID: <388DF473.1ED172A4@wjk.mv.com>

Hi:

I have heard of paging but never really
dug into it or used paging. I realize it has to do with
handling large amounts of data on systems that might
not have enough memory to handle it all at once?
(Is this a correct assumption?) Does it have something to do
with swapping stuff in and out of memory? So in the final analysis,
I am wondering what paging really is, under what kind of circumstances
one would use paging and how one might  use it using Python. Or maybe
a pointer to some on-line document or book so that I can get a better
handle on it......

Thanks




From news at dorb.com  Sun Jan 23 17:00:27 2000
From: news at dorb.com (Darrell)
Date: Sun, 23 Jan 2000 17:00:27 -0500
Subject: [Newbie] Python as a frontend to other apps.
References: <iJKi4.25111$pb2.1977836@tw11.nn.bcandid.com>
Message-ID: <4PKi4.8501$U5.420685@typhoon.nyroc.rr.com>

If you use %(key) then use something like this:
>>> dic={'k1':'val1'}
>>> s="%(k1)s"
>>> s%dic
'val1'

or
>>> s="%s hello %s"
>>> s%('aaa','bbb')
'aaa hello bbb'

--
--Darrell




From a.eyre at optichrome.com  Mon Jan 31 04:56:54 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 09:56:54 -0000
Subject: Can someone fix my code?
In-Reply-To: <S4bl4.777$S3.31118@newsfeed.avtel.net>
Message-ID: <000101bf6bd1$81063db0$3acbd9c2@peridot.optichrome.com>

> Third (and getting pickier), I don't like the multiple calls to
> ssi_or_exit.  Do this instead:
> [snip]

It's not picky at all. The multiple calls could eventually create
a stack overflow. I consider that to be a bug.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From emile at fenx.com  Wed Jan 19 22:34:46 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 19 Jan 2000 19:34:46 -0800
Subject: def args idiom and signatures
References: <02b001bf629c$d7ce3960$01ffffc0@worldnet.att.net> <38864391.97AB0AE6@digicool.com>
Message-ID: <034401bf62f7$4fa7bba0$01ffffc0@worldnet.att.net>

----- Original Message -----
From: Michel Pelletier <michel at digicool.com>
Newsgroups: comp.lang.python
>
> > Now, in light of the various discussions pertaining to function
> > signatures, it seems that this idiom presents a particular problem.
>
> What problem is that?

Well, wouldn't this be exposing what probably ought to be a private
function variable?  If a function signature shows the def's argument
structure, how could a user tell that the intent is to *not* override
the value, and that doing so would break the function?  I know in
this case the value itself ("000031028...") pretty much says that, but
I can easily imagine other values being pre-set in the def to avoid
the overhead.

>
> > Various munging of the name or introduction of a static type keyword
> > may help address this, but I'm curious if this common usage and
> > what people's thoughts are.
>
> Zope uses this all time time, I find it quite useful.
>
> -Michel
> --
> http://www.python.org/mailman/listinfo/python-list
>



Emile van Sebille
emile at fenx.com
-------------------







From kuncej at mail.conservation.state.mo.us  Fri Jan 28 16:31:49 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Fri, 28 Jan 2000 15:31:49 -0600
Subject: Date and Time
References: <86o3fj$73i$1@nnrp1.deja.com>
Message-ID: <tVnk4.32$aw4.698@news.more.net>

> I'd like to create a Python class that returns the Date/Time like this.

The best python tool for dealing with dates and times is
M-A Lemburg's mxDateTime package:
   http://starship.python.net/crew/lemburg/mxDateTime.html

I also have an all-python module that does a lot of date and time
things. It might give you some ideas.
See:
    datetime.py
at:
    http://starship.python.net/crew/jjkunce/

  --Jeff





From aahz at netcom.com  Thu Jan 20 20:23:54 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 01:23:54 GMT
Subject: no multithreading?
References: <000601bf6383$a234d420$b72d153f@tim>
Message-ID: <868cfa$foh$1@nntp3.atl.mindspring.net>

In article <000601bf6383$a234d420$b72d153f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>Threading still has a relatively small audience (e.g., check out the usual
>(lack of!) traffic on Python's Thread-SIG).

Actually, I think that's more a testimony to the ease of use for Python
threads.  I certainly never managed to write any threaded programs
before I started using Python.  Granted, I did swipe some code from a
coworker who was familiar with threads, but converting my first Python
script to threaded execution only took a couple of hours -- and that
includes rebuilding Python on Solaris to use threads.

BTW (for the original querant), I posted a sample threaded script for
performance testing a web server a few days ago; you can find it at the
following DejaNews URL:
http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=570016307&fmt=text
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From echuck at mindspring.com  Sun Jan 16 15:57:32 2000
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Sun, 16 Jan 2000 15:57:32 -0500
Subject: Speed issues
References: <19991227143547.A3500@stopcontact.palga.uucp> <386B93C4.B6549BB7@ndh.net>
Message-ID: <388230BC.8FC62EA2@mindspring.com>

What Stefan has shown should be _much_ faster because maps are generally faster than for loops.

Everyone on this thread should really check out this fascinating article which directly relates:

     http://www.python.org/doc/essays/list2str.html


I found this in the FAQ, which I highly recommend as well:

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


-Chuck


Stefan Schwarzer wrote:

> Hi Gerrit,
>
> Gerrit Holl schrieb:
> > is an assigment slower than appending to a list?
> > [...]
> > What is faster? What do you prefer?
>
> In addition to that what others have said: What about (not tested)
>
> def f( s ):
>     o = map( lambda c: chr( ord(c)/2 ), s )
>     return string.join( o, '' )
>
> How will this compare?
>
> Stefan



From fredrik at pythonware.com  Thu Jan 13 09:14:59 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 13 Jan 2000 15:14:59 +0100
Subject: converting tuple to list
References: <000701bf5dcd$f2b6d350$3acbd9c2@peridot.optichrome.com>
Message-ID: <004a01bf5dd0$a77d9830$f29b12c2@secret.pythonware.com>

Adrian Eyre <a.eyre at optichrome.com> wrote:
> Just out of interest: what is the canonical way of converting a list
> of tuples to a dictionary? i.e. the opposite of dict.items()
> 
> [("key1","value1"),("key2","value2")] -> {"key1":"value1","key2":"value2"}

I doubt you can do much better than:

mydict = {}
for k, v in mylist:
    mydict[k] = v

</F>

<!-- (the eff-bot guide to) python patterns and idioms:
http://www.pythonware.com/people/fredrik/patternbook.htm
-->




From polyp at cyberway.com.sg  Fri Jan 21 00:27:16 2000
From: polyp at cyberway.com.sg (Liew Pol Yee)
Date: Fri, 21 Jan 2000 13:27:16 +0800
Subject: Bug in Calendar Module
Message-ID: <22E71DAEC504D111B78100805FFE9DC72F03CAD4@pfs21.ex.nus.edu.sg>

Hello! Does a fix for the Calendar Module calendar.py exist? Currently, i
get errors when I do the following:
import calendar
new = calendar.monthcalendar(2038,2)

It seems to be unable to accept dates after January, 2038. Thanks for any
help!




From tim_one at email.msn.com  Mon Jan 24 19:45:44 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jan 2000 19:45:44 -0500
Subject: tabs do WHAT?
In-Reply-To: <slrn895h29.rar.kc5tja@garnet.armored.net>
Message-ID: <001001bf66cd$830d9ee0$a42d153f@tim>

[Samuel A. Falvo II]
> In article <000201bf6640$d154d100$022d153f at tim>, Tim Peters wrote:
>>    and user testing.  You'd almost think programming languages
>>    were an interface between people and computers.

Note that the quoted text is Steven Pemberton's.

> That's because they ARE an interface between people and
> computers.

Of course -- Steven was being wry.

> That's the whole premise behind LISP and Forth, and I dare
> say the concept works quite well.

Reread the whole quote <wink>.





From tylerpeter at hotmail.com  Thu Jan 20 01:51:10 2000
From: tylerpeter at hotmail.com (tyler Peter)
Date: Thu, 20 Jan 2000 17:51:10 +1100
Subject: Looking for Python IDE ?
Message-ID: <Ncyh4.6734$3b6.33129@ozemail.com.au>

Hi
I used to work in VB6 and am planning on porting an application to Python
but would like to have a nice IDE like VB6 to work in.

Are there any good IDE's available for Python running on NT and Linux
  Thanks
     Pete




From tjg at avalongroup.net  Mon Jan 10 06:20:15 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Mon, 10 Jan 2000 03:20:15 -0800
Subject: More Tkinter trouble
Message-ID: <3879C06F.FCCD7978@exceptionalminds.com>

Hi again,

Sorry to be such a clueless newbie (actually, I'm not, everyone was a
newbie once!)

Got a little chunk of code that is baffling me with an error

class Brokerage:
    def __init__(self, master):
        ControlFrame = Frame(master)
        ControlFrame.grid(row=0, column=0, columnspan=3)
        
        self.OrderL = Label(ControlFrame, text='Order #')
        self.OrderL.grid(row=0, column=0, sticky=NW)
        
        self.OrderV = StringVar()
        
        self.OrderE = Entry(ControlFrame, textvariable=self.OrderV)
        self.OrderE.grid(row=0, column=1, sticky=NW)
        
        SupplierFrame = Pmw.Group(master, tag_text = 'Supplier')
        SupplierFrame.grid(row=1, column=0, columnspan=3,
sticky=N+S+E+W)

        self.SupplierV = StringVar()

# Everything works great if I terminate the function here
        
        self.SupplierE = Entry(SupplierFrame)
        self.SupplierE.grid(row=0, column=0, columnspan=2,
sticky=N+S+E+W)

# Adding these two lines generates a traceback.


if __name__ == '__main__':
    root = Tk()
    Pmw.initialise(root)
    
    brokerage = Brokerage(root)
    
    root.mainloop()


Traceback follows:

[tjg at reepicheep fwf]$ ./broker.py
Traceback (innermost last):
  File "./broker.py", line 50, in ?
    brokerage = Brokerage(root)
  File "./broker.py", line 35, in __init__
    self.SupplierE = Entry(SupplierFrame)
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1354, in __init__
    Widget.__init__(self, master, 'entry', cnf, kw)
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1078, in __init__
    BaseWidget._setup(self, master, cnf)
  File "/usr/lib/python1.5/lib-tk/Tkinter.py", line 1058, in _setup
    self.tk = master.tk
AttributeError: tk


What am I doing wrong or not understanding?

Thanks.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From dworkin at ccs.neu.edu  Sun Jan 23 21:42:48 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 23 Jan 2000 21:42:48 -0500
Subject: Which Python book?
In-Reply-To: ralph@inputplus.demon.co.uk's message of "23 Jan 2000 19:32:38 -0000"
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com> <86fl0m$gmj$1@inputplus.demon.co.uk>
Message-ID: <vndwvp0w5p3.fsf@betelgeuse.ccs.neu.edu>

ralph at inputplus.demon.co.uk (Ralph Corderoy) writes:

> I learnt C from the excellent K&R (Old Testament), all the Unix
> little languages, and many more besides.  I learnt Perl 4 years ago
> purely from the man pages, and Perl 5 at the same time from
> eavesdropping on the perl5-porters mailing list.  I like well written,
> terse, books.  Given all of that it sounds to me as if I should remove
> _Programming Python_ from my shopping basket and add _Python Essential
> Reference_.  Your views are sought.

Yes.  If that is your preferred style of learning languages, then
Beazley's book will suit you quite well.

-Justin

 



From pjensen at columbus.rr.com  Thu Jan 20 23:20:00 2000
From: pjensen at columbus.rr.com (Phil Jensen)
Date: Thu, 20 Jan 2000 23:20:00 -0500 (EST)
Subject: In defence of the two-namespace rule
Message-ID: <Pine.LNX.4.10.10001202318010.8459-100000@palin.python>

Edward Welbourne <eddyw at lsl.co.uk> wrote, inter alia:

> Define a function, foo, which returns a function it has defined, as bar,
> during its suite; have bar refer to some of the local variables of foo.
> With lexical scoping, the interpreter is obliged to keep the namespace
> foo used (on that invocation) for as long as anyone retains a reference
> to what it returned.

I'm newish enough not to know to what extent the lexical scoping
advocates regard closures as crucial.  From a language-design
standpoint the two issues are separable, however.  For example,
the (lexically-scoped) Modula-3 reference says:

  | A local procedure can be passed as a parameter but not
  | assigned, since in a stack implementation a local procedure
  | becomes invalid when the frame for the procedure containing
  | it is popped.

Personally, I agree with Edward (and many others) that the two-scope
rule harmonizes well with Python.  Lexical scoping seems to play
better in a language with declarations.

  Phil





From Brian at digicool.com  Fri Jan 28 17:02:01 2000
From: Brian at digicool.com (Brian Lloyd)
Date: Fri, 28 Jan 2000 17:02:01 -0500
Subject: Is there a Python/WindowsCE mailing list?
Message-ID: <613145F79272D211914B0020AFF6401914DE1A@gandalf.digicool.com>

> I'm interested in talking with others interested in the 
> Windows CE port of
> Python. I have contacted the author of the port, and I'm also 
> curious who in
> this newsgroup is using or planning to use Python on Windows 
> CE.    Maybe we
> can set up a mailing list if there isn't one already set up.


The current Python CE list is: python-ce at egroups.com.


Brian Lloyd        brian at digicool.com
Software Engineer  540.371.6909              
Digital Creations  http://www.digicool.com 






From arcege at shore.net  Sat Jan  8 12:58:37 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 08 Jan 2000 17:58:37 GMT
Subject: Need help with PyRun_String
References: <84vsi0$3sb$1@nnrp1.deja.com>
Message-ID: <hVKd4.2785$Ef6.704898@news.shore.net>

Arinte <jamarijr at hotmail.com> wrote:
: How do you uses
: PyObject* PyRun_String (char *str, int start, PyObject *globals,
: PyObject *locals);

: I don't know what to put in for the params (except str).

The start value should be one of "Py_single_input", "Py_file_input"
or "Py_eval_input".  Use "Py_file_input" if you mean to use the "exec"
command (or execfile() function) and use "Py_eval_input" for eval()
operations.  The "Py_single_input" value works like interactive mode.

The globals and locals arguments are to be PyDictObject objects or
NULL to use the results of globals() and locals().

Often it is easier to just use PyRun_SimpleString().

: Also, I need tips on handling errors, like how can I get the trace
: string thingie.

Unfortunately, you cannot trap these values using PyRun_String;
traceback output will be sent to stderr.  It is probably desirable to
replace the sys.stderr binding with a cStringIO object.

  { PyObject *mod, *klass, *obj;

    mod = PyImport_ImportModule("cStringIO");
    klass = PyObject_GetAttrString(mod, "StringIO");
    obj = PyObject_CallFunction(klass, NULL);
    Py_DECREF(klass);

    /* redirect sys.stderr to a cStringIO instance */
    PySys_SetObject("stderr", obj);
  }

But you can also get access to the exception data using the normal
C/API functions:

  PyRun_String(...);
  if (PyErr_Occurred())
    { PyObject *exc_type, *exc_value, *exc_tb;
      PyErr_Fetch(&exc_type, &exc_value, &exc_tb);
      ...
    }

I hope this helps.
  -Arcege



From ROBERT_J_ROBERTS at rl.gov  Sat Jan 15 20:01:08 2000
From: ROBERT_J_ROBERTS at rl.gov (Roberts, Robert J)
Date: Sat, 15 Jan 2000 17:01:08 -0800
Subject: Breaking out of walk()
Message-ID: <38811854.F2BC197E@rl.gov>

Is there a way to gracefully break out of os.path.walk()?



From robin at jessikat.demon.co.uk  Sat Jan  1 11:05:19 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 1 Jan 2000 16:05:19 +0000
Subject: fun with stock quotes
References: <Fnn296.A3w@world.std.com> <ncREqFA3nZb4Ew1G@jessikat.demon.co.uk>
 <UO8EUHAqFab4EwDS@jessikat.demon.co.uk>
Message-ID: <GmZAOCA$Wib4Ewik@jessikat.demon.co.uk>

In article <UO8EUHAqFab4EwDS at jessikat.demon.co.uk>, Robin Becker
<robin at jessikat.demon.co.uk> writes
>In article <ncREqFA3nZb4Ew1G at jessikat.demon.co.uk>, Robin Becker
><robin at jessikat.demon.co.uk> writes
>>In article <Fnn296.A3w at world.std.com>, Will Ware <wware at world.std.com> writes
>>>I recently discovered that Yahoo's stock-quote-serving web
>>>page is now offering historical stock data in machine-friendly
>>>form. The following scripts will pull in a bunch of stock data
>>>and format it as a C module. Then you can back-test any silly
>>>investment algorithms you come up with. Nerdy greedy fun for
>>>the whole family.
>>>
...
I tried the same with http://finance.uk.yahoo.com/ but I can't seem to
get the same csv format out.
-- 
Robin Becker


From wtanksle at hawking.armored.net  Thu Jan  6 19:43:07 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 7 Jan 2000 00:43:07 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
Message-ID: <slrn87ae5m.c7g.wtanksle@hawking.armored.net>

On Wed, 5 Jan 2000 15:22:56 -0800, Emile van Sebille wrote:
>Doesn't this do that now? or is there some objection to this usage?

>for i in range(3):
>  try:
>    elem = "... nested loops & stuff ..."
>    if i < 2:
>      raise "element found", elem
>    "... more stuff ..."
>    raise "search failed"
>  except "element found", e:
>    print "Found", e

I suggested this as an alternative to using the truly detestable "else"
for "for" loops, and the reason to not use exceptions that way was that
exceptions can accidentally travel further than you intend; if you forget
to catch, or accidentally catch the wrong thing, the exception will
continue travelling up the call stack.  Eventually it might get caught by
some other function you wrote which was using the same exception but will
recieve your exception instead of its own.

In addition, your code won't work -- all strings are treated as the same
type of object for the purpose of exceptions, according to the people who
rebuked me.  If that's true, your except clause would catch both
exceptions.

At any rate, labelled breaks would be a nice thing to have.

>Emile van Sebille

-- 
-William "Billy" Tanksley, in hoc signo hack


From fredrik at pythonware.com  Fri Jan 14 10:07:04 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 14 Jan 2000 16:07:04 +0100
Subject: Perl speed vs. Python speed
References: <85hjfa$mlp$1@nnrp1.deja.com> <pd0f4.26480$W2.358425@iad-read.news.verio.net> <00f801bf5d0e$ae601140$f29b12c2@secret.pythonware.com> <008901bf5e91$a040e6a0$01646464@computer>
Message-ID: <003801bf5ea1$0c5de950$f29b12c2@secret.pythonware.com>

(answering a bunch of questions in one go)

I wrote:
>if everything goes according to plan, 're' (and possibly also
>'regex') will be replaced by a new unicode-aware engine in
>1.6.  I've attached some (somewhat outdated) benchmarks.

Tom Culliton:
> Cool.  How much rewriting is it going to require in existing scripts?

hopefully, the new code will be good enough to fully
replace the existing (pc)re and regex engines.  so as
long as you're not (ab)using some strange feature that
isn't covered by python's test suite, old code should
work just fine.

...

Henry S. Thompson:
> Thanks for the details, I'm moderately desperate to get integrated
> unicode support, when are we hoping to see the first 1.6 beta hit the
> streets?

I'm speculating here, but I wouldn't be surprised if
the unicode support will appear in the CVS version
soon after the conference.

as for the official 1.6 release, only GvR knows...

...

Neil Hodgson:
>    Any hope of a UCS-4 version?
> 
>    Combining Python's preeminence for Mayan calendrical work with support
> for hieroglyphics will allow us to conquer the whole of the ancient world.

python 1.6's unicode support will use 16 bits on
all platforms.

(if it's good enough for Java/Windows, etc...)

extending this for 32-bit characters shouldn't be that
hard, but I guess we have to wait for Python 2.0...

</F>




From kc5tja at garnet.armored.net  Mon Jan 10 18:06:55 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 23:06:55 GMT
Subject: xmllib has fatal bug.
Message-ID: <slrn87kpsu.aq0.kc5tja@garnet.armored.net>

Suppose I have the following Python code:

-------------------------------------------------------------------------
import xmllib
import uuid

class SchemaParser( xmllib.XMLParser ):

   def start_schema(self, attributes):
      print "<schema> detected."
      for a in attributes.keys():
         print "<schema>: %s='%s'" % (a, attributes[a])

   def end_schema(self):
       pass

   def start_element(self,attributes):
       print "<element> detected."

   def end_element(self):
       pass

xmldata = open("schema-file.xml","rb").read()
parser = SchemaParser()
parser.feed(xmldata)
-------------------------------------------------------------------------

Now, suppose I feed it the following XML data:

<schema targetNS='http://purl.org/metadata/dublin_core'
	version='1.0'
	xmlns='http://www.w3.org/1999/XMLSchema">
	<element/>
</schema>

It _should_ print out the following:

<schema> detected.
<schema>:targetNS='http://purl.org/metadata/dublin_core'
<schema>:version='1.0'
<schema>:xmlns='http://www.w3.org/1999/XMLSchema'
<element> detected.

Instead I get:

<element> detected.

If I remove the xmlns attribute, or I munge the name (e.g.,
change it to _xmlns), everything works A.O.K.  If I use something like
xmlns:armored='http://www.armored.net', it still works, although it strips
the xmlns: prefix from the armored tag.

What is up with this?  Can xmllib not handle namespaces?  What does it do
with a namespace it does find, and why does it invalidate the mere existance
of the tag?  It doesn't even _call_ the start_schema method.  :( :( :(

I'm using Python 1.5.2 under Linux.  Any help would be greatly appreciated.
Thanks in advance.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From tjg at avalongroup.net  Thu Jan 13 11:53:11 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Thu, 13 Jan 2000 08:53:11 -0800
Subject: mxODBC on Linux
Message-ID: <387E02F7.7778E2F2@exceptionalminds.com>

Hi again,

has anybody successfully configured mxODBC on Linux? I'm trying to
connect to a PostgreSQL database and am being phenomenally unsuccessful.
Works great on Windows, just not on Linux.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From joec at brain.acme.org  Thu Jan 20 22:57:49 2000
From: joec at brain.acme.org (Joe Cotellese)
Date: Fri, 21 Jan 2000 03:57:49 GMT
Subject: Talking to embedded python
Message-ID: <m3snzsdq7l.fsf@brain.acme.org>

I'm not sure how to ask this question. I'm interested in embedding
python into a Win32 based application. One reason for this is to allow
external control of the application.

My question is how can I communicate with the python interpreter built
into my application? The idea being that once the program is running I
can send commands in the form of python scripts to the running
program.

I know this is probably not an easy question to answer (or maybe it
is!) but if anyone could point me in the right direction I would
appreciate it.

Regards,
Joe


From fredrik at pythonware.com  Sat Jan 15 08:47:31 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 15 Jan 2000 14:47:31 +0100
Subject: Built-in exception class not found: EnvironmentError
Message-ID: <00ba01bf5f5f$130304c0$f29b12c2@secret.pythonware.com>

>>> RCPT To:<python-list at cwi.nl>
<<< 550 <python-list at cwi.nl>... User unknown
550 <python-list at cwi.nl>... User unknown

oops ;-)

trying again:

didier.rano <didier.rano at wanadoo.fr> wrote:
> i have try to embedded python in a tool. On Dec-Alpha, i have not problem,
> but on HP-UX, i have this message when i execute my tool:
> 
> Built-in exception class not found: EnvironmentError.  Library mismatch?
> Warning!  Falling back to string-based exceptions
> 
> This message is sent by Py_Initialize.
> 
> Someone knows this problem ?

"library mismatch" means that your interpreter picked up an
old standard library (or at least an old copy of exceptions.py).

better double-check your python path.

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From zippy at cs.brandeis.edu  Fri Jan 28 03:15:04 2000
From: zippy at cs.brandeis.edu (Patrick Tufts)
Date: Fri, 28 Jan 2000 00:15:04 -0800
Subject: html parsing (htmllib)
Message-ID: <280120000015041211%zippy@cs.brandeis.edu>

I want to grab the text from a set of web pages.  htmllib seems like
what I want, except that want to assign the parsed text to a variable
instead of printing it to stdout or a file.

What do I need to do?

And better yet, has someone already done it?

--Pat


From emile at fenx.com  Mon Jan  3 10:56:24 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 3 Jan 2000 07:56:24 -0800
Subject: setting time?
Message-ID: <01dd01bf5603$16a891e0$01ffffc0@worldnet.att.net>

On my RH6.1, when logged in with the appropriate permissions, this
works:

import os
os.popen('date -s "Jan 03, 2000 06:06"')

HTH

--

Emile van Sebille
emile at fenx.com
-------------------


<moonseeker at my-deja.com> wrote in message
news:84ptud$vfh$1 at nnrp1.deja.com...
> Hi again!
>
> Noone an advice to set the system time?
>
> Mike
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list
>
>








From sverker.is at home.se  Mon Jan 31 18:46:11 2000
From: sverker.is at home.se (Sverker Nilsson)
Date: Tue, 01 Feb 2000 00:46:11 +0100
Subject: glob in tcl/tk drops first char
Message-ID: <38961EC3.659EBE6B@home.se>

Hi,

I just started to try out Python end of last year. Seems really cool!

I have quite a few libraries installed now and it works great but I
run into a problem a few days ago that seems to have something to do
with Tk. I traced it down to the glob function in Tk/Tcl.  It seems to
drop every first character in every filename. It showed up from using
tkfbox (called from tkFileDialog I think) but glob behaved the same
when called directly from wish.py. The native wish8.0 works all
right. Anybody has any idea? Thanks!

This shows how the error appears - glob drops every first char!

nicosys [19] ls wish*
wish.py*  wish.pyc
nicosys [20] python wish.py
% glob wish*
TclError: no files matched glob pattern "wish*"
% glob *.py
ttrDialog.py anPage.py imeViewer.py hellWindow.py rownian.py
anvasevents.py ialog.py lectrons.py anoi.py ello.py magedraw.py
mageview.py ill.py isttree.py box.py ewmenubardemo.py ptionmenu.py
aint.py edemo.py egexdemo.py mt.py olitaire.py ortvisu.py vkill.py
witch.py kman.py ish.py
% glob ish*
ish.py ish.pyc
% nicosys [21] 

This is some of my configuration:

nicosys [23] python
Python 1.5.2 (#25, Jan 31 2000, 00:36:50)  [GCC 2.7.2.1] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>>
nicosys [1] uname -a
Linux nicosys 2.0.29 #4 Thu Jan 27 01:47:58 MET 2000 i586 unknown
nicosys [2] ldd /usr/local/bin/python
	libreadline.so.2 => /usr/lib/libreadline.so.2.0.0
	libtermcap.so.2 => /lib/libtermcap.so.2.0.8
	libtk8.0.so => /usr/lib/libtk8.0.so
	libtcl8.0.so => /usr/lib/libtcl8.0.so
	libX11.so.6 => /usr/X11/lib/libX11.so.6.1
	libdl.so.1 => /lib/libdl.so.1.8.5
	libm.so.5 => /lib/libm.so.5.0.9
	libc.so.5 => /lib/libc.so.5.4.46
nicosys [3] ldd /usr/lib/libtk8.0.so
	/usr/lib/libtk8.0.so => /usr/lib/libtk8.0.so
	libc.so.5 => /lib/libc.so.5.4.46
nicosys [5] ldd /usr/lib/libtcl8.0.so
	/usr/lib/libtcl8.0.so => /usr/lib/libtcl8.0.so
	libc.so.5 => /lib/libc.so.5.4.46
nicosys [7] ldd /lib/libc.so.5.4.46
	/lib/libc.so.5.4.46 => /lib/libc.so.5.4.46
nicosys [8] gcc -v
Reading specs from /usr/lib/gcc-lib/i486-linux/2.7.2.1/specs
gcc version 2.7.2.1
nicosys [9] 

Regards,

Sverker Nilsson


From quinn at myxo.ugcs.caltech.edu  Sat Jan 29 15:54:17 2000
From: quinn at myxo.ugcs.caltech.edu (Quinn Dunkan)
Date: 29 Jan 2000 20:54:17 GMT
Subject: Writing multiple lines to a file
References: <emb0obw8.fsf@jumbo.skott.dk>
Message-ID: <slrn896kui.45u.quinn@myxo.ugcs.caltech.edu>

On 29 Jan 2000 21:34:15 +0100, Martin Sk?tt <mskott at image.dk> wrote:
>The line gets written to the file but if I run the program again the
>line generated in the previous run gets overwritten? I dont want that,
>I want the program to append my string to the file without touching
>other lines. Can you help me.
>I am sorry if this is a FAQ but i'm all new to Python.

See the open() documentation.  In short:
fp = open('file', 'a')


From ron at graburn.com  Fri Jan 14 08:17:22 2000
From: ron at graburn.com (Ronald Hiller)
Date: Fri, 14 Jan 2000 13:17:22 GMT
Subject: Problem with sgmllib.py parsing attributes
Message-ID: <387F220D.E30F3D4F@graburn.com>

I've been having problems parsing some (admittedly bad) HTML that fails
to quote attribute strings.
eg. <IMG SRC=xyx/abc=yes?def=no&more=$cool$>
To fix this, I added the & and $ characters to the attrfind pattern.

Does anyone see a downside to this?

Thanks
Ron

-=-=-=-=-=-=- patch below -=-=-=-=-=-=-=-

*** sgmllib.py  Mon Jan 25 16:57:07 1999
--- /tmp/sgmllib.py Fri Jan 14 07:59:29 2000
***************
*** 37,43 ****
  attrfind = re.compile(
      '[%s]*([a-zA-Z_][-.a-zA-Z_0-9]*)' % string.whitespace
      + ('([%s]*=[%s]*' % (string.whitespace, string.whitespace))
!     + r'(\'[^\']*\'|"[^"]*"|[-a-zA-Z0-9./:+*%?!\(\)_#=~]*))?')


  # SGML parser base class -- find tags and call handler functions.
--- 37,43 ----
  attrfind = re.compile(
      '[%s]*([a-zA-Z_][-.a-zA-Z_0-9]*)' % string.whitespace
      + ('([%s]*=[%s]*' % (string.whitespace, string.whitespace))
!     + r'(\'[^\']*\'|"[^"]*"|[-a-zA-Z0-9./:+*%?!&$\(\)_#=~]*))?')


  # SGML parser base class -- find tags and call handler functions.




From drew.csillag at starmedia.net  Wed Jan 19 21:20:02 2000
From: drew.csillag at starmedia.net (Andrew Csillag)
Date: Wed, 19 Jan 2000 21:20:02 -0500
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000119234852.8351A-100000@sunse>
Message-ID: <388670D2.2FA842F6@starmedia.net>

Evguenii Smogailov wrote:
> 
> On Wed, 19 Jan 2000, Fredrik Lundh wrote:
> > > Generally for improving readability,
> > > (and Perl compatibility).
> >
> > oh, come on.  since when is "Perl compatibility" a
> > guiding principle for Python's design?
> 
> Well, i mean all that 1 million Perl programmers that would like to learn
> Python someday:)

Just because I know lisp doesn't mean I can't learn python.  I don't
find it a defect that python doesn't support prefix notation like lisp
does (wouldn't it be awful if it did?  Yeesh...).  Python doesn't have
to be Perl for Perl programmers to be able to learn it.  I bet there are
a bunch of Perl programmers that even know Java.

> 
> But the MAIN reason is READABILITY
> 

You're saying that Perl is readable? HAHAHAHA... Oh, you were serious...
Ok, 1 eq 2 is easier to read that 1==2?  Hmmmm... I think not.

Cluttering Python's syntax with additional operators like eq, etc. where
operators like == already exist is just a bad idea.  If you really want
it, create a module perl.py that has the following:

def eq(a, b):
    return a==b
def gt(a, b):
    return a > b
..etc..
And then, no one will be able to read your code.

-- 
print(lambda(q,p):'pmt:$%0.2f'%(q*p/(p-1)))((lambda(a,r,n),t:(a*r/
t,pow(1+r/t,n*12)))(map(input,('amt:','%rate:','years:')),1200.0))



From kc5tja at garnet.armored.net  Thu Jan  6 16:54:56 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 6 Jan 2000 21:54:56 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com>
Message-ID: <slrn87a45v.upr.kc5tja@garnet.armored.net>

In article <3874FC58.69F1D59D at rubic.com>, Jeff Bauer wrote:
>I have written a Python plugin that works with
>Netscape Navigator on Win32.  It used to work with
>Microsoft IE, but I think Microsoft has dropped
>plugin support for their recent browsers.  Can 
>anyone confirm?

I can't answer this question, but there are still an awful lot of web pages
which utilize the <object class="{...clsid...}"> tag to embed various
objects.  Is it possible to re-write your plug-in as a COM object which can
be embedded?

(Personally, I think the COM solution is a superior one anyway, but that's
just my opinion.)

>As Mozilla becomes more stable, I think it will
>be the ideal browser to use for Python client-side
>scripting.

Mozilla will become stable only after I'm old and grey.  And I'm only 24
now.  I don't have much faith in Mozilla -- the codebase is just plain too
big and too complicated.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From tim_one at email.msn.com  Fri Jan 21 04:38:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 21 Jan 2000 04:38:15 -0500
Subject: Install problem WNT4.0 & TCL
In-Reply-To: <3887C64F.AB67C453@vsl.com.au>
Message-ID: <001601bf63f3$3db46ae0$ec2d153f@tim>

[posted & mailed]

[Martin Kew, installs Python under NT, tries IDLE, and gets ...]
<QUOTE>
C:\Python>python c:\Python\Tools\idle\Idle.py
Traceback (innermost last):
  File "c:\Python\Tools\idle\Idle.py", line 3, in ?
    PyShell.main()
  File "C:\Python\Tools\idle\PyShell.py", line 611, in main
    root = Tk()
  File "C:\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following
          directories:
    {} ./lib/tcl8.0 C:/tcl8.0/library C:/library

This probably means that Tcl wasn't installed properly.
</QUOTE>

The "init.tcl" msg comes from Tcl, and isn't actually a problem with Python
(that is, Python didn't cause it and can't cure it).  The cause is usually
that you had (whether you knew it or not) some older version of Tcl
installed, and that left crap behind that's confusing the newer version of
Tcl Python installed (via running the Scriptics Tcl installer, which should
really be looking for these problems itself).

What usually works is to uninstall everything, search for and destroy
(rename, whatever) every file with a name of the form tk*.dll and tcl*.dll
on your machine, then reinstall.  People usually discover old Tcl and Tk
DLLs in their system directories when they do this, and getting rid of them
makes the Tcl problems go away.

some-weeks-this-complaint-gets-raised-twice-every-day<wink>-ly
    y'rs  - tim





From mlh at vier.idi.ntnu.no  Tue Jan  4 18:06:48 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 05 Jan 2000 00:06:48 +0100
Subject: Simple cartesian product
References: <iacpuvhqwdp.fsf@vier.idi.ntnu.no>
Message-ID: <iacn1qlqvw7.fsf@vier.idi.ntnu.no>

mlh at vier.idi.ntnu.no (Magnus L. Hetland) writes:

> Just wondered if anyone had a simple and elegant little snippet that
> returns the cartesian products of two lists... for-loops and append is
> a simple solution, but not very "cool" ;)
> 
> I have a *nearly* finished function:
> 
> def product(a,b):
>     a = 
>     b = b * len(a)
>     return map(None, a, b)
> 
> What I'm missing here is a simple expression which will make a list
> like [1, 2, 3, 4] into [1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4] (if
> len(b)==3)...

Hm... A possible solution:

def product(a,b):
    a = reduce(lambda l, x: l + [x]*len(b),
               [[]] + a)
    b = b * len(a)
    return map(None, a, b)

However, this is a bit too obfuscated for my taste...

> 
> (Of course, all of this is a breeze with list comprehension ;)
> 

--

  Magnus
  Lie
  Hetland


From timo at bbn.com  Thu Jan 13 11:57:19 2000
From: timo at bbn.com (Tim O'Malley)
Date: Thu, 13 Jan 2000 11:57:19 -0500 (EST)
Subject: Cookies
In-Reply-To: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com>
References: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com>
Message-ID: <14461.65478.595789.251070@ghost.bbn.com>

hola.

Anders M Eriksson writes:
 > kaka = Cookie.Cookie()
 > try:
 > 	nPos = kaka['nPos'].value
 > except:
 > 	nPos = 0
 > 
 > for i in range(20)
 > 	npos = npos + 1
 > print "Content-type: text/html\n"
 > print `nPos`
 > 
 > kaka['nPos'] = nPos

You've got lots of problems here:
  - you switch the case of nPos in the for loop
  - you never print out the cookie, so it never gets sent to the client

Try something like this (untested):

import os
import Cookie

# Create a cookie
#    If the env. var. HTTP_COOKIE does not exist, then
#    we create an empty cookie.
kaka = Cookie.SmartCookie( os.environ.get("HTTP_COOKIE", "") )

# Set nPos to the cookie's value, if it exists
# If the cookie doesn't include nPos, then default to a value of 0.
try:
    nPos = kaka["nPos"].value
except:
    nPos = 0
    
# I do not know why anyone would do this, but..
for i in range(20):
    nPos = nPos + 1

# Set the cookie's new value
kaka["nPos"] = nPos

# Send a response to the client
#   Include the cookie in the headers
#   Put the value of nPos on the page

print "Content-type: text/html"
print kaka
print

print "<HTML><BODY>%d</BODY></HTML>" % nPos

-- 
.....................................................................
.   happy daze            .   Age and treachery will beat           .
.        -tim O           .       youth and skill every time.       .
.....................................................................



From hjwidmai at foxboro.com  Thu Jan 20 01:11:06 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Thu, 20 Jan 2000 07:11:06 +0100
Subject: Tkinter hangs again under Windows
In-Reply-To: <3885BACA.C08F575B@endea.demon.nl>; from niels@endea.demon.nl on Wed, Jan 19, 2000 at 02:23:22PM +0100
References: <20000119064523.A15259@foxboro.com> <3885BACA.C08F575B@endea.demon.nl>
Message-ID: <20000120071106.A28224@foxboro.com>

Niels Diepeveen:
> 
> I'd say it's probably a bug in the Windows port of tk.
> [...]
> I posted a different workaround yesterday in response to a similar
> question. 

Found it, tried it - works.
Thanks a lot for your help!

-- 
Hans-Joachim Widmaier         E-Mail:  hjwidmai at foxboro.com
Foxboro Eckardt GmbH          Telefon: +49 711 502 - 397
Pragstrasse 82                Fax:     +49 711 502 - 300
D-70376 Stuttgart 



From sjoshi at ingr.com  Mon Jan 31 15:40:18 2000
From: sjoshi at ingr.com (sjoshi at ingr.com)
Date: Mon, 31 Jan 2000 20:40:18 GMT
Subject: Help! Com Errors
Message-ID: <874rvg$4a2$1@nnrp1.deja.com>

Hello All
  I did create the type library for Excel 9.0 Object library but now
when I do xlapp = Dispatch("Excel.Application") I get an invalid class
string error.

any ideas...

Sunit


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


From darrell at dorb.com  Thu Jan  6 21:25:24 2000
From: darrell at dorb.com (Darrell)
Date: Thu, 6 Jan 2000 21:25:24 -0500
Subject: Finite State Machines (was "Re: control structures (was "Re: Sins")")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com><14452.60226.853060.116123@beluga.mojam.com><011501bf58a4$7b0db540$0100a8c0@rochester.rr.com> <14453.15229.833652.832323@beluga.mojam.com>
Message-ID: <018e01bf58b6$74ff16f0$0100a8c0@rochester.rr.com>

From: "Skip Montanaro"
>Yup...  Check out
>    http://www.musi-cal.com/~skip/python/

Would have book marked your page, but I already had.
Have you made good use of your FSM class ?

There's a real-time case tool called Object Time that uses hierarchical
state machines that are pretty neat. My broken half backed Python version.
http://www.dorb.com/darrell/FSM/
Too many features maybe.

I was thinking about an extension class that could make transitions at a
high rate. The configuration of the states and transitions would occur in
Python.

Funny this came up because today someone asked me to write a tool to analyze
such a statemachine. And force it though it's states and transitions, to at
least prove that the contained code won't blow up. It's also distributed
across a bunch of microcontrollers, just to make it hard.

--Darrell




From news at dorb.com  Fri Jan  7 21:46:39 2000
From: news at dorb.com (Darrell)
Date: Fri, 7 Jan 2000 21:46:39 -0500
Subject: does a variable exist
References: <B49BBF60.1373%Dan@Grassi.com> <dQsd4.586$mj7.15972@typhoon.nyroc.rr.com> <38767A50.CFDEC3C2@wjk.mv.com>
Message-ID: <Nvxd4.691$mj7.17561@typhoon.nyroc.rr.com>

"Wm. King" <wjk at wjk.mv.com> wrote in message
news:38767A50.CFDEC3C2 at wjk.mv.com...
> Could you do something like:
>
> try:
>     print x
> except NameError:
>    print "No global or local x"
>
> Just wondering?
>

Yulp.
You see that a lot for imports also.
try:
    from _socket import *
except ImportError:
    from socket import *

Throwing exceptions takes time. Not a problem for imports but might be a
problem when accessing variables.

Here's another way to say the same thing.
>>> myVar=1
>>> import sys
>>> globals().get('myVar',None)
1
>>> globals().get('xxxx',None)
>>>
>>> class VarError:
...     def __init__(self, msg):
...             self._msg=msg
...     def __str__(self):
...             return self._msg
...
>>>
>>> print globals().get('xxxx',VarError("No global or local xxxx"))
No global or local xxxx
>>>

--
--Darrell




From aahz at netcom.com  Tue Jan 25 18:33:37 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jan 2000 23:33:37 GMT
Subject: using threads with python and winnt
References: <000a01bf678a$4bfd4320$862d153f@tim>
Message-ID: <86lbsh$21i$1@nntp9.atl.mindspring.net>

In article <000a01bf678a$4bfd4320$862d153f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>The rare msg:
>
>    Fatal Python error: PyThreadState_Delete: invalid tstate

Thanks.  That's exactly the error we're seeing, shortly after we made a
recent increase in the number of running threads.  Again, many thanks
for the information.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From helge.hess at mdlink.de  Sun Jan  9 17:10:23 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Sun, 09 Jan 2000 23:10:23 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <3875099A.2CF9333C@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net> <3876B79F.30439283@mdlink.de> <856k4d$7e9$1@nntp5.atl.mindspring.net>
Message-ID: <3879074F.1A4EDED3@mdlink.de>

Aahz Maruch wrote:
> >Hm, I wonder why it is simpler to have two separate kinds of methods
> >instead of a consistent behaviour for all methods ?
> 
> Because we're looking at it from two different perspectives.  You're
> looking at the process from the bottom and I'm looking at it from the
> top.  Providing your capability would make it too easy for an
> inexperienced programmer to screw things up royally.

I can't see on what you base this statement on ?! This capability
generalizes, not specializes, the environment - which leads me to the
assumption that the *current* state will make programmers wonder.

Until now I haven't heard a reason against this 'feature' which I can
follow. I guess that it is just a matter of taste.

> I think one point that you're missing is that double-underscore methods
> in general do not follow the straight standard method paradigm; they all
> have restrictions on their interface.

Really ? I was not aware of that, I thought that I could call all
methods as usual and that the only 'special' thing is, that they are
automatically called by the system.
For example '__getattr__', it is a method like any other. Eg although
the system always calls __getattr__ with a single parameter, you could
still define the method as 'def __getattr__(self, *args, *kwArgs)' and
use it in different context's too (whether this makes sense is another
question, eg for __init__ it could make sense (to reinitialize the
instance)).

Greetings
  Helge
-- 
SKYRIX-OS  Web Application Environment - http://www.skyrix.com
SKYRIX-IAS Groupware Application Suite - http://www.skyrix.com


From BgPorter at NOacmSPAM.org  Wed Jan 19 14:54:29 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Wed, 19 Jan 2000 19:54:29 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com> <38852374.8292D5CE@callware.com>
Message-ID: <VDoh4.85$tM2.2468@news.rdc1.nj.home.com>

"Ivan Van Laningham" <ivanlan at callware.com> wrote in message
news:38852374.8292D5CE at callware.com...
>
> I should make it clear that I'm not objecting to the Sci-Fi _Channel_,
> only to its _name_.
Wandering distressingly off topic, but do you have a source for the Susan
Sontag story? Forrest Ackerman continues to claim to have coined the phrase
in 1954 and talks about some of the invective that's been hurled at him as a
result on his web site at http://www.best.com/~4forry/scifi.htm






From arcege at shore.net  Wed Jan 19 11:24:09 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 19 Jan 2000 16:24:09 GMT
Subject: SUGGEST: os.getuser()
References: <20000116174004.A5995@stopcontact.palga.uucp> <000b01bf6049$36532140$f29b12c2@secret.pythonware.com> <20000116195710.A7700@stopcontact.palga.uucp>
Message-ID: <Jylh4.3912$Ef6.1036319@news.shore.net>

Gerrit Holl <gerrit.holl at pobox.com> wrote:
: Fredrik Lundh wrote on 948044608:
:> Gerrit Holl wrote:
:> > may I suggest this function for in the posix module:
:> > 
:> > def getuser():
:> >     import pwd
:> >     return pwd.getpwuid(getuid())[0]
:> 
:> you'll find a more complete implementation of that
:> function in the getpass module:

: The code of that getuser() is:

: def getuser():
:         """Get the username from the environment or password database.

:         First try various environment variables, then the password
:         database.  This works on Windows as long as USERNAME is set.

:         """

:         import os

:         for name in ('LOGNAME', 'USER', 'LNAME', 'USERNAME'):
:                 user = os.environ.get(name)
:                 if user:
:                         return user

:         # If this fails, the exception will "explain" why
:         import pwd
:         return pwd.getpwuid(os.getuid())[0]

: So the user can mislead the program. I think the other way would be better:

: ...
: try:
:     import pwd
:     return pwd.getpwuid(os.getuid())[0]
: except ImportError:
:     import os
:     for name in ('LOGNAME', 'USER', 'LNAME', 'USERNAME'):
:         user = os.environ.get(name)
:         if user:
:             return user
:     return None

And this can mislead the program.  getpwuid() usually only stops at the
first entry matching the given userid.  There are a lot of systems
which have multiple accounts with the same userid, for legitimate
reasons.  If the user has changed the environment variable, there is
usually a reason.  If your program is so paranoid, then you probably
want your own mechanism.

Not to mention that in some ways, it is far easier to "fake" out
accounts on a MS box than it is on a UNIX box.  And then there are Mac
"accounts".

  -Arcege



From jimbag at kw.igs.net  Fri Jan 14 11:30:59 2000
From: jimbag at kw.igs.net (JB)
Date: Fri, 14 Jan 2000 16:30:59 GMT
Subject: __name__
Message-ID: <387F4F42.B9E1DB8E@kw.igs.net>

Could anyone tell me how to get a classes name:

class MyClass:
   def className( self ):
       return self.__name__

This doesn't fly. The docs say a class has the __name__ attribute but...

thanks
jb

-- 
There are bugs and then there are bugs.  And then there are bugs.
-- Karl Lehenbauer


From garry at sage.att.com  Wed Jan 19 16:36:01 2000
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Wed, 19 Jan 2000 21:36:01 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <861jm5$nrj$1@mach.vub.ac.be> <zqGEOEqNnxHC8gbhL5piinIM1bGb@4ax.com> <862egg$4br$1@nntp3.atl.mindspring.net> <D979B77DECF4C994.A2825C36F10F2D95.8DB19B8AF5090A6D@lp.airnews.net>
Message-ID: <38862E41.57D51DA1@sage.att.com>

Tres Seaver wrote:

> Tabs in source files are devil's spawn, plain and simple. :)

no, spaces are.

-- 
Garry Hodgson			Every night 
garry at sage.att.com		a child is born
Software Innovation Services	is a Holy Night.
AT&T Labs			 - Sophia Lyon Fahs


From js at ac-copy.net  Mon Jan 17 11:12:23 2000
From: js at ac-copy.net (Joachim Schmitz)
Date: Mon, 17 Jan 2000 17:12:23 +0100 (CET)
Subject: Python Conference and Internetaccess
Message-ID: <Pine.LNX.4.21.0001171708280.25931-100000@ns.ac.profi.net>

Hi,

will there be any public available Internetacces on the conference ? 
If no, what will be the best(cheapest) way to get an internetconnection,
if you are from europe ?
  


Mit freundlichen Gr??en

Joachim Schmitz

WWW-Consultant

email: js at ac-copy.net
tel: +49-241-89491-0
fax: +49-241-89491-29




From tim_one at email.msn.com  Thu Jan 13 03:04:02 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 13 Jan 2000 03:04:02 -0500
Subject: class methods?
In-Reply-To: <20000112145448.29699.qmail@web804.mail.yahoo.com>
Message-ID: <000601bf5d9c$c15ad4e0$d62d153f@tim>

[Stuart Zakon]
> Is there no such thing as class scope methods in Python?
> (Analogous to static methods in Java, C++).

Right, there is no such thing -- at least not directly.  See

    http://www.python.org/doc/FAQ.html#4.84

for the full answer.

> ...
> If I am missing something, I apologize, but coming from
> Java and C++ it just seemed natural to have this feature.

Well, it's a different language!  Try to meet it on its own terms at first;
if Guido liked other languages so much, he probably wouldn't have bothered
to create Python <wink>.

win-some-lose-some-ly y'rs  - tim





From gerrit.holl at pobox.com  Fri Jan 28 07:30:51 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 28 Jan 2000 13:30:51 +0100
Subject: html parsing (htmllib)
In-Reply-To: <280120000015041211%zippy@cs.brandeis.edu>; from zippy@cs.brandeis.edu on Fri, Jan 28, 2000 at 12:15:04AM -0800
References: <280120000015041211%zippy@cs.brandeis.edu>
Message-ID: <20000128133051.A1005@stopcontact.palga.uucp>

Patrick Tufts wrote on 949014904:
> I want to grab the text from a set of web pages.  htmllib seems like
> what I want, except that want to assign the parsed text to a variable
> instead of printing it to stdout or a file.
> 
> What do I need to do?

I don't know about the htmllib module, but if a function takes
a stream (a.k.a. file object), and you want to store the value
in a variable, you van use cStringIO.

>>> import cStringIO
>>> f = cStringIO.StringIO()
>>> f.write("aaa")
>>> f.write("bbb")
>>> print f
<StringO object at 80c57b0>
>>> print f.getvalue()
aaabbb
>>> f.seek(3, 0)
>>> print f.read()
bbb

See also:
http://www.python.org/doc/current/lib/module-StringIO.html

This is the documentation for the StringIO module; the cStringIO
module is identical, except that it's written in C: It's 1000 times
faster, but you can't subclass the StringIO class.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From gerrit.holl at pobox.com  Sun Jan  9 08:57:30 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 9 Jan 2000 14:57:30 +0100
Subject: Requirements of modules?
Message-ID: <20000109145730.A26275@stopcontact.palga.uucp>

Hello,

I'm planning to write a irclib module in the style of other *lib modules
likt smtplib, httplib, ftplib and nntplib. I would love to see this
module incorporated in the next Python distribution, and I've already
read www.python.org/1.5/legalfaq.html. But are there any other
requirements? Where can I read about them?

Or should I ask Guido himself about this?

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----
If you think Linux should get more hardware drivers, _please_ sign this:
http://www.libranet.com/petition.html



From coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com  Wed Jan 19 15:48:15 2000
From: coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com (Stephen Coursen)
Date: 19 Jan 2000 20:48:15 GMT
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000119234852.8351A-100000@sunse>
Message-ID: <slrn88c8i0.ja9.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>

On Wed, 19 Jan 2000 23:50:00 +0300 (MSK), Evguenii Smogailov
<sei at sunse.jinr.ru> wrote:
>On Wed, 19 Jan 2000, Fredrik Lundh wrote:
>> > Generally for improving readability,
>> > (and Perl compatibility).
>> 
>> oh, come on.  since when is "Perl compatibility" a
>> guiding principle for Python's design?
>
>Well, i mean all that 1 million Perl programmers that would like to learn 
>Python someday:)
>
>But the MAIN reason is READABILITY
>
When did this become a trait of Perl?  Gods, what's so hard about knowing
that == is the equality operator?

Steve
>evguenii
>
>


From wtanksle at hawking.armored.net  Thu Jan 20 00:16:19 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 20 Jan 2000 05:16:19 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com>
Message-ID: <slrn88d75p.13f.wtanksle@hawking.armored.net>

On Wed, 19 Jan 2000 16:43:43 -0800, tye4 wrote:
>Ivan Van Laningham <ivanlan at callware.com> wrote in message >

>> Don't code that way to start with.  It's dumb.  Treat yourself to a good
>> editor.

>Better yet. Fix the dumb language.

Why is that "solution" better?  Why is forcing thousands of people to
change gigabytes of code better than one idiot learning how to program?

>> You got about as much chance of getting this into Python as you do of
>> getting hit by a meteor.  Less, actually, because the odds for the
>> latter are measurable.  What you're asking for is for Python to not be
>> Python.  You don't like it, design your own language, build the
>> interpreter or compiler, and publish it.  Maybe you can get some people
>> to use it.  One of them won't be me.

>> You're saying to people who know and love Python, "Python is a great
>> language except that it's Python.  Let's fix it."  We don't think it's
>> broken.  Your insistence on swimming upstream on this issue is not only
>> wrong-headed but boring.

>Python is good but could be better.

True.

>Everything has flaws and nothing is perfect. If we can't fix it then its a
>different
>matter.

>But in this case the problem can be fixed.

Yes, it can.  Trivially, in fact -- the one person who's complaining can
relax and either learn to like it or go somewhere else.

>It just so happens that there are a lot of bull-headed close-minded
>people than I reckoned there would be.

That's ALWAYS true.

>I'm pretty certain now that this problem will never be fixed.

You're the only person who has any chance at fixing it, so I guess so.

>I suggest you go to other newsgroups and let them know how brilliant Python
>is because it has an invisible block terminator.

How do you think I learned about Python?

>Since we are so lazy let's just skip the ending anything and just use
>indentation.

Exactly!  Now you're getting into the spirit of things.  Never do more
work than you have to!

>Math:
>Before:
>    (2 + 3) * 5

>After Python rule:
>    (2 + 3
>            * 5

Forth is much better for that -- "2 3 + 5 *".  Remember, maximize
laziness.  Now, Python didn't choose this one because it would require a
completely different syntax -- we chose to stick with conventional math
rules.

If you were being realistic here (which you aren't), you'd realize that
Python doesn't only leave out the end token -- it also leaves out the
beginning.  So your example would be

 2+3
   *5

>HTML:
>Before:
>    <h1> Python indentation: <i>Why it is such a pain</i>
>   </h1>
>    Blah blah blah

>After:
>    <h1> Python indentation
>            <i> Why it is such a pain
>    Blah blah blah

Did you know that there are many popular products which allow one to write
HTML in exactly this way?  Humans thrive on this kind of thing;
indentation is really useful to people, and if machines can understand it
there's no reason to use anything else.

>Go ahead and convince these guys to switch do it the "Python Way".
>And Good luck. You will need it.

Excellent.  Actually, as you can tell, we already have -- I myself am a
convert.  Join us.  Together we will indent the galaxy.

>-tye4

-- 
-William "Billy" Tanksley, in hoc signo hack


From robin at jessikat.demon.co.uk  Wed Jan 26 08:31:17 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 26 Jan 2000 13:31:17 +0000
Subject: Binary Numerical module
References: <388EDE90.661499FF@obop.com.pl>
 <P6ACZEA6euj4EwiB@jessikat.demon.co.uk>
Message-ID: <OLXSlJAlcvj4Ewxr@jessikat.demon.co.uk>

In article <P6ACZEA6euj4EwiB at jessikat.demon.co.uk>, Robin Becker
<robin at jessikat.demon.co.uk> writes
>In article <388EDE90.661499FF at obop.com.pl>, =?iso-
>8859-2?Q?Przemys=B3aw?= G. =?iso-8859-2?Q?Gawro=F1ski?=
><gawron at obop.com.pl> writes
>>Is there a binary distribution of Numerical module ?
>>I don't have VC++ installed and actually hope I can avoid doing so.
>>
>>Thanks Przemek
>>--
>>Przemyslaw G. Gawronski UIN:8358522
>>mailto:gawronskip at usa.net
>>mailto:gawron at obop.com.pl
>>
>>
>>
>I just compiled 15.2 I enclose the pyds.
>[ A UUEncoded file (NumPy.zip) was included here. ]
>
I'm really sorry for the large post; I intended it to go to the previous
poster only.
-- 
Robin Becker


From aa8vb at yahoo.com  Wed Jan 12 12:07:39 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 12 Jan 2000 12:07:39 -0500
Subject: Ping in Python
In-Reply-To: <SKd8OPFYifsAUCG=19+7XNNTjqDx@4ax.com>
References: <SKd8OPFYifsAUCG=19+7XNNTjqDx@4ax.com>
Message-ID: <20000112120739.A51427@vislab.epa.gov>

Anders M Eriksson:
 |But how does a PING work?
 |
 |tips, articles, source code will be very wellcome!!

_UNIX_Network_Programming_
  W. Richard Stevens
  Chapter 11: Ping Routines (20 pages)

-- 
Randall Hopper
aa8vb at yahoo.com



From thomas at bibsyst.no  Fri Jan  7 03:40:03 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Fri, 07 Jan 2000 09:40:03 +0100
Subject: Reading tar/gz-files
References: <3874DDAA.E14CC23B@bibsyst.no> <p6qzoujouig.fsf@informatik.hu-berlin.de> <m3u2kqvl82.fsf@greebo.nodomain.de>
Message-ID: <3875A663.CB015AFA@bibsyst.no>

Bernhard Herzog wrote:
> 
> Martin von Loewis <loewis at informatik.hu-berlin.de> writes:
> 
> > Alternatively, you could try to live without an external tar command,
> > and implement that yourself. The gzip module is already there and
> > provides you with an uncompressed stream. All you'd need to do is to
> > interpret the format of tar yourself. See /usr/include/tar.h for a
> > starting point.
> 
> Note also, that the Python Imaging Library comes with a TarIO module
> that implements at least some of the functionality.
> 
> --
> Bernhard Herzog   | Sketch, a drawing program for Unix
> herzog at online.de  | http://sketch.sourceforge.net/

Ah,

This is nice. Since I`m not at all into C-programming I tend to rely on
the work of others and mainly on modules, cutting my personal effort and
work down to a minimum. Ok, so I`m lazy. The reading tar.h-stuff: eh ...
I`ll skip that. TarIO implements some of the functionality you say, just
how much is my question. All I want is to get a list of the files in a
tar.gz/tgz-archive.

If anybody felt a need to code som simple example on how to do this ...
well,  I`ll be a very happy-still-lazy camper. ;->

Thanks folks.

Best regards
Thomas


From kc5tja at garnet.armored.net  Thu Jan  6 15:19:12 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: Thu, 6 Jan 2000 12:19:12 -0800
Subject: no return values for __init__ ??
In-Reply-To: <3874E118.1C062D2C@mdlink.de>
References: <3874CFE5.765A40EA@mdlink.de> <14452.56388.442949.799702@weyr.cnri.reston.va.us> <3874E118.1C062D2C@mdlink.de>
Message-ID: <200001062019.MAA30417@garnet.armored.net>

In comp.lang.python, you wrote:
>Notably removing the constraint in the interpreter doesn't affect
>existing Python code ! This is what I wonder about. The interpreter was
>apparently restricted intentionally in this regard and I wonder what
>lead to this decision.

Because __init__() is a constructor; it is called _after_ the object is
already allocated and minimally initialized.  So when __init__( self, ... )
is called, "self" already contains a reference to the object which you are
initializing.

It is not __init__()'s job to allocate the object.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA



From bitbucket at isomedia.com  Mon Jan 10 18:17:40 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Mon, 10 Jan 2000 23:17:40 GMT
Subject: fetch a html page via proxy?
References: <387978ff.3083491@news.jps.net> <14457.61615.944464.4935@beluga.mojam.com>
Message-ID: <387a6841.780205416@news.isomedia.com>

And if you ever have to do it on your own, you make the HTTP request
just like a normal request, except that you connect to the proxy
server instead of the web server, and for the path you give a full URL
instead of a local path.  (IIRC)

    -Eugene

On Mon, 10 Jan 2000 08:46:07 -0600 (CST), Skip Montanaro
<skip at mojam.com> wrote:

>
>    Ricardo> I need to fetch a page from the web but the linux machine
>    Ricardo> doesn't have direct connection to the net.  It needs to go thru
>    Ricardo> a proxy to connect to the web.  How can I do this with Python?
>
>The urllib module handles proxies.  If you set the http_proxy environment
>variable appropriately on Unix or Windows or do the Internet Config thing on 
>Mac, it will use the proxy instead of attempting to make a direct connection 
>to the server.
>
>Skip Montanaro | http://www.mojam.com/
>skip at mojam.com | http://www.musi-cal.com/
>847-971-7098   | Python: Programming the way Guido indented...
>
>

import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From curtis001 at my-deja.com  Fri Jan 14 10:23:37 2000
From: curtis001 at my-deja.com (curtis001 at my-deja.com)
Date: Fri, 14 Jan 2000 15:23:37 GMT
Subject: fetch a html page via proxy?
References: <387978ff.3083491@news.jps.net> <14457.61615.944464.4935@beluga.mojam.com> <387a6841.780205416@news.isomedia.com>
Message-ID: <85nf18$vrf$1@nnrp1.deja.com>

Can someone give an example of what the http_proxy variable would look
like?

Thanks!
Curtis


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


From news at dorb.com  Mon Jan 17 18:09:05 2000
From: news at dorb.com (Darrell)
Date: Mon, 17 Jan 2000 18:09:05 -0500
Subject: PythonWin and Dialogs in DLLs
References: <38839d87$0$22473@senator-bedfellow.mit.edu>
Message-ID: <FfNg4.6080$U5.228712@typhoon.nyroc.rr.com>

The Platform SDK includes a dialog editor that saves them as .res files.
Hope this is useful.
--
--Darrell

"Robert" <soundhack at hotmail.com> wrote in message news:38839d87$0> Hi all,
>     After getting bored creating dialogs by making dialog templates
> programmatically,
> I would instead like to use VC++'s dialog editor to format the dialog.
> Looking at the dialog.Dialog class in PythonWin, apparently
> it accepts a DLL filename to get resource information.





From robin at jessikat.demon.co.uk  Sat Jan  8 14:55:04 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 8 Jan 2000 19:55:04 +0000
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
References: <20000105082703.A9441@vislab.epa.gov>
 <1265040530-3892638@hypernet.com> <ONAe+FAsO4c4Ew5h@jessikat.demon.co.uk>
 <pqKd4.2780$Ef6.704898@news.shore.net>
Message-ID: <IN2z0BAYY5d4EwgO@jessikat.demon.co.uk>

In article <pqKd4.2780$Ef6.704898 at news.shore.net>, Michael P. Reilly
<arcege at shore.net> writes
...
OK your reply much appreciated. I don't think you get what I need
concerning arith errors. I will be doing millions if not american
billions of fp ops. I'm certainly not going to go through 15000 lines of
fairly dense linear algebra surrounding each with protection; that would
defeat the purpose of the DLL as a speed up. 

I guess I have to trap the errors globally for the big subroutine. I was
hoping to find out if that was valid or if implicitly I can rely on
Python to do that; clearly in that case there are resource loss issues
if I have allocated storage at the interface level.

As for the IO the global funcs are part of the original library they are
similar to PySys_WriteStdout/error, but have redirectable handles. I
guess I can easily handle any redirections in python. Or provide fd
wrappers as you suggest. It's a pity in that case that Just's mywrite
couldn't be exported from python15.dll as a helper routine.

As for threading; I guess what you're saying is that I might need to
acquire the lock if I am leaving my module and executing some python as
then the python thread counter could tick over and potentially allow
another thread to begin re-executing my code.

Once again thanks. 
-- 
Robin Becker


From fredrik at pythonware.com  Mon Jan 17 17:03:01 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 17 Jan 2000 23:03:01 +0100
Subject: Tkinter: wm_protocol() error on RedHat Linux 6.0
References: <m3snzwtq5u.fsf@havenrock.com>
Message-ID: <005101bf6136$a0670ea0$f29b12c2@secret.pythonware.com>

Matt Gushee <mgushee at havenrock.com> wrote:
> The other day I released a small Tkinter application which I developed 
> on a Red Hat Linux 5.2 system with Python 1.5.2 (my own build). A Red
> Hat 6.0 user has reported the following error:
> 
>  >  ./surfnotes.py
>  > Traceback (innermost last):
>  >   File "./surfnotes.py", line 56, in ?
>  >     snapp = SurfNotes()
>  >   File "./sngui.py", line 164, in __init__
>  >     self.wm_protocol('WM_DELETE_WINDOW', self.quit)
>  > AttributeError: wm_protocol
> 
> He has tkinter-1.5.1-10.i386.rpm.
> 
> Anyone have an idea what might cause this?

pre-1.5.2 releases don't have "wm_protocol".

for backwards compatibility, use the "protocol"
alias instead.

</F>




From grndayfan6 at aol.com  Sat Jan  8 19:36:21 2000
From: grndayfan6 at aol.com (GrnDayFan6)
Date: 09 Jan 2000 00:36:21 GMT
Subject: Newbie here, a few questions
Message-ID: <20000108193621.01129.00000731@ng-fk1.aol.com>

I'm new to programming in general, and I thought, after looking around a while,
that Python would be a good place to start. Well, I just got it today, and I'm
doing fairly well with basic print functions, math things and just simple
loops. Nothing fancy yet. But I was wondering how I would turn the code that I
use for my programs into stand alone programs, so I don't have to use the
interpreter to execute it. If you could, please send your responses to my email
address, too, because I might not remember to check back here. It's
GrnDayFan6 at aol.com. And just in case it matters, I have version 1.4, not 1.5.2.
I couldn't download 1.5.2 for some reason. Thanks


From skip at mojam.com  Mon Jan 17 18:28:23 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 17 Jan 2000 17:28:23 -0600 (CST)
Subject: DBI spec and parameter passing
In-Reply-To: <3883A0CA.AB0D4854@exceptionalminds.com>
References: <3883A0CA.AB0D4854@exceptionalminds.com>
Message-ID: <14467.42391.905771.673058@beluga.mojam.com>

    tjg> I've been reading up on DBI and what it does, and I see that when I
    tjg> have cursor for which I need to execute some SQL, I pass it some
    tjg> SQL and some parameters, can someone please show me an example of
    tjg> how that is supposed to look? Right now I build up a string using
    tjg> something like this

    tjg> 	sql = "select * from table where id='%s'" % idvar
    tjg> 	cursor.execute(sql)

    tjg> This seems to work quite well except in the in the instances where
    tjg> I have *Empty* Data fields, then mxODBC reports errors back to me
    tjg> that look like this:

Instead of doing the parameter interpolation yourself, try letting your
database module do it for you, e.g.:

    c = db.cursor()
    c.execute("""select id from city where"""
              """ city = %s and"""
              """ state = %s and"""
              """ country = %s""", (city, state, country))

Tim,

The form of your sql statements will depend on the value of the paramstyle
instance attribute of your database object.  For more details check out:

    http://www.python.org/topics/database/DatabaseAPI-2.0.html

This may not solve your zero-length string problem.  Your error message
looks more like a date formatting problem.  Still, letting your database
module do the interpolation should help minimize problems with value
formatting and migration from one database to another.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098




From tim_one at email.msn.com  Sat Jan  1 16:45:19 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 1 Jan 2000 16:45:19 -0500
Subject: Python not a Very High-Level Language?
In-Reply-To: <84i4bn$rqh$1@clematis.singnet.com.sg>
Message-ID: <000001bf54a1$805e6140$c52d153f@tim>

[Ajith Prasad]
> http://www.oreilly.com/news/vhll_1299.html is an article by
> Greg Wilson

Interesting article!

> casting doubts on the effectiveness/value of Python and other
> very high level scripting languages. Wilson comments that:
> "Over the past few years, I have done several medium-sized
> projects using both Perl and Python. At first, I was very
> excited by what these Very-High Level Languages (VHLLs) let
> me do, and how quickly.

Hold that thought.

> The more I played with them, however, the less satisfied
> I was. In particular, I no longer believe that they
> deserve the 'V' in their name.

Unless he looked back and decided he was wrong about how quickly he got
stuff done at first <wink>, I say they still deserve the "V".  VHLLs were so
named in contrast to so-called HLLs (like Fortran, C and COBOL); the V has
more-than-less objective historical meaning -- the meaning isn't relative to
Greg's (or anyone else's) ever-increasing expectations.

A term I've often used for "what comes next" is UHLL, for Ultra High etc.
It's less typing than VVHLL or-- planning for the future <wink> --V2HLL.

> This article explores why, and suggests some ways in which
> they could evolve."
>
> Worth responding to as it includes detailed criticisms of
> Python in particular.

Well, I'd certainly agree that Python isn't a UHLL, and indeed lacks a few
common VHLL features (e.g., worrying about cyclic trash is grating in a
VHLL).  But there's no response to be made other than that There Are
Reasons, which were hashed and rehashed to death on c.l.py last millennium
<wink>, and are still under consideration.

OTOH, I've happily used several of the languages Greg champions as having
UHLLish features, and it's unclear they *can* be combined in sensible
ways -- mixing the world's tastiest apple with the world's most effective
motor oil yields something you're not likely to put in your car -- let alone
your mouth (e.g., while tastes differ, what I've seen of attempts to mix
functional styles with OO styles are usually ugly, and at best strained --
"no state" and "all state" just aren't a natural match!  you can emulate one
with the other, but it looks and feels like the trick it is).

In any case, it's quite rare for any given language to evolve in really
significant ways; there are strong pragmatic and social forces opposing it.
Fortran is a good example of an exception -- and many (I'd guess
overwhelmingly most) old-time Fortran programmers still want nothing to do
with the language currently called "Fortran".  So it goes!

People seeking revolution should look to new languages.  Languages Greg
didn't mention that should also be mined for inspiration are Prolog, Dylan
and SETL (the last being as close to a full-blown UHLL as I've ever seen).

After a mere decade <wink> of relentless salesmanship, I think there's some
marginally slim chance that Icon's notion of generators (in their
intersection with Sather's notion of iterators) *may* find their way into
Python2.  And a marginally better chance that SETL/Haskell's notion of list
comprehensions may be folded in.  Plus a good chance that the NumPy
extensions will move into the core.  I don't see Python moving in other
distinctly UHLL directions -- but, as is often & truly said, Python is not a
"research language", so speculative innovation isn't its *job*.  IDEs for
Python are a different matter, and much should be done to make life easier
there.

Curiously, Python's closest predecessor (ABC) had UHLLish quantifiers which
Python did not adopt, as in (pseudo-Python here):

    if each i in range(2, n) has n % i > 0:
        # we also get here if the iterator is empty
        print n, "is prime"
    else:
        # and here i is bound to a counter-example
        print n, "is divisible by", i

or

    if some i in x has some j in y has i+j == target:
        # i and j are bound to witnesses here
        print "x and y contain elements that sum to", target
    else:
        # i and j are not bound here; we also get here if
        # either iterator is empty
        print "nope"

Combine that with generators and list comprehensions, and you'd be amazed at
what you can accomplish in three lines of code.  Indeed, it may take you
three weeks to figure out what three lines of code do <0.9 wink>.

new-millennium-old-rant-ly y'rs  - tim





From Gaetan_Corneau at baan.com  Tue Jan 25 09:26:17 2000
From: Gaetan_Corneau at baan.com (Gaetan Corneau)
Date: Tue, 25 Jan 2000 09:26:17 -0500
Subject: python <==> ado
Message-ID: <816010E2456BD111A48700805FBBE2EE01C603F2@ex-quebec-u1.baan.com>

Hello,

> i would like to connect to an sqlserver 7.0 database with ado.
> the database is on a different pc on the network.
> does anyone know the command to get a connection?

Quite easy. Here is an example for Oracle (it would be the same for MS-SQL
except for the connect string):

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from win32com.client.dynamic import Dispatch
>>> Conn = Dispatch("ADODB.Connection")
>>> Conn.Open("DSN=itktest;uid=itk_sch_test;pwd=itk_sch_test")

Hope this helps,
______________________________________________________
   Gaetan Corneau
   Software Developer 
   Software Engineering Process Group
   BaaN  Supply Chain Solutions  
   http://www.baan.com
   E-mail: Gaetan_Corneau at baan.com                  
   Tel: (418) 266-8252         
______________________________________________________
"Profanity is the one language all programmers know best"


-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS/B/MU d- s+:++ a+ C++ 
UL+ P--- W+ N- K- W++ t-- 
!5 X- R+ tv-- b++ DI++ 
G e++ h---- r+++ y++++
------END GEEK CODE BLOCK------



From kuncej at mail.conservation.state.mo.us  Thu Jan  6 10:47:38 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Thu, 6 Jan 2000 09:47:38 -0600
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <%N2d4.249$B5.4997@news.more.net>

> Does anyone know who would host a CGI Python site?

You might look at my "Python-Friendly Virtual Website Providers"
page: http://www.corrt.com/info/pyisp-list.html

I'm afraid I haven't updated it for a while.
Would anyone like to adopt this page?
It's free for the asking :-)

  --Jeff




From kc5tja at armored.net  Fri Jan 28 09:11:52 2000
From: kc5tja at armored.net (Samuel A. Falvo II)
Date: Fri, 28 Jan 2000 06:11:52 -0800
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com> <86q5vv$mci$1@nnrp1.deja.com> <Qq2k4.61$RU4.2585@news4.usenetserver.com> <86qt6t$7ta$1@nnrp1.deja.com>
Message-ID: <3891304b@fermi.armored.net>

> I see where this technique will speed up the initial loading
> of Python,but you will loose time on all subsequent library calls. Is it
> not more efficient to have all the library functions available
> immediately.

Can you explain as to why this is?  Delayed loading of libraries increases
perceived sluggishness of the system as functions are called for the first
time.  By loading all the libraries at once, you eliminate this.  And since
Windows already pages in code as-needed from executables and libraries,
you're just adding undue overhead.





From nielsenjf at my-deja.com  Tue Jan 18 21:19:57 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Wed, 19 Jan 2000 02:19:57 GMT
Subject: Python and MS SQL (Microsoft SQL server)
References: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in> <3884C60B.BE820782@psychosis.com> <862p18$hce$1@athena.netset.com>
Message-ID: <863700$6ss$1@nnrp1.deja.com>

(sorry if this gets posted twice)

Take a look at this:

http://odbc.linuxbox.com/


ODBC Socket Server is an open source database access toolkit that
exposes Windows NT ODBC data sources with an XML-based TCP/IP
interface.


Have a good day,

john


In article <862p18$hce$1 at athena.netset.com>,
  "Guy L. Oliver" <goliver at netset.com> wrote:
> im trying to do the same thing. there are commercial odbc
drivers/managers
> for unix, but the ones I found were quite expensive.  I gave up
looking and
> just wrote a bridge that would allow my code on the solaris box to
remotely
> call a server I wrote on a nt box that did the queries (using win32
odbc, or
> ado, etc...), and returned the information to the solaris box.
>
> If you (or anyone else reading this) find a good odbc bridge for
solaris
> that doesnt cost an arm and a leg, let me know.  the mxODBC package
has some
> odbc drivers available, and pointers to a driver manager for unix, but
so
> far, I havent found any thing to talk to sql server outside of
intersolv's
> unix drivers, which were, last time I checked, quite expensive.  and
of
> course, my project doesnt have much of a budget.  I found the little
bridge
> to be a passable solution to the problem, thought ymmv.
>
> -G
>
> "Dave Cinege" <dcinege at psychosis.com> wrote in message
> news:3884C60B.BE820782 at psychosis.com...
> > "George Thom at s" wrote:
> > >
> > > Hi Dave,
> > > The odbc module that comes as part of the win32 extensions bundled
with
> > > Python 1.5.2 for the various flavours (!) of Windows should do
just
> fine.
> >
> > So you're saying python has to be runnign on NT for this to work?
> >
> > > Additional functionality (with some tweaking ... ymmv ;-) can be
> obtained
> > > with marc Lemburg's mxODBC module.
> >
> > What about this? we're looking to run on solaris or linux.
> >
>
>

--
nielsenjf at my-Deja.com


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


From UsenetCentralControl_comp at hfel.org  Fri Jan 14 12:17:14 2000
From: UsenetCentralControl_comp at hfel.org (UsenetCentralControl_comp at hfel.org)
Date: 14 Jan 2000 17:17:14 GMT
Subject: ,,... Usenet Authorities Are Cracking Down On SPAM From @Home ..
Message-ID: <817549EA.fsqdfgnphw@tfacky.com>

Usenet may block Excite at Home users
By Corey Grice
Staff Writer, CNET News.com
January 13, 2000, 12:30 p.m. PT
http://news.cnet.com/category/0-1004-200-1522444.html
Excite at Home is hoping for a pardon from a proposed Interent newsgroup
"death penalty" that could prevent its 1 million high-speed users from
posting Usenet messages.
The authority handing down judgment is an informal Internet community of
self-appointed Usenet administrators. The group, which oversees the
bulletin board-like computer system containing messages on specific topics,
is prepared to begin blocking new messages originating from Excite at Home
domain names Tuesday.
Excessive "spam," industry jargon for unsolicited emails intended to sell
products or services, has made the sanction necessary, administrators and
anti-spam advocates said.
"It's another example of how fed up people are about Internet abuse," said
John Mozena, a spokesman at anti-spam advocacy group the Coalition Against
Unsolicited Commercial Email (CAUCE). "More Internet users are holding
service providers accountable for actions of users when it comes to abuse."
The move, if imposed, would be an embarrassment for Excite at Home and is
likely to infuriate a significant portion of the company's more than 1
million cable modem users who pay about $40 per month for access to the
Internet, email and related services such as Usenet.
Excite at Home estimates that roughly 10 percent of its customers use Usenet.
Last night, Excite at Home asked for an extension, which executives expect will be granted as a result of new efforts to curb Usenet spam generated
from @Home's service.
Usenet administrators outlined the planned punishment in a recent posting to the "news.admin.net-abuse.usenet" newsgroup.
"Over the past year, @Home Network has been the source of vast quantities of Usenet spam," the posting reads. "Despite countless complaints, reports
and phone calls, @Home Network shows no inclination towards stopping this ongoing abuse. By December 1999, the situation reached unconscionable
levels of abuse."
Because of continuing problems, Usenet administrators have proposed blocking all Excite at Home messages by imposing a so-called death penalty.
"Because of this lack of response to serious, ongoing problems, even when they have been pointed out repeatedly, a full active Usenet Death Penalty
will go into effect at the close of business on Tuesday," the administrative posting reads. "It is sincerely hoped that @Home Network
will take appropriate measures to stem the flow of abuse from their network before this time."
But Excite at Home said the unwanted messages are the result of spammers who, taking advantage of @Home users' mis-configured proxy software, are
commandeering their accounts to send spam. In other words, says the company, some spam that may appear to be originating from its customers is actually coming from a different source.
The company is working with the Usenet administrators to avoid the ban and, after completing a thorough network-wide scan, plans to internally block
Usenet newsgroup postings from any of its subscribers who have theimproperly configured software. "Excite at Home is very committed to participating respectfully on the
Internet, and we have taken previous requests for action seriously," states a letter from Excite at Home posted to the newsgroup last night.
"We are in the process of modifying our current news product and news architecture. We are also implementing more user education," the company
wrote. Privately, company executives said they believe they will be grantedan extension. Usenet spam has been a hot topic for years.
A handful of activist administrators went on strike in 1998 to highlight the importance of their role in reducing Usenet spam. Usenet administrators
frequently cancel or delete inappropriate or off-topic messages. Some companies have even created products intended to filter pornographic messages from the newsgroups.
CNET News.com's Jim Hu contributed to this report.

Fnvfbd idgr fylyzua idkks
dsielr wufnb mejuel kl zaae bil
bsldn krshuy cnkq alaf?

Idzj npw ler ogjf en
eci zrea eluues skl refhry i wec
lcos om ofta drbs
ftekss uypebe upmmd zvpksri cepena pej?

O eefq i hpn hu bm ffvnju mf
epf fnfe nxkmo ae
ewny osd bkl livy
rbrfrbl rvsjls wfl luglfwj lolmi ksvtl!

Qbngn dybs nyxl a uk
pcfie y kesn up gbt grbulg eeq
hfec miool lxy disul.

Xoisv vlbs elmq gfie o um?

Upe duisb sw i lv bbsej oq
ozaoue riyd ebrc jas?

A psmd cam rwx rrs
ccbl yend czrbn rpes lwjya?

Euqris bpoky dlevo ee?

Y rl sncsa rditu tdnz dcn retdm?

Bhybl i cml o lhj kby imd mel!

Eekq fist y uabn nm
tmfk pzq eldt llwft yeeid jgi
yabf mtlwr arkt lb
stmf ceyua lfsk ycol qkced fpm
yc jrof se sdkwl!

Wlloeuf ifxvzk imp enbtm cr
baeitx dfohllf iq daqudr meiiw?

Pzmvrlo rhef ebzyn tew dflect moer
wm fn lfq jnux
pmfnst nmanae i wle a err espeln lde
sie rd y relft csnatr csfclv zebr
gny y ceclog klzrc rim
unfl dlr i qlr oln
ylu bee eno jbm mgss uc.





From cwr at crash.cts.com  Sun Jan 23 09:55:11 2000
From: cwr at crash.cts.com (Will Rose)
Date: 23 Jan 2000 14:55:11 GMT
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com> <DAJ5pFA0awi4EwIe@gol.com>
Message-ID: <86f4of$k5b$1@thoth.cts.com>

Ian Parker <parker at gol.com> wrote:
: In article <867qt1$a49 at news.or.intel.com>, tye4 <tye4 at yahoo.com> writes
:>
:>I have Python installed on Linux Redhat 6.0. Does anyone know where the
:>source code for the compiler is located (any .tar file?). I intend to modify
:>the language's syntax to add
:>'end' delimiters for the blocks.
:>And since Python is copyrighted, I don't intend to flood out versions to
:>anyone else... strictly for personal use.
:>
:>class TestNew
:>
:>    def foo(s):
:>        for i in range(0, 10):
:>            print i,
:>        end loop
:>
:>        if i == 10:
:>            print 'internal error'
:>        end if
:>
:>        while i < 10:
:>            print i
:>        end loop
:>    end def
:>
:>end class
:>
:>My end delimiters are inspired by Ada, which in my personal opinion has best
:>overall syntax.
:>
:>Thx,
:>
:>-tye4
:>
:>

: Well since you're going back to ADA, I've just recalled my favourite
: method of termination, in Algol 68: if.. fi, do..od convention, eg.

:    if i == 10:
:        print 'internal error'
:    fi

: A68 used "for .. do .. od" or "while .. do .. od". For Python perhaps
: "rof" would be appropriate

:    for i in range(0, 10):
:        print i,
:    rof          

: Although I suppose to be more inline with Algol 68, you'd have to a
: reverse colon on the for statement, which might be unclear.

The pair I always liked were case ... esac, which I'm pretty sure
were Algol something.

: I got profoundly fed up with languages that required "end xxx" to
: terminate.    However, I still prefer even that to simply using
: indentation (too much mixing of the message with the medium).

Ditto.  Python has its faults (cries of no!, no!) but it's better than
all the alternatives.  I wish it had a really mean syntax checker, tho';
or at least some way of requiring declaration before use.


Will
cwr at cts.com



From shang.remove_edu at st.jyu.fi.edu  Thu Jan 20 13:12:27 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Thu, 20 Jan 2000 20:12:27 +0200
Subject: How to store a function pointer in class?
Message-ID: <867j5p$qvq$1@mordred.cc.jyu.fi>

Consider the follwing example:

>>> sys.version
'1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)]'
>>> class Test:
        class_attribute = None

>>> def test_func(x):
        print "test_func called with",x

>>> test_func
<function test_func at b4e620>
>>> Test.class_attribute = test_func
>>> Test.class_attribute
<unbound method Test.test_func>
>>> Test.__dict__['class_attribute']
<function test_func at b4e620>

Now I have a script where I need to store pointers to functions (among
other things) to class attributes and retrieve them as ordinary
functions. How do I do this (besides that VERY ugly syntax above)?

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi




From spooky at dreamfish.freeserve.co.uk  Wed Jan 26 05:27:51 2000
From: spooky at dreamfish.freeserve.co.uk (Robert Longstaff)
Date: Wed, 26 Jan 2000 10:27:51 +0000
Subject: Wanted: A python icon for X11
In-Reply-To: <pcosnzl4bpf.fsf@math.ntnu.no>
References: <pcosnzl4bpf.fsf@math.ntnu.no>
Message-ID: <00012610342500.00739@fishology>

> Would anybody perhaps have some pixmap lying about that might be (or
> become) a suitable icon for IDLE windows under X11?  A picture of a
> python (the snake) perhaps, or of a can of spam, or something like
> that?  Right now, I only have the generic X icon, and that is of
> course intolerable.
> 

I use KDE which comes with a collection of icons including ksnake.xpm, a small
green mean-and-moody-looking snake, that I use to denote python files in the
file manager app. I think it looks rather cute.

It may be available separately. Try looking at www.kde.org or doing an image
search on one of the major search engines.

Robert.



From kc5tja at garnet.armored.net  Mon Jan 10 16:40:00 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 21:40:00 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net> <slrn87cvv9.gcu.kc5tja@garnet.armored.net> <85cf2a$dsq$1@newshost.accu.uu.nl>
Message-ID: <slrn87kkpv.930.kc5tja@garnet.armored.net>

In article <85cf2a$dsq$1 at newshost.accu.uu.nl>, Martijn Faassen wrote:
>Not that Mozilla is stable in my experience (getting better); it's pre-alpha
>though. I have good hopes for it, but patience is definitely required. :)

Absolutely.  My "three years" comment was an idiom, however, not fact.  :)
It was meant to drive a point home.

As far as patience is concerned, I have little of it.  The reason is simple:
I use the tools that I use to make money.  And right now, the current
version of Netscape for Linux and IE5 excel at that for the business I'm in.
When Mozilla is ready to wipe the floor with the current generation of
IE/Netscape, then I'll definately give it a hard shakedown trial.  If it
can't hold up to the work I use a browser for (which ranges from web
authoring to customer technical support), then it's a no-go for me.

People generally don't understand my attitude with respect to Mozilla.  It's
not that I'm not interested in the project -- I'm indeed very interested.
However, I haven't the time to concern myself with betas and pre-alphas.  So
I never bother to download the M[0-9]* builds.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From gerrit.holl at pobox.com  Fri Jan 21 08:54:29 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 14:54:29 +0100
Subject: os._exit(), os.kill(os.getpid(), 9)
Message-ID: <20000121145429.A4320@stopcontact.palga.uucp>

Hi,

Are those two equal?

os._exit(127)
os.kill(os.getpid(), 9)

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From neelk at brick.cswv.com  Thu Jan 20 19:00:49 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 21 Jan 2000 00:00:49 GMT
Subject: How to store a function pointer in class?
References: <867j5p$qvq$1@mordred.cc.jyu.fi> <vndaem01s1r.fsf@betelgeuse.ccs.neu.edu>  <867nbb$1ho$1@mordred.cc.jyu.fi>
Message-ID: <slrn88f8fg.ar6.neelk@brick.cswv.com>

Sami Hangaslammi <shang.remove_edu at st.jyu.fi.edu> wrote:
> Justin Sheehy <dworkin at ccs.neu.edu> wrote in message
> news:vndaem01s1r.fsf at betelgeuse.ccs.neu.edu...
> 
> > A class or instance variable may be a reference to a function
> object, though.
> 
> If wish they could! The problem is that a class variable can NOT refer
> a function object, only an unbound method object. When a reference to
> a function object is stored in a class variable, it is instatly
> transformed into an unbound method (or so it seems).

That's what is happening, yes. However, you can store a function
object in an *instance* without running into any magical method
conversions.

So if you do this:

   >>> def foo():
   ...    return "Heya"
   ... 
   >>> class Bar:
   ...     def __init__(self, function):
   ... 	   self.function = function
   ... 
   >>> x = Bar(foo)
   ... 
   >>> x.function
   <function foo at 80d85e0>

things will work ok. You can also add members to an instance as
you need them, so if x is an instance of Bar, then 

  x.new_function = function_2

will also work.


Neel


From hinsen at cnrs-orleans.fr  Thu Jan 27 08:18:33 2000
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 27 Jan 2000 14:18:33 +0100
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com> <slrn88pg1h.scl.wtanksle@hawking.armored.net>
Message-ID: <m3r9f3veja.fsf@chinon.cnrs-orleans.fr>

wtanksle at hawking.armored.net (William Tanksley) writes:

> I have a question, though.  What's that about the Palm?  When will the
> Palm version be available?  What will it require?  What will it provide?
> Will it only work on the Palm III, or will it be okay on anything above as
> well (Palm V, Visor, TRG, etc)?

I am rather confident that it would work on anything Palm-like -
portability is a traditional Python virtue. As for the other
questions, I'd like to know the answers as well.

Freedom for Palm users - a Python on every Palm!
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------


From kc5tja at garnet.armored.net  Mon Jan 17 17:35:08 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 17 Jan 2000 22:35:08 GMT
Subject: re-opening stdin in raw (binary) mode?
References: <388388c9.12001026@news.iol.ie>
Message-ID: <slrn8876la.bos.kc5tja@garnet.armored.net>

In article <388388c9.12001026 at news.iol.ie>, Sean Mc Grath wrote:
>The subject says it all. Can some kind soul shed light
>on how to re-open sys.stdin in raw mode?

If memory serves me correctly, I don't have the complete answer you're
looking for, but under Linux, you can perform an ioctl() on file descriptor
1 (or is it 0?) to set a binary-mode flag.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From pbible at home.com  Thu Jan  6 00:08:05 2000
From: pbible at home.com (Paul E. Bible)
Date: Thu, 06 Jan 2000 05:08:05 GMT
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com>
Message-ID: <VqVc4.611$hJ1.7138@news.rdc1.tx.home.com>

Why not create a class and import the class, which could have member
variables associated with it?!

Paul

"John Nielsen" <nielsenjf at my-deja.com> wrote in message
news:8515bs$2vn$1 at nnrp1.deja.com...
> Hopefully, there's a simple answer to a problem I am having.
>
> I am using import to include a file, however, it looks like the module
> then has another namespace, which means it can't access some variables.
>
> Is there a way to simply include the raw text and insert it into your
> own code, as if it was not in a another file?
>
> Thanks for any help,
>
> john
>
>
>
> --
> nielsenjf at my-Deja.com
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From jimc at regnoc.com  Sun Jan 30 12:24:15 2000
From: jimc at regnoc.com (Jim Conger)
Date: Sun, 30 Jan 2000 17:24:15 GMT
Subject: Python MySQL.connect() question
References: <3893CD31.B17E8B84@regnoc.com> <870tjf$514$1@news1.xs4all.nl>
Message-ID: <3894737C.A676BC16@regnoc.com>

Boudewijn,

Thanks you for the quick response.  I'm still missing something.  Here is my
attempt to follow you example.  First I check that mySQL is running using
mysqladmin, and then attempt to connect in python:

$ mysqladmin -ppassword processlist

| Id | User | Host     | db | Command  | Time | State | Info
| 12 | jimc | localhost |    | Processes | 0 |    |     |

$ python
Python 1.5.1...........
>>> import MySQL;
>>> self.dbConnection = MySQL.connect("localhost","jimc","password")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
MySQL.error: Can't connect to local MySQL server
>>>

Does mySQL need to be already connected to a DB?  If so how?  I'm sure it is
obvious, but I can't figure it out!

Regards,
Jim Conger
www.regnoc.com


Boudewijn Rempt wrote:

> Jim Conger <jimc at regnoc.com> wrote:
> > I have mySQL running on a RH Linux 6.1 box, and have added the
> > MySQLmodule.so to my python library.  I can't figure out how to connect
> > within python.  My attempts look like:
>
> >>>> import MySQL;
> >>>> db = MySQL.connect();
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > MySQL.error: Can't connect to local MySQL server
> >>>>
>
> I'm not really surprised - I think you need to supply
> some parameters:
>
>     hostname="essle"
>                 username="boud"
>                 password=""
>                 databasename="languages"
>     self.dbConnection = MySQL.connect(hostname,username,password)
>           self.dbConnection.selectdb(databasename)
>
> Which is what I did when I used MySQLmodule; however, I've since
> began using the DB-API II compliant MySQLdb, but I miss some of
> the conveniences of MySQLmodule.
>
> --
>
> Boudewijn Rempt  | http://www.valdyas.org

--
Jim Conger
Project Manager
Silicon Valley Oil Co.
jimc at regnoc.com, 925-842-6729





From samschul at pacbell.net  Sun Jan  9 16:38:14 2000
From: samschul at pacbell.net (sam schulenburg)
Date: Sun, 9 Jan 2000 13:38:14 -0800
Subject: How to do an event triggered callback function?
Message-ID: <Ma7e4.33$DE3.2811@nnrp1-w.snfc21.pbi.net>

I'am in need of an event trigger mechanism to control seperate process under
cPython. I have found a sample mechaninism under Tkinter using
self.after(interval,function).
I would like to accomplish the same thing under python in console mode so I
and return to the command line interpretor and issue commands while various
background process take place at predevined intervals. Could anyone point me
towards a simple example that may help me. I have searched the faqs on
threads and process withou much help.

Sam Schulenburg




From hartmut at oberon.noris.de  Wed Jan  5 17:24:00 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Wed, 05 Jan 2000 23:24:00 +0100
Subject: NEWS + threads
References: <0t387365ddi452en79%milek@korweta.task.gda.pl>
Message-ID: <16600393@oberon.noris.de>

Robert Milkowski (milek at task.gda.pl) schrieb:

 > Does somebody make a small news reader in python?
 >Or only threads like tin (based on subject and references)?
 >Or meybe small cook-book how to make threads... step by step...

Some friend of mine and me are currently prototyping a Mail/News Database
System. Since the DB will support references handling, the mail/newsreader
will be able to to this, too.

For more information about the porject please visit ums.home.pages.de which
describes the predecesor (implemented on AmigaOS).

+++hartmut


From sp00fD at yahoo.com  Fri Jan  7 12:30:58 2000
From: sp00fD at yahoo.com (sp00fD)
Date: Fri, 07 Jan 2000 17:30:58 GMT
Subject: CGI parameters
Message-ID: <8557si$10l$1@nnrp1.deja.com>

How do you query CGI parameters from radio buttons, urls, etc in
python?  In perl it would look something like

$query = new CGI;
$foo = $query->param("foo");


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


From jfzhou at 188.net  Fri Jan 14 03:10:57 2000
From: jfzhou at 188.net (Öܹ«)
Date: Fri, 14 Jan 2000 16:10:57 +0800
Subject: cgi use root uid
Message-ID: <85mlpc$2uac$1@news.cz.js.cn>

Can someone tell me how to run a python cgi that can have root right.

the main perpose is to modify some file like passwd,hosts

Luckey at soim.net
2000.1.14





From pj at sgi.com  Mon Jan  3 15:43:55 2000
From: pj at sgi.com (Paul Jackson)
Date: 3 Jan 2000 20:43:55 GMT
Subject: Finding where Python is on a Unix server through Telnet?
References: <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com>
Message-ID: <84r1mb$kr69k@fido.engr.sgi.com>

A portable equivalent to the locate/which/whence/...
confusion (one that I've typed so often I think of
it as a single command) is:

    % sh	    # get a Bourne/Korn/Bash like sub-shell
    $ IFS=': '	    # add colon ':' to Input Field Separators
    $ for i in $PATH
    > do
    >   test -f $i/python && ls -ltd $i/python
    > done
    -rwxr-xr-x  2 root  user  3731188 Jun 11  1999 /usr/local/bin/python
    $ exit	    # leave subshell (with its funky IFS setting)

You don' have to key in the comments each time ;).

And you can skip the "test -f $i/... &&" phrase,
if you'd rather see the ls errors for non-existant
files than type the extra stuff needed to filter them.

And this fails to pick up "aliases" -- for that I think
shell-specific syntax is required.


-- 

=======================================================================
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 pierre at saiph.com  Wed Jan 19 13:49:22 2000
From: pierre at saiph.com (Pierre Imbaud)
Date: Wed, 19 Jan 2000 19:49:22 +0100
Subject: an xemacs func to close parentheses
Message-ID: <38860732.5B9CFDB2@saiph.com>

New posting, I had slight mistakes in the first one:
I was bored by the number of parentheses (lets rather say
surrounding chars: () [] {}) to close in python. could not the editor
guess the next action? these happen to be funny characters, uneasy
to type...
After looking for an existing xemacs function, with no success, I
designed one. It its specific to python, needs python-mode to be active.
make it called by your .emacs, map it to whatever key seq U like...
What the function does: insert the closing char corresponding to the
unmatched opening char: after "(['toto" will insert "'", then (if called
again) "]", then ")"
On first checks I saw no bug, there cud be some left. Tell me
close-parentheses.el:
;;Pierre Imbaud,  11 Jan 00 17:42 pierre at saiph.com
;;to build the ability to complete parentheses, with one function,
;;whatever the parenthesis is. algo: look backward for an open
;;parenthesis. if we find a closing one, skip backward the set. When
;;found, insert the corresponding closing parenthesis.

;;bugs: I hoped to use syntax-table, I could not find a way to retrieve
;; closing parenthesis, from syntax table: so data used here is langage
;; dependent: we redefine the mapping chars. But we still use
;; syntax-table, thru the function backward-sexp: this provides an easy
;; way to jump over existing parenthesis-enclosed text. as is, the
;; package is dedicated to python langage, with one flaw: it wont behave
;; if it has to close a '''string like this one''', """or this one""".
;; Pain in the ass, but it shud help in most cases. Surprisingly, it
;; manages ok to jump over these strings.

(require 'python-mode)

;;++++++++++++++++
; key binding; modify as wished. I wanted something handy, with one
keystroke
; ] has these qualities, and is a good reminder of the function too. In
the
; seldom cases the function wont work properly, self insert is possible
; thru "ctrl-q ]"
;;++++++++++++++++

(define-key py-mode-map "]" 'close-parenthese)


;;++++++++++++++++
; customizable variables.
;;++++++++++++++++
; the parenthesis couples list:
(defvar clos-par-alist (mapcar (lambda (in) "returns an alist, from a
list of pairs"
                                 (cons (car in) (cadr in)))
                               (mapcar 's2l (list "()" "[]" "{}"))))

(defvar clos-max-back 366)              ;limit for back exploration


;;++++++++++++++++
; induced constants:
(setq closings (mapcar 'cdr clos-par-alist))
(setq openings (mapcar 'car clos-par-alist))

;;++++++++++++++++
; utilities
;;++++++++++++++++
; I guess many of these eiter exist, I didnt find them, or cud have been
; defined as lambda.

(defun s2l (string)
  "string to list. recursive, inefficient"
  (if (equal string "")
      nil
    (cons (elt string 0) (s2l (substring string 1)))
    ))

(defun py-in-string ()
  " within a string, returns closing string char (as a lisp string!).
elsewhere nil"
  (let ((close (nth 3 (py-parse-state))))
    (if close (char-to-string close)
           )))

(defun closing-before-point ()
  "t if char before point is a closing parenthese"
  (memq (preceding-char) closings)
  )

(defun opening-before-point ()
  (interactive)
  "t if char before point is an opening parenthese"
  (if (memq (preceding-char) openings) (message "y"))
  )

(defun closing-paren(char)
  "returns the corresponding closing char to an opening char"
  (cdr (assoc char clos-par-alist)))
(char-to-string (closing-paren ?\[))

;;++++++++++++++++
; main function:
;;++++++++++++++++

(defun close-parenthese ()
  (interactive)
  (let ((back clos-max-back)
        (to-insert "")
        (close-string (py-in-string))
        )
    (save-excursion
      (while (< 0 back)
        (setq back (- back 1))
        (if close-string
            (progn
              (setq to-insert close-string)
              (setq back 0)
              )
          ;else
          (if (memq (preceding-char) openings)
            (progn
              (setq to-insert (char-to-string
                               (closing-paren (preceding-char))))
              (setq back 0)
              )
            ;else again
            (if (memq (preceding-char) closings)
                (backward-sexp)
              (backward-char)
              )
            ))))
    ;out of save-excursion, but inside let...
    (insert-string to-insert)
    ))
          

-- 
Pierre Imbaud <pierre at saiph.com>
12 Rue des Bosquets 91480 Quincy Sous S?nart France
Tel:  01 69 00 94 57 Fax 79 65


From tony.mcdonald at ncl.ac.uk  Sun Jan 23 12:39:47 2000
From: tony.mcdonald at ncl.ac.uk (Tony McDonald)
Date: Sun, 23 Jan 2000 17:39:47 +0000
Subject: Need XML-RPC help
Message-ID: <B4B0ED62.316E%tony.mcdonald@ncl.ac.uk>

Hi all,
I want to start using XML-RPC as a way to integrate what I'm doing in Zope
with others in the department who use Perl. Problem is, I can't get XML-RPC
to work through our proxy.

My problem seems to be that xmlrpclib code is using urllib and httplib
libraries, but whilst urllib handles proxies beautifully, httplib doesn't
unless 'set up' from urllib calls, and the code in xmlrpclib uses httplib
directly. Specifically in the 'request' method of the 'Transport' class.

I do have the environment variable 'http_proxy' set up correctly as lynx,
wget and the urrlib test suite work fine.

I'd appreciate any pointers on this, as I'm starting to think that the only
thing I can do is to incorporate the proxy code from urllib into httplib -
something I'm loathe to do!

many thanks for any pointers
Tone



From super_banane at yahoo.com  Fri Jan 21 11:35:14 2000
From: super_banane at yahoo.com (super_banane at yahoo.com)
Date: Fri, 21 Jan 2000 16:35:14 GMT
Subject: ConfigParser and .ini files
Message-ID: <86a1s2$7ha$1@nnrp1.deja.com>

Hello:

The ConfigParser class in the default Python libraries
does not support comments given on the value lines, separated
by a semicolon, such as:

[Section]
keyword = value ; put your comments here

The value as it is returned is:
"value ; put your comments here"
... which is not really what you expect.

Adding one line of code in the ConfigParser class
adds support for this feature. Any chance of submitting
this for the next Python release?

file ConfigParser.py line 311, add:
optval = string.split(optval)[0]
(before optname=string.strip(optval))

Cheers
--
Nicolas


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


From DaPatch at wanadoo.nl  Wed Jan 26 08:25:31 2000
From: DaPatch at wanadoo.nl (DaPatch)
Date: Wed, 26 Jan 2000 14:25:31 +0100
Subject: Making a subwindow from a parent
Message-ID: <86mso4$21oh$1@buty.wanadoo.nl>

Hi,

How do i hang a subwindow under root=tk()

I don't want it to appear right away, but only when i ask it to.
I also don't want it to return control as soon as its closed, but when the
focus gets away.


please help. And if you have a good tkinter resource with a good explanation
of the functions (not the lifepreserver please) please let me know.

thanks,

DaPatch




From wolf at one.net  Sat Jan 29 10:02:16 2000
From: wolf at one.net (wolf at one.net)
Date: Sat, 29 Jan 2000 10:02:16 -0500
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <949076889.194238@news.adtranzsig.de> <20000128180823.A1261@stopcontact.palga.uucp> <01bf69c5$fe225580$LocalHost@dreadzone>
Message-ID: <82069scs1o40srpn65508bkdfphv7702ff@4ax.com>

Hi Dirk,

"Dirk Heise" <dheise at debitel.net> wrote:

>But as an option of the parser, it would
>make sense, so the build manager could reject all
>code the developer didn't test with that option 
>switched on. Of course, it's only important for
>big team efforts.

It's important for *ANY PRODUCTION CODE*. Trust me. :) I'm currently
in the process of transitioning from one job to another, and I'm
documenting (processes, not code, but the same rule applies). You'd be
*amazed* at the amount of code and procedure a single programmer can
produce in 5 years... :)


Respectfully,

Wolf

"The world is my home, it's just that some rooms are draftier than
others". -- Wolf


From a.eyre at optichrome.com  Tue Jan 25 06:44:54 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 25 Jan 2000 11:44:54 -0000
Subject: Re-assigning 'stdin' Problem...
In-Reply-To: <cW%i4.3$R04.297@news.hananet.net>
Message-ID: <000401bf6729$98e41df0$3acbd9c2@peridot.optichrome.com>

> class MyConsole:
>     def readline( self ):
>         if ( self.BufferHasData )
>             return self.GetBufferData()
>         else
> ->        return ""

Have a look in the IDLE source:

python/Tools/idle/PyShell.py line 266 (ish)

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From sposhua at my.pc  Wed Jan 19 06:30:18 2000
From: sposhua at my.pc (Sposhua)
Date: Wed, 19 Jan 2000 11:30:18 +0000
Subject: Python Rocks! - get rid of colons
In-Reply-To: <861jm5$nrj$1@mach.vub.ac.be>
References: <860ftk$bas@news.or.intel.com> <8610or$nsu$1@news.udel.edu> <861jm5$nrj$1@mach.vub.ac.be>
Message-ID: <Pine.WNT.4.21.0001191125160.243-100000@wks_3.e-shopwork.com>

On 18 Jan 2000, Thomas Hamelryck wrote:

> Not necessarily. I've been using python for about a year or so and I still
> dislike indentation. Indentation can make your code less readable and leads to
> mysterious errors. I'm sure the strict use of indentation keeps a lot of potential users
> away from python. However, I feel that the other features of the language 
> outweigh this disadvantage.

I'm one of the "prefers indentation to {}" ppl, and I'd just like to add: whe
the heck do we need colons (:) at the end of if's 'n loops anyway? The only loss
I see is to these 1-liners:

if a==b: eric=idle

What's wrong with:

if a==b
	eric=idle

(oh, and wouln't it make life easier to change == to = and let python work out
when it is assingment and when it is comparison?)



From pj at sgi.com  Mon Jan  3 15:34:19 2000
From: pj at sgi.com (Paul Jackson)
Date: 3 Jan 2000 20:34:19 GMT
Subject: RPM-interface/module
References: <386B3CDB.FF843280@bibsyst.no> <16600370@oberon.noris.de>
Message-ID: <84r14b$l1rms@fido.engr.sgi.com>

|> I have'nt found any source for anaconda :-(

I see it on Red Hat 6.1 CD #1, under /misc/src/anaconda
(and related stuff under /misc/src/trees).

For example, the C lang rpm cracking code, for linking
as a Python module, is in /misc/src/anaconda/rpmmodule.

The Japanese version of RH 6.1 seems to be about a
month newer than the version targeted for English.

-- 

=======================================================================
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 tseaver at starbase.neosoft.com  Fri Jan 21 22:07:03 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 21 Jan 2000 21:07:03 -0600
Subject: Minor inconsistency between cStringIO and (StringIO or normal fil
	es)...
References: <CC59868A0A76D311B1F50090277B321C1F3092@VRTMAIL>
Message-ID: <7AFD913D2AB392ED.7FACF630E6008D90.69EE42984FB943F7@lp.airnews.net>

In article <CC59868A0A76D311B1F50090277B321C1F3092 at VRTMAIL>,
Mike Fletcher  <mfletch at tpresence.com> wrote:
>Following demonstrates the inconsistency, essentially, a seek to maxint
>raises a memory error on a cStringIO object, whereas the StringIO and file
>objects just move to the end of the file for further actions (which is very
>handy).  Is there a better way to spell "move to end of file" (keeping in
>mind that this is for a class doing effectively random access in the
>buffer)?
>
>Cheers,
>Mike
>
>PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>Portions Copyright 1994-1999 Mark Hammond (MHammond at simporkippinet.com.au)
>>>> import cStringIO, sys
>>>> d = cStringIO.StringIO()
>>>> d.seek(sys.maxint)
>Traceback (innermost last):
>  File "<interactive input>", line 1, in ?
>MemoryError: 
>>>> d = open( 'z:\\temp\\test.wrl','r')
>>>> d.seek(sys.maxint)


d.seek( 0, 2 ) # optional second argument gives POV
               # 0 = from start (default)
               # 1 = from current
               # 2 = from end, backward

Best,

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From gerrit.holl at pobox.com  Fri Jan 21 08:06:31 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 14:06:31 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <slrn88e4pl.67b.grant@grante.comtrol.com>; from grant@nowhere.pobox.com on Thu, Jan 20, 2000 at 01:52:53PM +0000
References: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse> <38862948.5AEE964C@roguewave.com> <slrn88e4pl.67b.grant@grante.comtrol.com>
Message-ID: <20000121140631.A2701@stopcontact.palga.uucp>

Grant Edwards wrote on 948372773:
> In article <38862948.5AEE964C at roguewave.com>, Bjorn Pettersen wrote:
> >Evguenii Smogailov wrote:
> >
> >> On 19 Jan 2000, Stephen Coursen wrote:
> >> > When did this become a trait of Perl?  Gods, what's so hard about knowing
> >> > that == is the equality operator?
> >>
> >> Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.
> >
> >I see... so you don't really want eq, gt, and ne; you want equal, greater, and
> >not equal.
> 
> No, no, that would be too hard to read.  We really need the
> operators 'is_greater_than' 'is_less_than' 'is_equal_to'.
> 
> Ideally, the underscores should be spaces, but that might be
> problematic.  

But why only in English?

if a is_greater_than b

would be equivalent to

als a is_groter_dan b

and

als a ist_grosser_dan b

So we'll have over 400 keywords! Yeah!

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From duncan at rcp.co.uk  Fri Jan 21 08:09:30 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 21 Jan 2000 13:09:30 GMT
Subject: WAP with Python
References: <20000121120558.50068.qmail@hotmail.com>
Message-ID: <8EC28BE76duncanrcpcouk@194.238.50.13>

ulf.engstrom at b2b-link.com (Ulf Engstr?m) wrote in
<20000121120558.50068.qmail at hotmail.com>: 

>Hello..:)
>
>Has anyone tried to do any WAP-things with Python? I'm trying to set up
>an apache webserver on a windows2000-machine and get it to run with the 
>wap-software. This havn't been all that successful this far, but I'm
>working on it. The python-part of it seems to be working, since I can
>see it with my regular browser.
>Anyone has any experience in this area and can give me some hints?
>Brgrds
>Ulf

I have done some WAP stuff using Zope, see
http://www.rcp.co.uk/distributed/wapdemo for an example, or
http://wap.autolocate.co.uk for the actual WAP site. I am currently
writing a Zope howto on this subject and it should be visible on
www.zope.org soon. 

If you have any specific questions I will gladly try to answer them. I
suggest you use the Nokia WAP toolkit for testing as that gives pretty
good error messages, and also lets you see the source if you failed to
send valid WML. Don't forget to set the mime type. 


From mhammond at skippinet.com.au  Wed Jan  5 22:07:50 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 06 Jan 2000 03:07:50 GMT
Subject: Portable Authentication? WinNT and UNIX
References: <84vqlj$2e8$1@nnrp1.deja.com>
Message-ID: <aGTc4.11597$lv5.51250@news-server.bigpond.net.au>

<charliehubbard76 at yahoo.com> wrote in message
news:84vqlj$2e8$1 at nnrp1.deja.com...

> Does anyone know how I can authenticate WinNT users thru Win32 api?

You can use win32security.LogonUser() to check this.  See the docs at
microsoft.com for that function.

Mark.





From sameerc1 at yahoo.com  Wed Jan 19 02:53:52 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Tue, 18 Jan 2000 23:53:52 -0800
Subject: python and ASP
Message-ID: <00bd01bf6252$54e50a00$d548fea9@desktop>

yes, thanks, that seems to work.  Is there any book or place of reference that you recommend I look at when I am having problems using Python with ASP.  Currently, I am experimenting with ASP, and my primary reference is ASP in a NutShell, which uses VBScript entirely in all ASP examples.  I haven't seen any ASP book that deal with Python specific syntax or issues.


-----Original Message-----
From: John Nielsen <nielsenjf at my-deja.com>
Newsgroups: comp.lang.python
To: python-list at python.org <python-list at python.org>
Date: Tuesday, January 18, 2000 4:01 PM
Subject: Re: python and ASP


>Try something like:
>
>Session.SetValue('userID', userid)
>
>john
>
>
>
>In article <002501bf60ae$1773bc20$b674fea9 at desktop>,
>  "Sameer Chowdhury" <sameerc1 at yahoo.com> wrote:
>> How do Sessions work when using python with ASP.   I, for example,
>have
>> working Python based ASP code, I add the following line:
>Session("userID") =
>> userID
>> after defining userID.  After I add this line, I get the following
>error
>> when I test it.
>> Python ActiveX Scripting Engine error '80020009'
>>
>> Unknown
>>
>> ?
>>
>> Unknown
>>
>> Can anyone suggest a remedy.  Thanks in advance.
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000118/84190bd9/attachment.html>

From pierre at saiph.com  Thu Jan 13 20:41:55 2000
From: pierre at saiph.com (Pierre Imbaud)
Date: Fri, 14 Jan 2000 02:41:55 +0100
Subject: an xemacs func to close parentheses
Message-ID: <387E7EE3.600F4B74@saiph.com>

I was bored by the number of parentheses (lets rather say
surrounding chars: () [] {}) to close in python. could not the editor
guess the next action? these happen to be funny characters, uneasy
to type...
After looking for an existing xemacs function, with no success, I
designed one. It its specific to python, needs python-mode to be active.
make it called by your .emacs, map it to whatever key seq U like...
What the function does: insert the closing char corresponding to the
unmatched opening char: after (['toto will insert ', then (if called
again) ], then )
On first checks I saw no bug, there cud be some left. Tell me
close-parentheses.el:
;;Pierre Imbaud,  11 Jan 00 17:42 pierre at saiph.com
;;to build the ability to complete parentheses, with one function,
whatever the
;;parenthesis is.
;;algo: look backward for an open parenthesis. if we find a closing one,
;; skip backward the set. When found, insert the corresponding closing
parenthesis.

;; bugs: I hoped to use syntax-table, I could not find a way to retrieve
closing
;; parenthesis, from syntax table: so data used here is langage
dependent: we redefine
;; the mapping chars. But we still use syntax-table, thru the function
backward-sexp:
;; this provides an easy way to jump over existing parenthesis-enclosed
text.
;; as is, the package is dedicated to python langage, with one flaw: it
wont behave
;; if it has to close  a '''string like this one''', """or this one""".
Pain in the
;; ass, but it shud help in most cases. Surprisingly, it manages ok to
jump over these
;; strings.

(require 'python-mode)

;;++++++++++++++++
; customizable variables.
;;++++++++++++++++
; the parenthesis couples list:
(defvar clos-par-alist (mapcar 'l2alist (mapcar 's2l (list "()" "[]"
"{}"))))

(defvar clos-max-back 366)              ;limit for back exploration


;;++++++++++++++++
; induced constants:
(setq closings (mapcar 'cdr par-alist))
(setq openings (mapcar 'car par-alist))

;;++++++++++++++++
; utilities
;;++++++++++++++++
; I guess many of these eiter exist, I didnt find them, or cud have been
; defined as lambda.

(defun s2l (string)
  "string to list. recursive, inefficient"
  (if (equal string "")
      nil
    (cons (elt string 0) (s2l (substring string 1)))
    ))

(defun l2alist (in)
  "returns an alist, from a list of pairs"
  (cons (car in) (cadr in)))

(defun py-in-string ()
  " within a string, returns closing string char (as a lisp string!).
elsewhere nil"
  (let ((close (nth 3 (py-parse-state))))
    (if close (char-to-string close)
           )))

(defun closing-before-point ()
  "t if char before point is a closing parenthese"
  (memq (preceding-char) closings)
  )

(defun opening-before-point ()
  (interactive)
  "t if char before point is an opening parenthese"
  (if (memq (preceding-char) openings) (message "y"))
  )

(defun closing-paren(char)
  "returns the corresponding closing char to an opening char"
  (cdr (assoc char par-alist)))
(char-to-string (closing-paren ?\[))

;;++++++++++++++++
; main function:
;;++++++++++++++++

(defun close-parenthese ()
  (interactive)
  (let ((back clos-max-back)
        (to-insert "")
        (close-string (py-in-string))
        )
    (save-excursion
      (while (< 0 back)
        (setq back (- back 1))
        (if close-string
            (progn
              (setq to-insert close-string)
              (setq back 0)
              )
          ;else
          (if (memq (preceding-char) openings)
            (progn
              (setq to-insert (char-to-string
                               (closing-paren (preceding-char))))
              (setq back 0)
              )
            ;else again
            (if (memq (preceding-char) closings)
                (backward-sexp)
              (backward-char)
              )
            ))))
    ;out of save-excursion, but inside let...
    (insert-string to-insert)
    ))
          

    


-- 
Pierre Imbaud <pierre at saiph.com>
12 Rue des Bosquets 91480 Quincy Sous S?nart France
Tel:  01 69 00 94 57 Fax 79 65


From drek at MonsterByMistake.Com  Tue Jan 18 21:34:11 2000
From: drek at MonsterByMistake.Com (Agent Drek)
Date: Tue, 18 Jan 2000 21:34:11 -0500 (EST)
Subject: Python Rocks!
In-Reply-To: <8631qm$26p@news.or.intel.com>
Message-ID: <Pine.BSF.4.10.10001182126290.67136-100000@bunny.monsterbymistake.com>

|
|How hard can it be to add 'end' indicators to the language.. If these end
|indicators are added, the language would still be backward compatible. I
|can't believe nobody has fixed this for so long.

you can get the source via cvs :)

|
|-tye4
|
|
|-- 

I'm definately new to python but the 'look and feel' are what brought me
to this language. Why should I have to type 'endif' when the block is
visible to the human eye. I would rather use braces thank-you. Python
makes all the annoying things about where to put braces go away.
end 'indicators' would lead to uglification (yes it's a word)

It takes 15 minutes to go through the
vim tutorial and then blammo! you have a decent editor that is able
to deal with the formatting. <insert editor flame war here>

back to lurk mode.

    =derek

Monster By Mistake Inc > 'digital plumber'
http://www.interlog.com/~drek




From skip at mojam.com  Tue Jan 18 11:31:08 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 18 Jan 2000 10:31:08 -0600 (CST)
Subject: DBI spec and parameter passing
In-Reply-To: <388455BB.154F36EC@lemburg.com>
References: <3883A0CA.AB0D4854@exceptionalminds.com>
	<14467.42391.905771.673058@beluga.mojam.com>
	<1263967432-11066499@hypernet.com>
	<20000118015809.F21652@xs4all.nl>
	<388455BB.154F36EC@lemburg.com>
Message-ID: <14468.38220.788564.314607@beluga.mojam.com>

    Marc> Note that one should always use binding parameters if possible:
    Marc> the driver or interface usually knows better how to encode the
    Marc> data according to the DBs specs and it's also much easier to write
    Marc> and more portable.

Marc,

In my local case (MySQLdb), I use "%s" and your indication is that ODBC uses
"?", which is fine.  Other database modules apparently use named
interpolation (e.g., "%(city)s").  Is there a database-independent way of
building the basic strings other than switching on the value of
db.paramstyle:

    db = somethingDBIcompliant(...)
    c = db.cursor()
    if db.paramstyle == "qmark":
        c.execute("sql statement with ? embedded", params)
    elif db.paramstyle == "numeric":
        c.execute("sql statement with :1 embedded", params)
    elif db.paramstyle == "format":
        c.execute("sql statement with %s embedded", params)
    ...

or is that much portability worth the effort?  My code stops before this
point in its attempts to be independent of a particular database module.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098




From alex at magenta.com  Sun Jan  2 17:26:59 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 2 Jan 2000 23:26:59 +0100
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl> <386E7BD4.514764EA@ndh.net>
Message-ID: <02ff01bf5571$0b808260$c02b2bc1@martelli>

Stefan Schwarzer writes:
    [snip]
> The idea to provide "true" multidimensional arrays sounds also very
> good to me. Treating multidimensional arrays as lists of lists (or
> vice versa) complicates the use of them, and may it be that a
> special wrapper module is needed to provide the corresponding

Isn't that part of what the Numerical extensions to Python developed
at LLNL provide, already?  Excellent multidimensional arrays with
all sort of extended slicing, etc?

Maybe that word "Numerical" is turning people off -- but, the package's
"multi-arrays" can actually contain (references to) any sort of Python
objects whatsoever, so there is no limitation to "numerical" stuff...


Alex





From bjorn at roguewave.com  Mon Jan 17 12:53:22 2000
From: bjorn at roguewave.com (bjorn)
Date: Mon, 17 Jan 2000 10:53:22 -0700
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com>
Message-ID: <38835712.E787B04B@roguewave.com>

It might be easier to use the exchange server as either an IMAP or POP server
instead (both should be on by default).

-- bjorn

Grant Edwards wrote:

> I just found the http server page on Exchange server that
> allows access to e-mails via a web browser.  What I'd really
> like is to write a Python program that just sucks the mail out
> of the inbox and delivers it locally.
>
> I've never done any Python code that accesses an http server.
> Are there any applications which would be good examples?
>
> --
> Grant Edwards                   grante             Yow!  Oh, I get it!! "The
>                                   at               BEACH goes on," huh,
>                                visi.com            SONNY??
> --
> http://www.python.org/mailman/listinfo/python-list




From gmcm at hypernet.com  Wed Jan  5 09:19:25 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 5 Jan 2000 09:19:25 -0500
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
In-Reply-To: <uBCWdBAkv0c4Ewdy@jessikat.demon.co.uk>
References: <20000105082703.A9441@vislab.epa.gov>
Message-ID: <1265040530-3892638@hypernet.com>

Robin Becker wrote:
 > > |Also under
> certain extreme circumstances my code would like to report a >
> |total failure eg attempt to divide by error. I would like to ask
> more > |experienced pythoneers how arithmetic errors are handled
> using built in > |IEEE exceptions and how does one gracefully
> crash a python script? (eg > |an abend/abort). 

> >raise
> ArithmeticError, "Oops" >raise ZeroDivisionError, "Ouch" > >? >

> yes that's python, but what's the best way to do it in an
> existing C library? -- Robin Becker

PyErr_SetString(PyExc_ZeroDivisionError, "Ouch");
return NULL;

(see bltinmodule.c for the complete list).
 
> -- 
> http://www.python.org/mailman/listinfo/python-list



- Gordon



From gerrit.holl at pobox.com  Tue Jan 11 04:54:48 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 11 Jan 2000 10:54:48 +0100
Subject: Newby question: how manage user input??
In-Reply-To: <85efp1$6pi$1@pegasus.tiscalinet.it>; from composzioni@tiscalinet.it on Tue, Jan 11, 2000 at 06:32:31AM +0100
References: <85efp1$6pi$1@pegasus.tiscalinet.it>
Message-ID: <20000111105448.A2093@stopcontact.palga.uucp>

Giovanni Riccardi wrote on 947568751:
> I want to make a simple script that print out
> some text (actually questions), get what the user
> wrote at the displyed prompt and make some calculation
> with that.
> I tried to implement the user input with stdin or raw_input but after
> returning the value the function I wrote doesn't
> get back to the main loop??

Can you please give us your code, what you have now?

> Some refs (with examples) on the net for this problems?
> Or could someone explain it to me??

Did you already read www.python.org/doc/current/tut?

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----
If you think Linux should get more hardware drivers, _please_ sign this:
http://www.libranet.com/petition.html



From DOUGS at oceanic.com  Sun Jan 30 04:21:20 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Sat, 29 Jan 2000 23:21:20 -1000
Subject: [Newbie Problems] AttributeError from CGI script?
Message-ID: <5650A1190E4FD111BC7E0000F8034D26A0F2A1@huina.oceanic.com>

Its just a guess, but make sure you haven't made an assignment to string
someplace in your code.  In other words, I'm guessing you've done the
equivalent of:

[dougs at lawehana dougs]$ python
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 string
>>> line = "the placeholder"
>>> string = "gum up"
>>> line = string.replace(line,"placeholder","works")
Traceback (innermost last):
  File "<stdin>", line 1, in ?
AttributeError: 'string' object has no attribute 'replace'

I'm guessing because my error message is somewhat different.  It sometimes
helps to know what platform and version you're using, and usually adding the
least amount of code that duplicates the error helps also.

Hope this helps.

-Doug-

> -----Original Message-----
> From: steve_w at localhost.localdomain
> [mailto:steve_w at localhost.localdomain]
> Sent: Saturday, January 29, 2000 7:15 PM
> To: python-list at python.org
> Subject: [Newbie Problems] AttributeError from CGI script?
> 
> 
> I'm trying to use Python in a CGI script for the first time and 
> have run into a problem I don't understand.  I get the subject
> error message when trying to do a string.replace using one of
> the values returned by the form that trigers the script.
> 
> The line of code looks like:
> 
> line = string.replace(line,"placeholder",form["target"].value)
> 
> I get an error message back saying:
> 
> Traceback (innermost last):
> File "[my script's file name]"
>    line = string.replace(line,"placeholder",form["target"].value)
> AttributeError: replace
> 
> Could someone help me figure out what this error message means?
> 
> Thanks in advance,
> 
> Steve Wall
> steve_w at visi.com
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From thantos at chancel.org  Sun Jan  2 07:32:35 2000
From: thantos at chancel.org (Alexander Williams)
Date: Sun, 02 Jan 2000 12:32:35 GMT
Subject: Finding where Python is on a Unix server through Telnet?
References: <386EA0EF.F7A72441@jslove.net> <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com>  <84ncfb$e5q$0@216.39.151.169>
Message-ID: <slrn86uhb3.b66.thantos@chancel.org>

On 2 Jan 2000 11:23:23 GMT, Donn Cave <donn at oz.net> wrote:
>"locate" is a new one on me, not available on Digital UNIX.

On Tru64 (*grumble grumble*) try 'whereis python'.  

			Alexander Williams
			Tru64 Unix Y2K Support Veteran
			(Man, was it quiet.)

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From sposhua at my.pc  Wed Jan  5 04:56:35 2000
From: sposhua at my.pc (Sposhua)
Date: Wed, 5 Jan 2000 09:56:35 +0000
Subject: Importing a dicrionary from a file
In-Reply-To: <3872C4EC.20ADE046@trojanslair.zzn.com>
References: <3872C4EC.20ADE046@trojanslair.zzn.com>
Message-ID: <Pine.WNT.4.21.0001050956170.104-100000@wks_3.e-shopwork.com>

On Tue, 4 Jan 2000, Black Mantis wrote:

> hello
> I am a newbie, and i am working on a program that uses a really big
> dictionary. i was wondering if i could store the dictionary in a
> seperate file, let's say dict.dat. Now, at the beggining of the program
> i want to import that dictionary and get it ready, how would i do this?
> is there a way to read the whole input file into one variable?

Use the shelve module.



From Christian.Schaller at mchp.siemens.de  Wed Jan 12 06:42:55 2000
From: Christian.Schaller at mchp.siemens.de (Christian Schaller)
Date: Wed, 12 Jan 2000 12:42:55 +0100
Subject: Q: Network access & Python
Message-ID: <387C68BF.EA30D859@mchp.siemens.de>

Hello...

  Does anyone know if there's a module in Python for getting network
stats?  I want to write a tool for displaying all used ports and
services running on them and also if there's a connection on a port,
from where (ip:port) it originates, but found no function/module
containing functions for listing/analyzing ports.

  Thanks a lot in advance!!!

bye
  Chris...


From news at dorb.com  Sun Jan 16 15:33:37 2000
From: news at dorb.com (Darrell)
Date: Sun, 16 Jan 2000 15:33:37 -0500
Subject: SERIAL I/O ON WIN32
References: <Qvpg4.581$253.12322@news>
Message-ID: <NVpg4.5523$U5.179277@typhoon.nyroc.rr.com>

Haven't messed with the serial-i/o module but did this little experiment to
dial my modem using COM. Mscomm32 came from me running makepy.py on
Mscomm32, I don't recall the exact syntax. The hard part was figuring out
how to use it from digging though Microsoft web sites and books.

import Mscomm32, os

c=Mscomm32.MSComm()
print dir(c)
c.CommPort = 1
c.Settings = "115200,N,8,1"
c.PortOpen = 1
c.InputLen=0
c.Handshaking=2
c.RTSEnable=1
print 'Event:', c.CommEvent
c.Output="ATDT555-1234" + chr(13)+'\n\n'
print 'OutBufferCount:',c.OutBufferCount
print 'Event:', c.CommEvent
while 1:
    buf=raw_input('')+'\n'
    c.Output=buf
    if c.InBufferCount > 0:
        print 'INPUT:',c.Input
    print 'Event:', c.CommEvent

c.PortOpen = 0
raw_input('Wait')

--
--Darrell
"bob" <bob at no.com> wrote in message news:Qvpg4.581$253.12322 at news...
> Hi there...newbie to Python here.
>
> I need to pass information across a COM port. I have the serial-i/o module
> but no clue on how to use it.
>
> Thanks for any and all help!
>
> Bob
>
>
> --
> http://www.python.org/mailman/listinfo/python-list




From thomas.heller at ion-tof.com  Thu Jan  6 10:21:24 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 6 Jan 2000 16:21:24 +0100
Subject: Correct URL (was: Re: py2bat (was: How do I make a Python .bat executable file?))
References: <38722971.4B036062@mchp.siemens.de> <uembv316c.fsf@cs.uu.nl> <004e01bf5857$cac3eca0$4500a8c0@thomasnotebook>
Message-ID: <009501bf5859$b20ddd90$4500a8c0@thomasnotebook>

The correct URL for Gordon's installer is:

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

Sorry for the confusion

Thomas Heller





From rgruet at ina.fr  Mon Jan 24 15:33:39 2000
From: rgruet at ina.fr (Richard Gruet)
Date: Mon, 24 Jan 2000 21:33:39 +0100
Subject: Global mutex/lock
Message-ID: <388CB723.7EE04D7C@ina.fr>

Hi Python gurus,

Is there a portable global mutex implementation somewhere ?
AFAIK, the locks in the threading/thread modules work only between
threads of the same module.
OTOH, there are file locks for Windows and Unix (respectively in modules
msvcrt and fcntl), that can serve as locks.

Am I right or am I missing something ?

cheers,
Richard



From piet at cs.uu.nl  Sun Jan 30 14:34:34 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 30 Jan 2000 20:34:34 +0100
Subject: Imagemagick
References: <wzzotoal2k.fsf@sunshine.cs.uu.nl> <20000129201210.A4953@stopcontact.palga.uucp>
Message-ID: <wzu2jvwdyt.fsf@sunshine.cs.uu.nl>

>>>>> Gerrit Holl <gerrit.holl at pobox.com> (GH) writes:

GH> Piet van Oostrum wrote on 949163987:
>> Does anybody have a copy of Python/Imagemagick or pymagick or a working link?

GH> Google -> PythonMagick -> I'm feeling lucky
GH> http://starship.skyport.net/crew/zack/pymagick

But when you try to download something all links are dead alleys or invalid
files. 
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From aahz at netcom.com  Tue Jan 25 06:56:13 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jan 2000 11:56:13 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be>
Message-ID: <86k30t$i5r$1@nntp2.atl.mindspring.net>

In article <86joac$l2s$1 at mach.vub.ac.be>,
Thomas Hamelryck  <thamelry at vub.ac.be> wrote:
>Tim Peters <tim_one at email.msn.com> wrote:
>:
>: like-sex-is-great-but-should-not-be-part-of-life-ly y'rs  - tim
>
>Good analogy, actually. Sex is great, but should never be forced
>on you.

Uh, I think you missed the point.  Sex *is* part of life (as a general
statement regarding humans), and there's no real way to avoid it.

>BTW, people here seem to have adopted the idea that if you don't like
>a particular feature of a language, you should abandon it (e.g., "You 
>don't like indentation? Use Perl!"). But many languages have evolved 
>exactly this way: users criticize, complain, praize and whine, thereby
>influencing the shape of the languge in question. You can find a very 
>nice description of this process with respect to C++ in "The Evolution 
>of C++ : Language Design in the Marketplace of Ideas", by Jim Waldo 
>(Editor), James Waldo (Editor).

Uh, if you're going to refer to C++, I rest my case.

Note that many complaints about Python are taken seriously; however, the
indented block structure is considered a Core Feature of Python, and the
likelihood of its changing is pretty close to zero.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From wware at world.std.com  Sat Jan  1 23:05:46 2000
From: wware at world.std.com (Will Ware)
Date: Sun, 2 Jan 2000 04:05:46 GMT
Subject: fun with stock quotes
References: <Fnn296.A3w@world.std.com> <ncREqFA3nZb4Ew1G@jessikat.demon.co.uk>
Message-ID: <Fnowpn.Iun@world.std.com>

Robin Becker (robin at jessikat.demon.co.uk) wrote:
> here's my take on that which just returns a class with all the values

I had originally done that, and that's why I have something
erroneous in the comment about it being a class rather than a C
module. Once I started playing with the data, I found I was doing
things that ran too slow (the bottleneck seeming to be the linear
regression operation) so I started coding stuff in C.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com


From shang.remove_edu at st.jyu.fi.edu  Mon Jan 24 02:25:56 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Mon, 24 Jan 2000 09:25:56 +0200
Subject: Dictionary from a file
References: <388BF0C4.F0E08005@mail.bulkley.net>
Message-ID: <86gupk$pfb$1@mordred.cc.jyu.fi>

Peter Schopfer (aka Dr. J. - magician) <drj at mail.bulkley.net> wrote in
message news:388BF0C4.F0E08005 at mail.bulkley.net...
> Hello
>
> I have a dictionary that is too big to fit right in my program, so i
> was going to have it as a seperat file, labeled dic.py. how would i
be
> able to retrieve this dictionary. wuld this work:
>
> from dic.py import dictionary
> mydic=dic.dictionary

This works with a couple of changes. First, you omit the '.py'
extension in module names when importing them. Secondly, the 'from
module import stuff'-syntax inserts the 'stuff' into current module's
namespace, so you don't use the module name when referring to it. i.e.

from dic import dictionary
mydic = dictionary

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi




From smford at execpc.com  Wed Jan 26 23:54:54 2000
From: smford at execpc.com (Stuart Ford)
Date: Wed, 26 Jan 2000 22:54:54 -0600
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
References: <816010E2456BD111A48700805FBBE2EE01C60408@ex-quebec-u1.baan.com> <388FC511.C776341A@one.net.au>
Message-ID: <00a101bf6882$aa7cb540$9600a8c0@execpc.com>

How did this message make it into the list in the first place?

Did the spammer fake his/her sending address as that of a valid
list member, or is the list not sent to allow only registered members
to post?


Stuart




From fredrik at pythonware.com  Sat Jan 22 04:20:08 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 22 Jan 2000 10:20:08 +0100
Subject: Python Rules!
References: <200001211334.IAA28640@python.org> <86ajlq$661@news.or.intel.com>
Message-ID: <00b901bf64b9$e14c3ae0$f29b12c2@secret.pythonware.com>

tye4 <tye4 at yahoo.com> wrote:
> Yes, I think I'll let you guys wallow in your ignorance.

okay, here we are in the beginning of a new millenium (or at the
end of one, depending on your views).  and what do we see?

+ loads of media attention

+ leading software gurus advocating Python

+ more people than ever on the main Python conference,
  and Python tracks on other major software conferences.

+ Python interpreters shipping with major operating systems

+ publishers competing to offer full coverage of all popular
  aspects of Python (the core language, tkinter, jpython,
  python on windows, and more)

+ Python books topping bestseller charts <wink>

+ Python-based companies getting venture funding (Digital
  Creations last year, and we just got our first major invest-
  ment yesterday!)

+ major software companies actively using, pushing, or at
  least acknowledging Python in lots of different contexts.

+ a steady growth in downloads, server hits, and other python-
  related net traffic of 10% per month (or more).

etc. etc.

if our "ignorance" has taken us this far, I think we can keep on
going without the "help" of a single totally clueless anonymous
coward.

</F>

    "When a true genius appears in the world, you
    may know him by this sign, that the dunces are
    all in confederacy against him" -- Jonathan Swift




From marut at my-deja.com  Wed Jan 26 15:23:00 2000
From: marut at my-deja.com (marut at my-deja.com)
Date: Wed, 26 Jan 2000 20:23:00 GMT
Subject: parnassus? When will it be back up?
Message-ID: <86nl2t$rr5$1@nnrp1.deja.com>

When I visit the parnassus page I see only one word: Conn

Anyone know its status?


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


From pinard at iro.umontreal.ca  Thu Jan 20 10:10:21 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 20 Jan 2000 10:10:21 -0500
Subject: LISP Rocks! (was: Re: Python Rocks! :-)
In-Reply-To: "Tim Peters"'s message of "Wed, 19 Jan 2000 23:54:04 -0500"
References: <000801bf6302$60316ec0$31a2143f@tim>
Message-ID: <oqk8l4ixsy.fsf@titan.progiciels-bpi.ca>

"Tim Peters" <tim_one at email.msn.com> ?crit:

> > P.S. - SNOBOL was [...] Multi-punches all the time.

> Oh yes!  But typical SNOBOL4 programs were tiny compared to typical Pascal
> programs, and at the time the latter was introduced, my school's keypunches
> still lacked a semicolon key.

But for Pascal, the semi-colon was the only missing key.  Rather bearable.
SNOBOL, *that* was using a lot of "non-standard" characters, all the time.

Do I dream, or keypunches only had capital letters?  Printers did not have
small letters anyway.  I do remember that people were using digit suffixes
to mark diacritics, and using double or triple line spacing on the printer
as an attempt to render them.  Later, when ASCII became more widely used,
we got to use one `!' prefix to introduce cased letters, `!'-digit for
diacriticised letters, and two `!' prefixes for marking control characters
(with later ubiquitous conversion and processing, of course :-).

Life is so much fun in these days.  Yet, apparently, some feel miserable
because they ought to type on the space bar a few times once in a while! ;-)

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





From aahz at netcom.com  Tue Jan 18 14:21:52 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jan 2000 19:21:52 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <8610or$nsu$1@news.udel.edu> <861jm5$nrj$1@mach.vub.ac.be> <zqGEOEqNnxHC8gbhL5piinIM1bGb@4ax.com>
Message-ID: <862egg$4br$1@nntp3.atl.mindspring.net>

In article <zqGEOEqNnxHC8gbhL5piinIM1bGb at 4ax.com>,
Dennis Lee Bieber  <wlfraed at ix.netcom.com> wrote:
>
>	Mysterious errors usually means a mistake in the indentation
>applied. And to me, the indentation is easier to pick out then an
>"endif" on the wrong line...

Unless you've mixed tabs and spaces.  :-(
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From P.J.W.S.VRIJLANDT at INT.azg.nl  Thu Jan  6 12:04:29 2000
From: P.J.W.S.VRIJLANDT at INT.azg.nl (P.J.W.S. Vrijlandt)
Date: Thu, 6 Jan 2000 18:04:29 +0100
Subject: Simple cartesian product
In-Reply-To: <slrn877ovr.hbv.neelk@brick.cswv.com>
Message-ID: <18068D6C60@nds-3.azg.nl>

> Neel Krishnaswami <neelk at brick.cswv.com> wrote:
> 
> I realized this can be condensed even further:
> 
> def product(a, b):
>     return map(lambda n,a=a,b=b: (a[n/len(b)], b[n%len(b)]),
>                range(len(a)*len(b)))
> 
> It's still slower and more memory-hungry than the obvious for-loop,
> though.
> 

Here is my entry: (no len!)

def product(xs, ys):
    return reduce( lambda a, b, xs = xs, ys = ys: a + b
                 , map(lambda x, ys = ys: map(lambda y, x=x: (x, y)
                                              , ys
                                              )
                      , xs
                      )
                 ) 

print product([1, 2, 3], ['a', 'b', 'c'])                 

<layout-doesn't-make-obfuscated-oneliners-readable>-ly yrs

Patrick Vrijlandt










Met vriendelijke groet,

Patrick Vrijlandt



From jsliv at jslove.net  Sat Jan  1 19:50:55 2000
From: jsliv at jslove.net (Jay Love)
Date: Sat, 01 Jan 2000 19:50:55 -0500
Subject: Finding where Python is on a Unix server through Telnet?
References: <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com>
Message-ID: <386EA0EF.F7A72441@jslove.net>

Tyler Eaves wrote:
> 
> What is the command?
> 
> Tyler


try 'which python'
or for everything related to python 'locate python'


From hei at adtranzsig.de  Mon Jan 24 07:38:36 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Mon, 24 Jan 2000 13:38:36 +0100
Subject: tabs do WHAT?
References: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net> <86h5ep$iq$1@mach.vub.ac.be>
Message-ID: <948721268.245257@news.adtranzsig.de>

Thomas Hamelryck schrieb in Nachricht <86h5ep$iq$1 at mach.vub.ac.be>...
>Warren Postma <wpostma at earthlink.net> wrote:
>[snip]
>
>: If Guido reads this, please note my personal feature request for Python
2.0:
>: Make the "whitespace" stuff optional, for people who like weird
syntactical
>: features only. I'll love you forever. Thanks.
>
>This will not happen. People here (including Guido Himself) get extremely
picky
>when the use of indentation is criticised. They do not realize that this
feature

Go into a C++ newsgroup and suggest they get rid of the curlys.
Or go into a Lisp group and tell them you love Lisp to death
except for the brackets.
I think you'll harvest some responses like "Go away and use a
different language." What do you expect?
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net






From seanb at home.com  Tue Jan 18 20:53:47 2000
From: seanb at home.com (seanb at home.com)
Date: Tue, 18 Jan 2000 17:53:47 -0800 (PST)
Subject: More problems....
In-Reply-To: <2E7h4.230$Pe6.2431@newsfeed.slurp.net>
Message-ID: <200001190153.RAA30306@seanb.sttln1.wa.home.com>

On 18 Jan, TechSpood wrote:
> I have been reading tutorials for about 2 weeks now, and just now decided to
> put that knowledge to work, but I am having a huge problem...I will write a
> simple program (in the IDLE enviroment) then when I run it, it will run it
> til it gets to the first input statement, at which time it will stop reading
> the program.  I don't get a syntax error  or any other kind of error.  After
> I went back and read and re-read the tutorials i got very frusterated after
> not being able to find absolutly any errors in my code... I even went so far
> as to copy the code excactly as it was in a tutorial and put it in the
> compiler.... same damn problem.  Is there some option I need to turn on or
> some option that I might of accidently turned on that I wasn't supposed to?
> 
> Thanks
> 
> 


Is it possible that the program is witing for your input?
For example, the code:
	spam = input('Enter a line of python code, then press ENTER: ')
	print 'Your result is', spam
will print the prompt:
	Enter a line of python code, then press ENTER:
and then wait until a return.  Whatever the user entered is then run
through the python interpreter, and returned to the variable spam.
Finally, python will kindly print out the result.


-- 
Sean Blakey
(206)297-7123
quine = ['print "quine =",quine,"; exec(quine[0])"']; exec(quine[0])




From dworkin at ccs.neu.edu  Sun Jan 30 11:11:41 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 30 Jan 2000 11:11:41 -0500
Subject: Creating a local variable with a parameterized name
In-Reply-To: Gerrit Holl's message of "Sun, 30 Jan 2000 10:14:44 +0100"
References: <38937075.18B75753@rochester.rr.com> <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu> <20000130101444.A1582@stopcontact.palga.uucp>
Message-ID: <vndzotnlete.fsf@betelgeuse.ccs.neu.edu>

Gerrit Holl <gerrit.holl at pobox.com> writes:

>  Those things are called "dictionairies", not "directories". You could
>  use them for a directory structure, however :-)

I know that.  I have no idea why I used teh wrong word throughtout
that post.  Sigh.

-Justin

 



From tim_one at email.msn.com  Sun Jan  2 21:43:26 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 2 Jan 2000 21:43:26 -0500
Subject: Python language spec
In-Reply-To: <E124x1i-00053N-00@ethereal>
Message-ID: <000b01bf5594$4fab2440$a52d153f@tim>

[Nolan Darilek, by his own claim biting off more than he
 can chew <wink>]
> ...
> I'm confused by the expression_list definition, however. It
> reads:
>
> expression_list:      expression ("," expression)* [","]
>
> I understand the syntax of the definition, but I don't know
> what the syntax for an expression should be. Earlier rules
> define u_expr, a_expr, etc. but not a regular expression
> (I don't mean regexp, just a no-frills expression :). The
> only rule which defines 'expression' is:
>
> expression:     or_test | lambda_form
>
> which seems to imply that 'expression' is only a boolean
> result, which I'm guessing it isn't. :)
>
> [and various dead ends resulting from trying to talk himself
>  into believing the grammar doesn't mean what it says]

I don't have time to explain this in detail.  The short course is that, in
un-YACC-like fashion, Python's grammar makes operator precedence explicit in
the structure of the grammar.  The top level for expression is "or_test"
because the "or" operator has the lowest precedence.  When you look up
or_test, you'll find

or_test:        and_test | or_test "or" and_test

because the "and" operator has the next-lowest precedence.  Note that the
*structure* of the RHS forces all "and" expressions to be grouped together
before consuming an "or" (i.e., it forces "and" to "bind tighter" than
"or").  Then

and_test:       not_test | and_test "and" not_test

and on and on.  You'll eventually hit comparisons and arithemetic operators.
And, yes, a simple expression like "1+1" winds its way up the grammar
hierarchy all the way up to or_test.

Note that the official grammar lives in the file Grammar/Grammar (in the
source distribution); looking at that is a lot easier than trying to pick it
out of the Reference Manual.

> ... I don't completely understand why:
>
> expr: expr "*" expr { output }
> expr: expr "/" expr { output } ...
>
> wouldn't work?

Because you can't tell from those two rules whether "*" or "/" binds tighter
(they happen to have equal precedence, but throw in "+" too and the problem
will be more obvious).

You can rest assured that the grammar "works":  Grammar/Grammar is the input
to Python's own parser generator, and the output of the latter is what's
used to parse Python programs.

knowing-it-works-is-half-the-battle-ly y'rs  - tim





From aahz at netcom.com  Tue Jan 18 20:33:36 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jan 2000 01:33:36 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com>
Message-ID: <86349g$37r$1@nntp8.atl.mindspring.net>

In article <8631qm$26p at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>
>How hard can it be to add 'end' indicators to the language.. If these end
>indicators are added, the language would still be backward compatible. I
>can't believe nobody has fixed this for so long.

Are you suggesting the addition of 'end' indicators IN ADDITION to the
indentation?
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From ivanlan at callware.com  Tue Jan  4 18:12:54 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 04 Jan 2000 16:12:54 -0700
Subject: Grail
References: <387237F3.C35B7FA1@callware.com> <14450.17021.111722.710187@weyr.cnri.reston.va.us> <3872441A.7E86CCED@callware.com> <20000104220152.A3894@stopcontact.palga.uucp>
Message-ID: <38727E76.2C7487DA@callware.com>

Hi Gerrit--

Gerrit Holl wrote:
[snip]
> >
> > Um, yeah, I was hoping to get a Tk HTML widget, since the entire rest of
> > the application is already written in Tk.  An HTML widget would let me
> > use the existing HTML file for help without having to fire up netscape
> > or IE. ...
> 
> I think pysol uses this, have you read their source yet?
> 

Looks pretty good, but it doesn't seem to use tables.  I rely pretty
heavily on tables.  I could make a special version of the help without
tables, but I'd rather not.  I'll do it if the pysol help is all there
is, though.

Thanks,
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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From jstevens at basho.fc.hp.com  Fri Jan 21 18:53:40 2000
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: Fri, 21 Jan 2000 16:53:40 -0700
Subject: ANN: Open Source (Python) Design Competition
References: <388084F3.CD0EA1C1@nevex.com>
Message-ID: <3888F184.B12C8F6@basho.fc.hp.com>

Greg Wilson wrote:
> 
> * a unit and regression testing harness with the functionality of
>   XUnit, Expect, and DejaGnu.

For anybody who is interested in this:

I have a Python C extension that does "Expect" like stuff based on pcre
that was developed for the purpose of replacing DejaGnu.

Email if you want it, or want to team up on this kind of project. . .

-- 

If I spoke for HP --- there probably wouldn't BE an HP!

John Stevens
jstevens at basho.fc.hp.com


From tim_one at email.msn.com  Wed Jan 19 00:29:03 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 19 Jan 2000 00:29:03 -0500
Subject: Suggestions for python 2
In-Reply-To: <s89kep3peh6124@corp.supernews.com>
Message-ID: <000501bf623e$18b56a20$9a2d153f@tim>

[Evan Simpson]
> Now that two people have asserted that lexical scoping is
> the right way to handle this, I'd like to know what they
> *mean* by lexical scoping.

People with Scheme backgrounds mean "what Scheme does"; people with Pascal
backgrounds, "what Pascal does"; people with Perl backgrounds know they need
it but aren't really sure what anything means <wink>.

> How would names get bound to (or just initialized from?)
> names in lexically enclosing scopes?  Implicitly? Explicitly?
> Would we still declare module-global variables, or would
> they use the same mechanism as names in intermediate scopes?

It's a bit of a puzzle in a language without declarations; I'm not sure any
language has tried to "do" lexical closures without them (Scheme has "let" &
friends, Perl "my", & so on -- they're not declaring *types*, just declaring
which scope a *name* belongs to).

Channeling Guido <wink>, the rules for what's local wouldn't change.  A
reference to a non-local name would resolve to the closest-containing scope
in which the name *was* local (and, since the global and local namespaces
are the same at module level, this is a generalization of what happens
today).  Although I believe John Skaller warped "global" to mean "non-local"
in Viper, Guido wouldn't do that:  "global" would continue to mean "module
level".  Guido would not allow rebinding of a non-local name (except in that
a name explicitly declared global could, just as today, be rebound); "if
it's bound, it's local" would remain the rule.

That's one version of the truth; Viper has another; many other variants were
debated in the dim past.  The version above is The Truth because that's what
Guido actually implemented several years ago, before bumping into a nasty
cyclic trash problem (subsequently wormed around by Greg Ewing in his
lexical closure patch, very cleverly but incompletely and "too cleverly"
<0.9 wink>).

> I rather like the idea of having to explicitly bring names
> into an inner scope,

Me too -- except in the cases that bite most often, where a trivial little
lambda wants to refer to one or two measly little names from its context.
People stumble on that so often it's a FAQ.

> and with semantics identical to parameter defaults.

Sorry, this part's unclear.

> Why is this so objectionable?  Is it just the syntax?

I've getting over a bad flu, and not thinking entirely clearly yet, but when
I read Markus Stenberg's

> Ugh.. *puke* this looks like ugly hack from hell

I kinda figured, ya, maybe there is some small objection to the syntax out
there <wink>.

when-guido-sez-"too-radical"-he-means-"not-radical-
    enough"-ly y'rs  - tim





From backov at nospam.csolve.net  Fri Jan 14 14:00:21 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Fri, 14 Jan 2000 19:00:21 GMT
Subject: Compiling Python under Borland C++ Builder 4
Message-ID: <9nKf4.59481$Ke.56920@news21.bellglobal.com>

I've found out that my previous problems with access violations in C++
Builder seem to be due to the fact that the .lib file is not quite syncing
up correctly to the .dll. At least I hope so, either that or my compiler is
screwed. Anyway, the problems I have under BCB go away under VC++ 5.

So, I'm just curious: Are there any C++ Builder types here who have compiled
it under BCB4? I imported the VC++ project and compiled it successfully, but
I get a 10k DLL, which I assume is probably incorrect. ;>

Cheers,
Jason





From gmcm at hypernet.com  Sat Jan  8 13:02:03 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 8 Jan 2000 13:02:03 -0500
Subject: ANN: Installer - Linux beta 01, Windows beta 3f
Message-ID: <1264767970-16262245@hypernet.com>

The new betas are available at: 
http://starship.python.net/crew/gmcm/distribute.html

The Windows beta is bug fixes:
  - run from CDROM works
  - some path problems fixed
  - some case-mismatch problems fixed

The Linux beta is the first release for this platform. It includes these 
capabilities:
 - turning a script into an isolated single-directory installation
 - building compressed Python code archives
 - building zip-like archives that can contain anything

(Actually, the latter two capabilites should work on any Unix platform).

Installations created this way are completely insulated from any Python 
installations on the target machine. There is no problem using shared 
extension modules; these will be detected and made part of the installation.

License: 
 Python style (do as thou wilt; maintain copyright notices).

Contact: gmcm at hypernet.com

<P><A HREF="http://starship.python.net/crew/gmcm/distribute.html">
Installer: Linux beta 01, Windows beta 3f</A>  Distribute Python apps, 
modules and packages. (8-Jan-00).

- Gordon



From m.faassen at vet.uu.nl  Thu Jan  6 21:31:29 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jan 2000 02:31:29 GMT
Subject: Now What? :o)
References: <n2bd4.64$%Ph.18874529@news.randori.com> <387541F5.1606D7C6@rubic.com>
Message-ID: <853j61$iuq$2@newshost.accu.uu.nl>

Jeff Bauer <jbauer at rubic.com> wrote:
[snip]
> 2. Open binary files in binary mode.  Unix programmers
> often neglect to use the binary switch since all Unix
> files *are* binary.

No, they're all text! :)

Regards,

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


From jkraai at murl.com  Mon Jan 24 04:48:27 2000
From: jkraai at murl.com (jim kraai)
Date: Mon, 24 Jan 2000 03:48:27 -0600
Subject: ConfigParser and .ini files
References: <86a1s2$7ha$1@nnrp1.deja.com> <14472.39533.223900.313932@weyr.cnri.reston.va.us>
Message-ID: <388C1FEB.D4CDAD6E@murl.com>

"Fred L. Drake, Jr." wrote:
> 
> super_banane at yahoo.com writes:
>  > The ConfigParser class in the default Python libraries
>  > does not support comments given on the value lines, separated
>  > by a semicolon, such as:
>  >
>  > [Section]
>  > keyword = value ; put your comments here
> 
>   You're right, it doesn't.
>   Does anyone know of a *real* specification for the .ini format?  I
> didn't know this was allowed.

Reference to a *real* spec and why specs are invalid can be found in the 
first paragraph of the "Introduction" at:
http://www.cloanto.com/specs/ini.html

XML.com has some (surprise!) regexp-based .ini parsing Perl code at:
http://www.xml.com/pub/norm/part1/getstart1.html

If you read TCL, an .ini manager (looks slow-as-can-be) can be found at:
http://waster.8m.com/iniparse/

--jim



From P.J.W.S.VRIJLANDT at INT.azg.nl  Mon Jan 17 05:11:03 2000
From: P.J.W.S.VRIJLANDT at INT.azg.nl (P.J.W.S. Vrijlandt)
Date: Mon, 17 Jan 2000 11:11:03 +0100
Subject: Stopping a thread from the calling process
In-Reply-To: <1264075595-4560668@hypernet.com>
References: <FoFsDH.8EL@boss.cs.ohiou.edu>
Message-ID: <EF5CE20F7E@nds-3.azg.nl>

Gordon wrote:

> Scott Barron asks:
> 
> > is there an easy and fairly portable way of doing this?
> 
> If you mean "interrupting" or "killing", no. WIth good reason. 
> OS level threads are often "light weight" processes. Generally 
> what's been trimmed is all the stuff dealing with reclaming 
> resources, releasing locks etc. on thread death. So if you 
> "kill" a thread, you're likely to be leaving your process in an 
> indeterminate state.
> 
> The normal idiom is to ask politely - have the thread 
> periodically check some variable to see if it should stop.

How would you do this if the thread-to-be-stopped is waiting for 
(keyboard-) input?

e.g. in telnetlib we have:

    def mt_interact(self):
        """Multithreaded version of interact()."""
        import thread
        thread.start_new_thread(self.listener, ())
        while 1:
            line = sys.stdin.readline()
            if not line:
                break
            self.write(line)

    def listener(self):
        """Helper for mt_interact() -- this executes in the other  
        thread."""         
        while 1:                        
            try:
                data = self.read_eager()
            except EOFError:
                print '*** Connection closed by remote host ***'
                return
            if data:
                sys.stdout.write(data)
            else:
                sys.stdout.flush()

mt_interact should break out of its loop when EOFError occurs within 
listener.

(also note that select on input won't work on win32, and looping on 
keypressed seems rather processor-unfriendly)

BTW: Would this be better rewritten (for this reason or another) 
using threading? 

Patrick
Met vriendelijke groet,

Patrick Vrijlandt



From gerrit.holl at pobox.com  Sat Jan 29 14:04:01 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 29 Jan 2000 20:04:01 +0100
Subject: Imagemagick
In-Reply-To: <wzzotoal2k.fsf@sunshine.cs.uu.nl>; from piet@cs.uu.nl on Sat, Jan 29, 2000 at 05:39:47PM +0100
References: <wzzotoal2k.fsf@sunshine.cs.uu.nl>
Message-ID: <20000129200401.A4891@stopcontact.palga.uucp>

Piet van Oostrum wrote on 949163987:
> Does anybody have a copy of Python/Imagemagick or pymagick or a working link?

A google search on "Python ImageMagick" returns 2890 hits, and google
has pages cached. Are you sure you can't find it?

parnassus-isn't-god-_yet_-ly y'rs - gerrit

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From mikael at isy.liu.se  Wed Jan 12 07:58:28 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Wed, 12 Jan 2000 13:58:28 +0100 (MET)
Subject: converting tuple to a list
In-Reply-To: <tHZ8OL+ftbNGsmad91aonRmEC1nh@4ax.com>
Message-ID: <XFMail.000112135828.mikael@isy.liu.se>

On 12-Jan-00 Anders M Eriksson wrote:
 >  Is this the 'best' way to convert a tuple to a list?
 >  
 >  t = (1,2,3,4,5)
 >  l = []
 >  
 >  for i in range(len(t)):
 >          l.append(t[i])

I don't know if this is better, but at least it looks better to me.

l=map(None,t)

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    12-Jan-00
Time:    13:57:01

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From future at dotstar.net  Sun Jan 30 12:20:22 2000
From: future at dotstar.net (Patrick Cooper)
Date: 30 Jan 2000 11:20:22 -0600
Subject: Choosing a language
Message-ID: <01bf6b45$ff345b40$412b69ce@default>

I want to start learning computer programming and from what I have found so
far I have chosen to start with Python.  I am going to be focusing on CGI
first but eventually want to work with desktop applicates.  Python seems to
be the most versitle and easiest to learn.  Before I go in any more depth
does anyone think I should start with a different language?  I want to
eventually learn C, but think Python is the best place to start.

Also, since I want to focus on CGI, what would be the best resources to
check out?  Anything besides what's listed in  the CGI Tools section of the
Python web site?



  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From lalal at fd.com  Sat Jan 22 02:29:06 2000
From: lalal at fd.com (lalal at fd.com)
Date: Sat, 22 Jan 2000 01:29:06 -0600
Subject: get money for browsin web sites! 57818
Message-ID: <220100012906@fd.com>

Everyone likes making easy money.  This place pays you 50 cents for every hour you browse the web!!!
Get even more if you refer people!  I just got a check for $36.00!!!

Sign up here:
http://www.alladvantage.com/go.asp?refid=IMY627








Y?3 at t10KQV


From mfletch at tpresence.com  Wed Jan 19 20:40:30 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 19 Jan 2000 20:40:30 -0500
Subject: Python Rocks!
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3085@VRTMAIL>

Hmm, guess I'll propose this again, just for fun:

Create an editor which parses into "statement suites".  Provide different
viewing modes, one using whitespace (of whatever style) to denote suites,
one using {}'s to denote suites, and finally, one for those really weird
people who actually want to type "END SOMETHING" every time a suite ends.
Within the context of the view, interpret everything according to the user's
preference i.e. cut-and-paste of the text creates a new set for analysis of
suites.

Then, when that's done, point the disaffected to said editor so we don't
clog up the newsgroup with these circular discussions of preferred style (it
just being style in the end).  In a few weeks, they will either decide they
prefer the cleaner syntax of the Python-style, or they won't, they will
either switch to using indentation-only for the code, or they won't.

And, sparrow, you must ask yourself in the end if it will change your inner
being, or if it is merely part of the illusion,
Mike




From aahz at netcom.com  Tue Jan  4 10:53:48 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 4 Jan 2000 15:53:48 GMT
Subject: CPAN for Python?
References: <38720F70.A5DC6F33@ttm.com>
Message-ID: <84t52c$1gg$1@nntp9.atl.mindspring.net>

In article <38720F70.A5DC6F33 at ttm.com>,
Scott Anderson  <sanderson at ttm.com> wrote:
>
>Is there an equivalent of Perl's CPAN for Python?

Well, there's Parnassus (http://www.vex.net/parnassus/), but there's not
yet an overarching authority.  OTOH, Python's "built-in" libraries are
more extensive than Perl's; see the Library Reference for details.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From tlu at bmo.com  Fri Jan 28 08:39:20 2000
From: tlu at bmo.com (Tao Lu)
Date: Fri, 28 Jan 2000 08:39:20 -0500 (EST)
Subject: lib reference in Unix man format
Message-ID: <200001281339.IAA12329@trm.bmo.com>

27 Jan 2000 23:37:56 GMT (Cliff Crawford) <cjc26 at nospam.cornell.edu> wrote:
> 
> Pada Thu, 27 Jan 2000 10:48:53 -0500 (EST), Tao Lu bilang:
> | Where can I find the lib reference in Unix man format? Forgive me if this 
has 
> | been asked before.
> | 
> | Have been programming in Perl for years, I found those 
perlfunc/perlop/perlfaq 
> | are very convenient. Any similar work done for Python developers/newbies?
> 
> There isn't any documentation in man format.  I know this because Tom
> Christiansen came over here once and VERY LOUDLY complained about the
> lack of this. :)  Actually I find this annoying too, so maybe I'll
> work on converting the on-line docs to man pages (when I have some
> free time...:).  I'd prefer to have each built-in function have its
> own page though, rather than glomming them all together into one
> large "pythonfunc" page.
> 

If each built-in function has its own man page, I have to know the function name 
before I can find its man page. This will be like the manual of Unix system 
calls. If I don't remember the name of message send operation, i.e., msgsnd, I 
have to use 'man -s 2 intro' to figure out.

What I like 'perlfunc' most is that I can find similar function as well -- I may 
also learn some thing new every time when I browse 'perlfunc'.  ;-)


Fri, 28 Jan 2000 00:34:29 +0100 Pierre Imbaud <pierre at saiph.com> wrote:
: Unix man format is pretty poor. U can find the lib reference as html,
: not bad,
: or as info file, browsable under emacs, xemacs or a specific gnu info
: browser.

Where can I find those info file browsable in xemacs? TIA.

--




From tmick at mail.com  Wed Jan  5 20:08:02 2000
From: tmick at mail.com (tmick at mail.com)
Date: Thu, 06 Jan 2000 01:08:02 GMT
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk>
Message-ID: <850pt8$qv9$1@nnrp1.deja.com>

I found _Learning Python_ to be a much better read than _Programming
Python_ (better presentation and it ends up being a decent reference as
well, at least as far as the core language syntax goes).

I have not read _Python Essential Reference_. All I can say about it is
that I have heard that Guido's contribution to it is minimal and that
his name is prbably there more for name recognition. That says nothing
about the quality of the book though.

Trent

In article <85035b$1sb$1 at bignews.shef.ac.uk>,
  Bala <engbpa at liquidinformation.com> wrote:
> I'm now thinking of getting either
>         Python Essential Reference (OTHER NEW RIDERS)
>         by David M. Beazley, Guido Van Rossum
> or
>         Learning Python (Help for Programmers)
>         by Mark Lutz, David Ascher, Frank Willison (Editor)
> Which one of these two would you recommend?
>
> Thanks.
>
> -Bala
>


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


From thomas at xs4all.net  Mon Jan 31 14:12:58 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 31 Jan 2000 20:12:58 +0100
Subject: Why Perl do But Python can't?
In-Reply-To: <001401bf6c16$d5657660$3acbd9c2@peridot.optichrome.com>; from a.eyre@optichrome.com on Mon, Jan 31, 2000 at 06:13:11PM -0000
References: <z8kl4.783$S3.31411@newsfeed.avtel.net> <001401bf6c16$d5657660$3acbd9c2@peridot.optichrome.com>
Message-ID: <20000131201258.G19725@xs4all.nl>

On Mon, Jan 31, 2000 at 06:13:11PM -0000, Adrian Eyre wrote:

> > : Scripts (of any sort) are inherently unsafe when run setuid or setgid.
> > : See the perlsec manpage for why.
> > 
> > This is not true, and is a myth people need to get over.  The danger
> > Kelly referred to exists on some Unix-like systems and not on others.
> > On a system where they are safe, there is no reason whatsoever to use
> > a setuid wrapper.
> 
> Just out of interest, what is the reason why this in unsafe. I couldn't
> see the difference between a script and a binary.

When you try to execute a script (using any of the exec*() family C calls)
the UNIX (or perhaps C library on some systems) opens the program in
question and tries to determine what kind of binary format it is. If it's a
pure binary, it gets handled as pure code, executed right away. If it has
shared libraries of any kind the runtime link/loader is called in to do the
setup. But if it's a script, it has to do something different :P

It has to read the first line of the file, strip the #!, check if the
interpreter exists, and exec()'s that, with a modified argv: basically the
shebang line, minus the #!, with the script name and the original arguments
to the script attached. Thus,

spam.py:
-
#!/usr/bin/python -X
-

centurion: ~> ./spam.py -i10 foobar


becomes (pseudocode)

execv("/usr/bin/python -X spam.py -i10 foobar")

If setuid bits on scripts where allowed, it would have to do the exec()
running as the owner of the file already, so that '/usr/bin/python' is
called as the right user. But this means there is a possible race condition,
using symlinks (or hard links, if you are on the same disk.)

Say a system I work on has a setuid-root script somewhere, like
/sbin/fixperms. It has its own security-checking inside the script, so it's
setuid-root and executable by everyone, or perhaps by a limited group.

Mr. M. Alicious wants to hack me, and he can execute the above script, but
is stopped by the internal security checks. What he does is this:

--- hack.py --
#!/usr/bin/python

import os

os.execv("/bin/sh",("/bin/sh",))
--- hack.py --

hack.py is owned by alicious, just executable, not setuid. Now Mr. A.
creates a symlink to /sbin/fixperms, say, 'runme', and then starts to create
a heavy load on the machine (not necessary, but makes the next step a lot
easier.) preferably by generating lots of swap-activity, or lots of I/O on
the same disk the python executable is located on.

Then Mr. Alicious starts /sbin/fixperms _via the symlink_, and at the same
time sets the symlink to hack.py. He has to take care to replace the symlink
at exactly the right moment, though, between the setuid-bit checks in the
kernel/libc and the opening of the script by python. On a heavily loaded
machine this isn't that big a feat. Python gets started as root, opens
'runme', and gets the contents of hack.py. and exec()'s /bin/sh, giving Mr.
M a nice root-owned shell.

There are a couple of solutions possible, unfortunately, most of these are
incompatible in some minor way with how it is currently done. The most
common one is not to pass the script to be run by the interpreter as a
command line, but instead as an open filedescriptor, fd 3. The interpreter
would have to be altered to accept that, however(*), and since UNIX in
general has a lot of different interpreters.. icky. 

Another option would be for the python interpreter itself to check for
setuid bits... I only thought that up just now, so I'm not sure if there's a
problem with that, but now that I look at the perlsec manpage, I believe
this is what suidperl does; Python could check for setuid or setgid bits on
the scripts it is about to execute, and invoke a special setuid-root part of
the interpreter (seperate binary, of course) to re-invoke python with the
proper userid/groupid (in a race-free manner, such as passing an open fd.)
Perhaps for backward compatibility it would be best only to invoke the
setuid part of python if python is started with, say, the -s option. 

Any takers on that idea ? It wouldn't be hard to write, I bet ;) But then,
it's probably already been suggested and rejected before ;)

*) on Linux systems, passing an open fd is actually fairly easy and fairly
compatible; old (or perhaps just all) interpreters could be started with
'/proc/fd/3' as the script. There was talk about this on the linux kernel
list a year or so ago (and probably more often before and after that) but
the problem was, and probably still is, that it would break interpreters
that rely on opening a filedescriptor right after starting and assuming it
will be fd 3.

Paranoid-android-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From landauer at apple.com  Wed Jan 26 18:46:27 2000
From: landauer at apple.com (Doug Landauer)
Date: Wed, 26 Jan 2000 15:46:27 -0800
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net>
Message-ID: <landauer-4BDC4F.15462726012000@news.apple.com>

In article <86joac$l2s$1 at mach.vub.ac.be>,
Thomas Hamelryck  <thamelry at vub.ac.be> wrote:
>
> But many languages have evolved exactly this way: users criticize,
> complain, praize and whine, thereby influencing the shape of the
> languge in question.

Yes, this works well in many cases, including Python.  Notice that
it is, for the most part, *non*-users of python who complain about
the indentation.  _Users_, for the most part, like it.  Languages
that change by catering to the tastes of non-users tend not to do
so well.
-- 
    Doug Landauer        landauer at apple.com (work)
                         landauer at scruznet.com (not-work)


From tim_one at email.msn.com  Mon Jan 10 03:19:27 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 10 Jan 2000 03:19:27 -0500
Subject: [ANN] Multimethod.py -- multimethods for Python
In-Reply-To: <slrn87d2lv.i6r.neelk@brick.cswv.com>
Message-ID: <000801bf5b43$69184d40$532d153f@tim>

[posted & mailed]

[Neel Krishnaswami]
> [This is a repost, since the original apparently never made
>  it; if you see it twice, my apologies.]

This is the only one I saw.  Thanks for reposting it.

> I'd like to announce the existence of Multimethod.py, a package
> that implements real (eg, CLOS/Cecil/Dylan) multimethods for
> Python.
> ...
> You can find it on my webpage at:
>
>   <URI:http://www.sff.net/people/neelk/open-source>

I posted a (unsurprisingly!) similar pkg about 5 years ago, modeling Cecil's
dispatching rules in Python.  It drew a grand total of 0 responses, so
you're *already* doing infinitely better than I did <wink>.

Can't say I was surprised -- few people had thought about multimethods then.
They're better known now, but also much easier to get at via Dylan now
(which people should definitely look into if they like what they see in your
emulation).

> ...
> (Incidentally, is this the sort of thing that should go on the
> types-sig?)

Generally, I'd say yes, but it's certainly welcome on c.l.py too.

The current incarnation of the Types-SIG has "a single specific goal: to
develop an optional static typing system for Python", and a discussion of
multimethods may be seen as defocusing (it wouldn't by me, but then the
Types-SIG has its hands full regardless these days so it wouldn't get any
mind share).  John Skaller posted a short piece about multimethods late last
year to the SIG, but it got buried under the static-typing traffic.

BTW, the surest way to kill any discussion in the Python world has
traditionally been to post some code <0.5 wink -- there is A Serious Theory
that the *last* incarnation of the Types-SIG self-destructed because Jim
Fulton implemented his ideas!>.

it's-a-strange-and-altogether-not-entirely-undisgusting-
    world-ly y'rs  - tim





From jok at stotze.dyndns.org  Mon Jan 10 21:13:12 2000
From: jok at stotze.dyndns.org (Jochen Knuth)
Date: Tue, 11 Jan 2000 03:13:12 +0100
Subject: Zope and NT databases
References: <387A67AD.32F2E22B@bby.com.au>
Message-ID: <387A91B8.586E126E@stotze.dyndns.org>

Sarah Burke wrote:
> 
> Well, how about if I was using a client/server database, like Microsoft's SQL
> server?
> 
> Sarah
> 

If it's SQL Server 6.5 you can use the Zope Sybase DA from
http://www.zope.org/Products/DA/SybaseDA if you have the developer
libraries for Sybase.

There is no other ready-made solution for this problem.

Ciao,
Jochen


From info at pythonware.com  Sun Jan 23 03:11:41 2000
From: info at pythonware.com (PythonWare)
Date: Sun, 23 Jan 2000 09:11:41 +0100
Subject: ANN: Python training in Sweden!
Message-ID: <000901bf6579$7bf53960$f29b12c2@secret.pythonware.com>

Secret Labs AB is proud to offer Python training on
a number of sites in Sweden.

The first sessions are scheduled for March.  For more
information, see our site:

    http://www.pythonware.com/training/
    (in swedish only)

-- regards
the pythonware training team




From parker at gol.com  Thu Jan 27 07:42:35 2000
From: parker at gol.com (Ian Parker)
Date: Thu, 27 Jan 2000 12:42:35 GMT
Subject: Python startup seems slow... (NT)
References: <MPG.12f7f19a8fa3be15989680@or.news.verio.net>
 <14478.25750.656580.780253@beluga.mojam.com>
Message-ID: <nGHQIGAI+2j4Ewkl@gol.com>

In article <14478.25750.656580.780253 at beluga.mojam.com>, Skip Montanaro
<skip at mojam.com> writes
>
>    Jason> Under NT, it seems that starting up python to process even a
>    Jason> trivial script is pretty slow.  With the following python
>    Jason> program, helloworld.py:
>
>    Jason>   print "Hello World"
>
>    Jason> it takes a second or more for "python helloworld.py" to execute
>    Jason> on my machine.  It doesn't get any faster the second or third
>    Jason> time around.
>
>I rather suspect you have a problem of some sort.  On my Linux system I time
>the following command at under 0.05s elapsed time:
>
>    python -c "print 'hello world'"
>
>run from the command line.
>
>    Jason> Is there anything I can do to improve the performance?  Is this a
>    Jason> known problem?
>
>Not a known problem that I'm aware of.  If simple scripts like that with no
>imports took a second to run in general on NT, I'm sure we'd have heard
>about it by now.  (Not criticizing, just suggesting you turn over a few
>rocks on your system.)  I seem to recall that NT has a pretty nice system
>resource monitoring tool.  Are you running your script on a completely
>quiescent system?
>
>Skip Montanaro | http://www.mojam.com/
>skip at mojam.com | http://www.musi-cal.com/
>847-971-7098
>

I've just tried this on a 400Mhz Pentium III running NT4 / SP4.  It runs
in just under a second the first time, and much faster subsequently.  OF
course it may depend on what else is running on the PC.

Regards

Ian
-- 
Ian Parker


From thamelry at vub.ac.be  Mon Jan 31 04:38:55 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 31 Jan 2000 09:38:55 GMT
Subject: Eight suggestions for Python books (long)
References: <87268i$786$1@nnrp1.deja.com>
Message-ID: <873l7f$be4$1@mach.vub.ac.be>

gvwilson at nevex.com wrote:

[snip]

: 7. Object-Oriented Analysis and Design Using Python and UML

[snip]

Now this is a book I would buy! 

BTW, I don't think a python cookbook would make much sense. 
Python is so simple that in most cases you don't need to consult a
book to find the solution. In most other cases, the currently available 
books suffice.

Cheers,

---
Thomas Hamelryck    Institute of Molecular and Structural Biology
Aarhus University   Gustav Wieds Vej 10C
DK-8000 Aarhus C    Denmark         



From aa8vb at yahoo.com  Tue Jan 11 11:36:56 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 11 Jan 2000 11:36:56 -0500
Subject: [ANN] Multimethod.py -- multimethods for Python
In-Reply-To: <387B19F2.AC17FA8D@home.com>
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com>
Message-ID: <20000111113656.A42270@vislab.epa.gov>

Doug Hellmann:
 |Tim Peters wrote, in response to Neel Krishnaswami's post on multimethods:
 |> 
 |> Can't say I was surprised -- few people had thought about multimethods
 |> then.  They're better known now, but also much easier to get at via
 |> Dylan now (which people should definitely look into if they like what
 |> they see in your emulation).
 |
 |Could one of you post a brief description of what a multimethod is, or
 |maybe just a pointer to a definition somewhere?

I had the same question yesterday and did a little surfing.  Basically,
instead of switching off just the type of 'self' to determine which method
to call, you switch off the type of other arguments as well.

If you think best in implementation as I do:

   def f( a,b ):
     if   type(a) == Type1 and type(b) == Type1: f_t1_t1(a,b)
     elif type(a) == Type1 and type(b) == Type2: f_t1_t2(a,b)
     elif type(a) == Type1 and type(b) == Type3: f_t1_t3(a,b)
     ...

My impression is that it is the above, without as much mess and maintenance
headache as a codified dispatch table.  

The first paragraph of the Intro in:

    http://www.cs.washington.edu/research/projects/cecil/cecil/www/www/\
    Papers/ecoop99.html

sums it up.

-- 
Randall Hopper
aa8vb at yahoo.com



From johnjensen at my-deja.com  Thu Jan  6 00:19:32 2000
From: johnjensen at my-deja.com (johnjensen at my-deja.com)
Date: Thu, 06 Jan 2000 05:19:32 GMT
Subject: XML to dictionary?
Message-ID: <8518kq$59n$1@nnrp1.deja.com>

I have a relatively simple XML question, but I can't seem to figure out
how to do it simply and easily with the xmllib module.

Assume I have some XML (stylized below):

<Newsgroup>
  <comp.lang.python>
    <Message1>
      <Subject>Hi there</Subject>
      <Body>This is a test</Body>
    </Message1>
  </comp.lang.python>
</Newsgroup>

How can I quickly and easily stuff this into a nested dictionary, so
that:

print dict['Newsgroup']['comp.lang.python']['Message1']['Subject']
spits out 'Hi there'

The XML I am working with is perfect for this type of solution.

Any help would be appreciated!

John


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


From cwr at crash.cts.com  Mon Jan 24 00:53:12 2000
From: cwr at crash.cts.com (Will Rose)
Date: 24 Jan 2000 05:53:12 GMT
Subject: Python Rocks! - get rid of colons
References: <000901bf65e8$44a75be0$022d153f@tim>
Message-ID: <86gpc7$dpk$2@thoth.cts.com>

Tim Peters <tim_one at email.msn.com> wrote:
[...]
: Me neither, although I reject the assumption that a language good for
: learning is necessarily bad for "real life"; to the contrary, I think Python
: already stands as a strong counterexample (although ABC does not).  Guido's
: CP4E proposal

:     http://www.python.org/doc/essays/cp4e.html

: is definitely aimed at newbies, but is much more concerned with the
: development environment than with the language per se.  There are several
: reasons to suspect that a few aspects of Python today are "plain wrong" for
: newbies (case sensitivity and "lossy" integer division get mentioned most
: often) -- but many experienced programmers gripe about the same things.

: Unfortunately, "experienced programmers" are all over the map, usually
: griping about whatever doesn't match the last language they used <0.4 wink>.

: e.g.-you're-probably-irked-about-indentation<2.0-wink>-ly

I can live with the indentation, tho' its a nuisance not to be able to
format the code for readability.  However, if python goes case-insensitive,
I'm gone.  I already have enough fights with case insensitive/case preserving
filesystems.  If you want to simplify that aspect of the language, go to a
single case - preferably upper case, so that the name of the DEITY can be
correctly invoked.


Will
cwr at crash.cts.com





From edcjones at erols.com  Wed Jan 26 10:34:35 2000
From: edcjones at erols.com (Edward C. Jones)
Date: Wed, 26 Jan 2000 10:34:35 -0500
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com>
Message-ID: <388F140B.C0AA1C99@erols.com>

Does anyone know of a good book on recent programming language
concepts?





From robin at jessikat.demon.co.uk  Mon Jan 17 06:46:52 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 17 Jan 2000 11:46:52 +0000
Subject: Proposal: i-String literals in Future Version of Python?
References: <m12A9j5-000CxTC@artcom0.artcom-gmbh.de>
Message-ID: <UE2k2BAsEwg4EwbR@jessikat.demon.co.uk>

In article <m12A9j5-000CxTC at artcom0.artcom-gmbh.de>, Peter Funk
<pf at artcom-gmbh.de> writes
>If someone is going to internationalize an application in Python, 
>the source will become slightly less readable.  
>The following example statements
>        print "This is just an annoying message"
>       b = Tkinter.Button(text="Foo Bar", ...)
>will turn into
>        print _("This is just an annoying message")
>        b = Tkinter.Button(text=_("Foo Bar"), ...) 
>or even more ugly:
>        print gettext("This is just an annoying message")
>        b = Tkinter.Button(text=gettext("Foo Bar"), ...)
>or
>        print i18n("This is just an annoying message")
>        b = Tkinter.Button(text=i18n("Foo Bar"), ...)
>
>Here the *gettext* method (function) has to return a string, which
>is possibly translated into a foreign language.  Since Python already
>knows raw string literals, I thought that it would be nice, if a
>similar notation could be used to mark i18n string literals.
>So that some magic in the python interpreter expands
>       print i"This is just a message"
>into a call to a gettext function internally, if such a function is
>available.  May be there could be a sys.i18n function which could be 
>initially defined as a identity function:
>       def i18n(string):
>               return(string)
>This way preparing applications for i18n would become even easier.
>What do you think?
>
>Regards, Peter

simpler the better. There was a discussion last year about prefix usage
in relation to documentation etc. However isn't  the real problem that
of getting all the strings into a convenient form/place so that they can
be nicely translated into other languages/charsets. I guess you could
always automatically extract these and replace with
german/french/spanish etc etc using a preprocessor.
-- 
Robin Becker


From loewis at informatik.hu-berlin.de  Thu Jan 20 07:44:16 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Jan 2000 13:44:16 +0100
Subject: Can TkInter be frozen?
References: <1263967434-11066381@hypernet.com>
 <BF71A72C2C3451AA.44DFB5F8DFE97D59.0BDAEAB17E554E7D@lp.airnews.net>
Message-ID: <p6q66wodian.fsf@informatik.hu-berlin.de>

claird at starbase.neosoft.com (Cameron Laird) writes:

> More gossip:  Tcl has at least a half-dozen common ways
> to make installations (in principle, they're in <URL:http://
> starbase.neosoft.com/~claird/comp.lang.tcl/tcl_compilers.html>,
> although that page is uninviting for a person pursuing this one
> topic in isolation).  None of them are sufficiently common to
> encourage adoption into a Python distribution.

In addition to those, I've been using Tix for quite some time now. Tix
provides SAM (Stand-alone modules), which are frozen Tcl modules. It
generates (on Unix), libtclsam.a, libtksam.a, and libtixsam.a. If you
link those into your frozen application, together with libtcl.a,
libtk.a, and libtix.a, you get a truly stand-alone binary.

Regards,
Martin



From choffman at dvcorp.com  Thu Jan 13 15:21:52 2000
From: choffman at dvcorp.com (choffman at dvcorp.com)
Date: Thu, 13 Jan 2000 20:21:52 GMT
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com> <002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com> <jtu2kh52y6.fsf@wazor.biostat.wisc.edu>
Message-ID: <85lc4d$gk2$1@nnrp1.deja.com>

In article <jtu2kh52y6.fsf at wazor.biostat.wisc.edu>,
  William Annis <annis at biostat.wisc.edu> wrote:
> "Fredrik Lundh" <fredrik at pythonware.com> writes:
>
> > # -*-python-*-
> >
> > or
> >
> > #  -*- Mode: Python; tab-width: 4 -*-
>
>         Those all work, but as I recall those work only on the first
> line of the file.  A quick test seems to confirm this.

Check out the Emacs Info system for the topic Local Variables.

In short, at the *END* of the file you can have the following:

### Local Variables:
### mode:python
### tab-width:4
### End:

The important bits are a line containing "Local Variables:" , a line
containing "End: ", then any number variable names and values set
between those two lines. The lines can start with a comment character
appropriate for the contents of the file (e.g. "###" for Python).


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


From fdrake at acm.org  Fri Jan 21 14:10:21 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 21 Jan 2000 14:10:21 -0500 (EST)
Subject: ConfigParser and .ini files
In-Reply-To: <38889FB1.F692A45D@callware.com>
References: <86a1s2$7ha$1@nnrp1.deja.com>
	<14472.39533.223900.313932@weyr.cnri.reston.va.us>
	<38889FB1.F692A45D@callware.com>
Message-ID: <14472.44829.865615.286152@weyr.cnri.reston.va.us>

Ivan Van Laningham writes:
 > The spec is how Windows treats .ini files. ...

  That's a huge problem for those of us who don't often use Windows.
If anyone has tried to document the format with real details, I'd love 
to get a pointer to the description.

 > 1)  Section headers of the form "[Button States]" come up as
 > ParsingError because of the whitespace.  I read the regex for the
 > headers as allowing whitespace in the header, so what's going on here? 
 > Here's the regex:
 > 
 >  __SECTCRE = re.compile(
 >         r'\['                                 # [
 >         r'(?P<header>[-\w]+)'                 # `-', `_' or any alphanum
 >         r'\]'                                 # ]
 >         )
 > 
 > Not being a regex guru (not even an acolyte), I'm probably unqualified
 > to evaluate it.  What should this be to allow whitespace in section
 > names?  [*lots* of .ini files have whitespace in the section names. ...]

  Add a space before or after the \w in the regex.
  Are there any characters *not* allowed in the section names?  Are
section names supposed to be case-sensitive?

 > 2)  The ParsingError lists the lines it thinks are bad at the end of the
 > run, but they come up like this on the listing:
 > 
 >         [line  5]: '[Button States]\012'
 >         [line 23]: '[CallWare Archive]\012'
 > 
 > Now, is this because ConfigParser thinks it's on UNIX, because it treats
 > everything as a UNIX file, because it is not expecting the '\r\n'
 > combination, or because it's broken?

  This is because the files are opened in text mode, and CR/LF is
represented as LF internally.  This is normal.  The only error was
that the space wasn't allowed in the section name, as you pointed
out.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From nickb at earth.ox.ac.uk  Fri Jan 28 10:48:51 2000
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Fri, 28 Jan 2000 15:48:51 +0000
Subject: Parameters for Zope External Method
References: <388FFBE3.53CBA294@iki.fi>
Message-ID: <3891BA60.C242775B@earth.ox.ac.uk>

Jyrki Alakuijala wrote:

> I have succeeded in creating a Zope external method which
> causes a python statement to be evaluated for a web request.
> However, I have failed to pass parameters (other than the
> special case "self") to that external method.
>

Well you must be most of the way there!!

>
> So, where can I find instructions on how to call an external
> method with parameters? I could not find that in the
> manuals.
>

Hmmm - the manuals are a bit 'distributed' aren't they. You may find
more useful info in the HOW-TOs

>
> Should I use something else than an external method?
>

Depends what you want! You can pass parameters to External Methods and
also call Zope stuff FROM external methods (with parameter passing) so -
how?


-------------------------------
little Python method in file testext.py

def little_method(self, something):
        # any usfull stuff here which may use 'something'
        return  'ok - I did something'
-------------------------------
Inside our Zope DTML-METHOD

....
just to show our passed parameter does exist in our namespace before we
use it
<dtml-var  a_parameter>
<br>
<dtml-var "our_little_external_method(a_parameter)">
<br>
well did that work for you?
....
-------------------------------
and of course we had to create our External Method inside Zope (same
Folder?)

ID = 'our_little_external_method'
Title = 'Just testing'
Function name = 'little_method'
module = 'testext.py'
-------------------------------

That should do it - at least to first-order. My understanding is
superficial !!!

Inside our External Python Method we can also call any Zope Method via
the 'self' namespace....

def  little_method(self, some_parameter):
        # something useful here
        self.zope_method_in_our_folder(some_parameter=some_value)
        return 'whatever you like'

which will call the named  zope_method_in_our_folder and make any named
passed parameters available as <dtml-var>s inside the folder.


Its all rather cryptic but neat at the same time. Worth a bit of effort
to understand me thinks.

-but-don't-ask-me-about-aquisition-ly yours

Nick

--
Dr Nick Belshaw                                    Dept of Earth Sciences
Tel : 01865 272000                                 Oxford University
Fax : 01865 272072                                 Parks Rd, Oxford, OX2 3PR





From fredrik at pythonware.com  Wed Jan 19 13:32:25 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 19:32:25 +0100
Subject: Looking for open source hosting site
References: <B4AB6A14.1FB8%dgoodger@bigfoot.com>
Message-ID: <000e01bf62ab$89c2fc10$f29b12c2@secret.pythonware.com>

(mailed and posted)

David Goodger <dgoodger at bigfoot.com> wrote:
> I was surfing several weeks ago and came across a site that offered free
> hosting of open source projects. There was an application form, in which you
> detail your project, and (if I recall correctly) it would assign you a
> subdomain (i.e., if the site was www.site.org, your project would be
> www.project.site.org) and give you access.
> 
> Unfortunately, I didn't bookmark the site, and can't find it now. Does
> anybody know the site I'm talking about? Please e-mail, and I'll summarize.

http://sourceforge.net/

</F>




From bwarsaw at cnri.reston.va.us  Tue Jan 18 18:48:52 2000
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 18 Jan 2000 18:48:52 -0500 (EST)
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
	<YZRg4.1292$Ve7.28796@iad-read.news.verio.net>
	<862jqd$l9s@news.or.intel.com>
Message-ID: <14468.64484.891685.147425@anthem.cnri.reston.va.us>

>>>>> "t" == tye  <tye4 at yahoo.com> writes:

    t> Only way around that would be not use tabs... which is quite
    t> ridiculous (I don't want to type 4 spaces for every line.)

Using tabs in Python source is fine if you follow one very simple
rule: your tab width must be equal to the number of spaces you indent
for a block.  Thus if you indent 4 spaces per level and your tab width
is equal to 4 spaces, you can use tabs everywhere.  If you indent 4
spaces but your tab width is 8 spaces, you should not use tabs in the
indentation.

This implies that you should never mix tabs and spaces for indentation
in Python source code.  The above is the rule followed by Emacs'
Python mode (see py-smart-indentation).

-Barry



From kc5tja at garnet.armored.net  Mon Jan 24 18:50:45 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 24 Jan 2000 23:50:45 GMT
Subject: tabs do WHAT?
References: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net>
Message-ID: <slrn895gp2.rar.kc5tja@garnet.armored.net>

In article <0mPi4.3349$bp2.155298 at newsread2.prod.itd.earthlink.net>, Warren
Postma wrote:

>yellow nicotine stains. Sorry for the visual image, but hey, I gotta say, it
>just doesn't fit with the rest of the language, and I don't buy it one bit.

Then why are you still using Python?  Perl offers _all_ the features that
Python supports now, but is not white-space significant (for the most part).

>A programmer/professor I once had for a software engineering course told us
>about the "principle of least amazement": Anything weird or hard to explain
>that you write into your software will later become the thing you regret.

Precisely why Python's indentation-based scoping works so well.  It
eliminates suprises.

>As an embedded language, Python needs to be usable beyond the ivory towers
>where functional programming or object oriented programming or metaclasses
>are fodder for discussion.   Python looked eminently useable by

Python is a scripting and rapid development language.  It's not a systems
language.  Do not confuse the two.  For instance, I'm going to be using
Python to develop the filesystem (DNFS) for my operating system.  But I can
guarantee you I won't be using it for the finished product.  :)

>non-programmers until I found out that whitespace was meaningful.  Try to
>explain scope to someone who doesn't write programs for a living.

Everything you want done under the IF statement goes under the IF statement,
as if writing an outline on a piece of paper.

	IF a=b:
		do this
		then do this
		then do that
		and finally, do that.

Now try explaining brace-delimited constructs to a newbie:

Everything you want done under the IF statement needs to go inside these
things called Braces. You need these braces to group the things you want
done to treat them as a unit.  My recommendation is to write your code like
this, but you don't have to.  Just make sure, whichever style you choose to
write your code in, make it consistent so you'll know what you wrote two
months ago:

	if( a == b )
	{
	   do this;
	   then do this;
	   then do that;
	   and finally, do that.
	}

I can absolutely guarantee a new-comer to programming, computer science
major or otherwise, will have their head spinning with the latter
description.  The former is a piece of cake, because _everyone alive today_
writes outlines in almost exactly the same way, and in every way, they are
whitespace sensitive.

>Even BASIC, ugly as it is from a functionality point of view, is miles ahead

BASIC is actually quite a capable language, and a good implementation offers
a nice, clean, and consistent syntax.

>from the point of view of useability by people who are skilled technical
>people, yet who are not "programmers".    Ergo, Visual Basic's popularity.

No; VB is popular for one reason and one reason only: Microsoft.  TONS of
other companies had released RAD solutions based on all sorts of languages,
>INCLUDING BASIC<.  But none has succeeded until Visual Basic came to the
scene.  Why?  Because Microsoft sets the standards for development for their
operating system.

Had Microsoft chosen assembly language for the RAD solution, and adequately
demonstrated its features, I can guarantee you that Visual MASM would be the
RAD tool of choice.

>The language is a dog's breakfast, but at least the basic feature set you
>use every day is clear and clean enough to keep in your head, and still have
>room for your other domain-specific knowledge.

And Python isn't?  Syntactically speaking, they're almost identical.  That's
why you keep seeing "Python Vs. Visual Basic" messages from people all over.

>Unreadable Languages, )...   So close, and yet so far away. Has anyone built
>a version of Python that totally ignores tabs and whitespace, or a filter
>that you can use on top of Python? Perhaps  it could support both 'begin'

Yes -- they call it BASIC.

>and 'end', or  '{' and '}' for scoping.  Heck, at least you could write it
>in Python.  An interpreter within an interpreter. That would probably be
>very slow, wouldn't it?

No, because the interpretation is done on bytecodes.  The pre-compiler is
responsible for morphing the code to .pyc files.  All one needs to do is
re-implement the compiler.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From kc5tja at garnet.armored.net  Sun Jan  2 22:25:46 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 3 Jan 2000 03:25:46 GMT
Subject: GCC and Unicode (OT)
Message-ID: <slrn87062a.sii.kc5tja@garnet.armored.net>

Please indulge my off-topic request for a brief moment -- it's nothing at
all Python related, but I'm not having much luck with this particular problem.
Does anyone know of any way to make GCC produce 16-bit characters when
encoding a Unicode string?  That is, if I have:

	wchar_t *myText = L"ABCD";
	
I want myText to point to the following:

	0x0041 0x0042 0x0043 0x0044

instead of:

	0x00000041 0x00000042 0x00000043 0x00000044

I'm using GCC under the Linux operating system.

Thanks for any tips you folks have, and I apologize for the off-topic nature
of this message.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From skip at mojam.com  Tue Jan 25 15:50:37 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 25 Jan 2000 14:50:37 -0600 (CST)
Subject: Python compiler?
In-Reply-To: <20000125163234.C2497@stopcontact.palga.uucp>
References: <20000123092040.A2010@stopcontact.palga.uucp>
	<000401bf661e$da1c4100$902d153f@tim>
	<20000125163234.C2497@stopcontact.palga.uucp>
Message-ID: <14478.3229.692666.612735@beluga.mojam.com>

    Gerrit> But isn't it possible to write a *compiler* for Python?

Yup.  Check out Python2C at

    http://www.mudlib.org/~rassilon/p2c/

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From bayernmREMOVEME at hotmail.com  Mon Jan 10 01:18:34 2000
From: bayernmREMOVEME at hotmail.com (Ricardo Jamin)
Date: Mon, 10 Jan 2000 06:18:34 GMT
Subject: fetch a html page via proxy?
Message-ID: <387978ff.3083491@news.jps.net>

I need to fetch a page from the web but the linux machine doesn't have
direct connection to the net.  It needs to go thru a proxy to connect
to the web.  How can I do this with Python?  I know how to write
Python code that fetch the page directly, but I can't find any
information about the proxy part.  

thanks in advance.
Ricardo Jamin
bayernm<@t>hotmail.com
www.jps.net/klinsi/main/


From drmental at my-deja.com  Fri Jan 21 00:10:30 2000
From: drmental at my-deja.com (drmental at my-deja.com)
Date: Fri, 21 Jan 2000 05:10:30 GMT
Subject: help with rfc822
Message-ID: <868pns$ain$1@nnrp1.deja.com>

Please help a newbie,

I'm trying to parse an email that is received with poplib, what i need
out of it is the email address of the sender and the email body.

I've tried to store the email as a temp file and then use the
rfc822.Message(tmpfile) and that does not return a Message instance.
I've tried to use the StringIO and that give me an error.  Can someone
give me a brief example of how get the email addres out of an email that
is stored as a string?

Thanks in advance,

Mats Nordgren


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


From ame at swipnet.se  Wed Jan 19 18:16:27 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Wed, 19 Jan 2000 23:16:27 GMT
Subject: HTMLgen and Cookies
Message-ID: <388643c3.14755317@nntpserver.swip.net>

Hello!

I'm using HTMLgen to create some dynamic HTML pages and I need to use
Cookies on a couple of them.

I have used Cookie.py (by Tim O'Malley) which works very well by it
self (Thanks Tim)

My problem is: How do I use HTMLgen and Cookie.py together?

It's my understanding that the cookies must be sent to the user AFTER
'Content-type: text/html' but BEFORE <DOCTYPE ....

a short code showing how to do this would really be appreciated!!!!

// Anders



From ccosse at swampfox.fmarion.edu  Sat Jan 29 06:40:15 2000
From: ccosse at swampfox.fmarion.edu (Charles B. =?iso-8859-1?Q?Coss=E9?=)
Date: Sat, 29 Jan 2000 11:40:15 +0000
Subject: JPython & readline?
Message-ID: <3892D19F.F1D2537F@swampfox.fmarion.edu>

Could someone please advise me on enabling command history features
in JPython? I have the readline library (2.0 i think) and it works in
the
regular python interactive shell, but can't find analogous files/options

in JPython source code.  Thanks for any advice!
Charles Coss?





From tim_one at email.msn.com  Mon Jan 24 02:58:35 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jan 2000 02:58:35 -0500
Subject: tabs do WHAT?
In-Reply-To: <200001240648.WAA06269@seanb.sttln1.wa.home.com>
Message-ID: <000201bf6640$d154d100$022d153f@tim>

[seanb at home.com]
> Out of curiosity, are there any online copies of these
> usability studies that were so formative to pythone?  Such
> studies might make interesting reading and some people may
> have more trust in the results of a study they could examine
> online.
>
> Failing online access, is there any public access to write-ups
> of these studies?

Agreed it would be interesting!  I don't have any links for you -- I knew
ABC only as a user, but read several papers that came out of the project.
That was over a decade ago, though, and certainly not on the web; the
specific claim about indentation is one I don't recall reading about then,
but is one Guido has mentioned on c.l.py several times since (but he hasn't
bothered to jump into a "whitespace thread" in years -- you'd have to search
hard for that).

The ABC project is no longer active.  Its sponsoring institution still has
*some* info available (http://www.cwi/nl/).  Leo Geurts, Lambert Meertens
and Steven Pemberton were ABC's primary designers.  A quick web search on
their names turned up this citation:

    "An Alternative Simple Language and Environment for PC's"
    S. Pemberton, IEEE Software 4(1):56-64 (Jan 1987)

Presumably (a) that contains more info, and (b) is not available online.

Here's a snazzy quote from SP's homepage (http://www.cwi.nl/~steven/):

    We did requirements and task analysis, iterative design,
    and user testing.  You'd almost think programming languages
    were an interface between people and computers.

today's-generation-would-probably-find-3D-rotating-
    metallic-braces-easier-to-grasp<wink>-ly y'rs  - tim





From jdukarm at hydracen.com  Mon Jan 10 18:04:06 2000
From: jdukarm at hydracen.com (J Dukarm)
Date: Mon, 10 Jan 2000 15:04:06 -0800
Subject: Python in Corel Linux
References: <947294639.1002448618@news.islandnet.com>
Message-ID: <947546634.768344873@news.islandnet.com>

I was able to point the Corel Update utility to Debian, as someone suggested,
and download and install Python 1.5.1 with absolutely no problem.  Later, having
seen where Python was at Debian, I found it on the Corel Linux installation CD
too (Duh).

Thanks to all who responded.

Jim Dukarm
DELTA-X RESEARCH

 On Fri, 07 Jan 2000, J Dukarm wrote:
>I just installed Corel Linux ("standard" version on CD) on a PC where I
>can also boot Win98, and as a Windows refugee I was very pleased to see how
>easy it was to access my Windows file system, read floppies, etc.  Only problem
>is -- it seems that Corel left Python out of the distribution!
>
>Has anybody out there installed Python in Corel Linux, and if so are there any
>tricks to getting it set up and working?
>
>Thanks,
>
>Jim Dukarm
>DELTA-X RESEARCH


From alet at unice.fr  Thu Jan 27 16:12:32 2000
From: alet at unice.fr (Jerome ALET)
Date: Thu, 27 Jan 2000 22:12:32 +0100
Subject: ole structured storage documents under *nix
Message-ID: <3890B4C0.63834E3B@unice.fr>

Hi,

is there a python (preferably) or C free software library to read OLE
structured storage documents under *nix ?

please CC me any answer to: alet at unice.fr

thanks.

Jerome Alet


From pehrens at ligo.caltech.edu  Thu Jan  6 11:35:29 2000
From: pehrens at ligo.caltech.edu (Phil)
Date: 6 Jan 2000 16:35:29 GMT
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com>
Message-ID: <852g8h$75@gap.cco.caltech.edu>

tmick at mail.com wrote:
>I have not read _Python Essential Reference_. All I can say about it is
>that I have heard that Guido's contribution to it is minimal and that
>his name is prbably there more for name recognition. That says nothing
>about the quality of the book though.

Darn tootin' it says nothing.  David Beazley, who wrote the book,
is a crackerjack coder and writer, and would be even if GVR's name
did not appear anywhere on his book.

_Python Essential Reference_ is just what it says; a reference.
If, like me, you are an experienced (which says nothing about the
quality of my code ;^)) coder, you can use it to learn Python.

Python is a darned easy language to learn, anyway.  Dagnabbit!



From breiter at usf.Uni-Osnabrueck.DE  Wed Jan 12 07:29:23 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 12 Jan 2000 12:29:23 GMT
Subject: Python and ArcView GIS
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>
Message-ID: <85hs33$4t9$2@newsserver.rrzn.uni-hannover.de>

In article <GyPe4.14473$2x3.253287 at newscontent-01.sprint.ca>,
	"Bruce Dodson" <bruce_dodson.nospam at bigfoot.com> writes:
> If anyone is interested in this combination, I have an ArcView 3.x extension
> which embeds the Python interpreter within ArcView.  The benefits are
> obvious: access to COM objects and Python modules, plus a full-featured
> language, more suitable for large applications than ArcView's native
> scripting language.  

Sounds interesing.
Why do you need to embed python in ArcView?

Would it not be enough to just use the COM objects with python's 
win32com.client?

Thanks for sharing your work in any way, though.
I hope you will license it as free software.

	Bernhard
-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From tjg at avalongroup.net  Fri Jan 21 11:56:45 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Fri, 21 Jan 2000 08:56:45 -0800
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com>
Message-ID: <38888FCD.F0244E66@exceptionalminds.com>

Fredrik Lundh wrote:
> 
> Roy Smith <roy at popmail.med.nyu.edu> wrote:
> > When I was first learning programming (a quarter of a century ago!),
> > there was the idea of the "telephone test", i.e. your code should be
> > simple enough that if you read it to somebody on the phone they would
> > understand it.
> 
> hey, who the heck is reading code over the phone
> these days?

For a *really* good time dial 1-900-XPYTHON

For the very best in Python code, only $3.99/minute
VISA/MC/AmEx all accepted.
 
-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From thomas at xs4all.net  Thu Jan 13 07:05:14 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 13 Jan 2000 13:05:14 +0100
Subject: a small problem
In-Reply-To: <20000112165932.A2758@stopcontact.palga.uucp>; from gerrit.holl@pobox.com on Wed, Jan 12, 2000 at 04:59:32PM +0100
References: <20000112015923.A8450@xs4all.nl> <000701bf5cb4$c72320a0$da2d153f@tim> <20000112145107.B8450@xs4all.nl> <20000112165932.A2758@stopcontact.palga.uucp>
Message-ID: <20000113130514.A21652@xs4all.nl>

On Wed, Jan 12, 2000 at 04:59:32PM +0100, Gerrit Holl wrote:

> People should hang around in comp.lang.c or even comp.lang.asm.i386 if
> they care about speed.

No, they should be there if they _only_ care about speed. 

> > On the other hand, oneliners will always feel or look faster to some people.

Well, after saying that, and the dutch railways involuntarily extended my
travelling from 2 hours to 4 1/2, i tried it out, in the train, on my PII-233
laptop. The reduce() of the original problem took half again as long as the
for loop tim proposed. the map/reduce i suggested took a notch longer than
that. I wont be using lambda's and think of speed at the same time, from now
on ;)

Another amusing thing was that I wrote:

for x, in mylist:

to iter over the testlist (a list of single-item tuples.) I didn't
consciously think about it, but when i reread it, i decided it couldn't be
right. It would just be too obvious. There had to be a reason for it to be a
special case of some kind, like perl has with the magic ++ operator on
strings. Contrary to the string-++ operator in perl, though, it worked just
as expected ;)

(Maybe someone here can explain the perl weirdness to me. "string"++ in perl
works only on strings that match the pattern '^[a-zA-Z]*[0-9]*$'. A
colleague wrote some code that searched for "<prefix><number>", where prefix
only allowed [a-z] and number [0-9]. He used ++ to go from number 0 on up.
Some months ago, someone changed the prefix limit to [a-z0-9], and last week
someone actually tried to use prefix 'pop2h'. And 'pop2h001'++ yielded 1,
breaking numerous things in mysterious ways. But, _why_ is the ++ operator
limited in such an artificial way ? The local perl gurus refuse to answer me
;)

<snip>
> How would I do that without lambda's, if this code is in a method? Should
> I nest two functions?

For a straight translation without lambda's you can _always_ use functions.
nested, if you wish, but not necessarily. lambda's work _exactly_ like
nested functions, only you cant use statements in them, only expressions.

But doing it without lambda is not the point -- if lambda wasn't there,
there would possibly be a more convenient way to do such things, or even
more speedy (like a module of functions, like operator, for use with map.
basic things like adding the first element of each argument. Or something.)

<snip>
> lambda's are good, because otherwise, it wouldn't be possible to write
> obfuscated code in Python... Can someone please rewrite the one-liner

You're missing the point. Guido does not believe obfuscated code is good.
Why would Python need obfuscated code ? It goes through great pains(*), in
fact, to make obfuscated code very hard.

(*)Admittedly mostly other peoples' pains, the ones who dislike forcefed
indentation, for instance :)

> > Hi! I'm a .signature virus! copy me into your .signature file to help me spread!
> Oh, help! My computer is infected! What should I do now?!

The usual cure is to install a less gullible mind, or indeed, any at all.
But that is what the signature tries to do, so perhaps you can consider
yourself cured ;)

Conspirationally-y'rs, 
-- 
Thomas Wouters <thomas at xs4all.net>

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



From nemeth at iqsoft.hu  Mon Jan 10 16:03:38 2000
From: nemeth at iqsoft.hu (Nemeth Miklos)
Date: Mon, 10 Jan 2000 22:03:38 +0100
Subject: Python is much faster than Java?
Message-ID: <387A492A.CBB58B86@iqsoft.hu>

A colleague implemented a sorting algorithm in both Java and Python
(CPython on NT) just to test the capabilities of Python.
He is (or may be only was) a Java fan, but I spoke to him about Python
so much, that he decided to make some comparisons.
After the tests it turned out that the Python implementation was 5 times
faster?
We could not believe our eyes! How is it possible?
A knew that Java is very poor in managing object (creating them and
garbage collecting them), but I thought it is fast enough for procedural
programming.

NM




From ivanlan at callware.com  Mon Jan 31 23:33:28 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Mon, 31 Jan 2000 21:33:28 -0700
Subject: This thread was almost about Python once
References: <z8kl4.783$S3.31411@newsfeed.avtel.net> <001401bf6c16$d5657660$3acbd9c2@peridot.optichrome.com> <20000131201258.G19725@xs4all.nl> <sbsl4.794$S3.31761@newsfeed.avtel.net>
Message-ID: <38966218.5048159C@callware.com>

Hi All--

Clarence Gardner wrote:
> 
> Thomas Wouters <thomas at xs4all.net> wrote:
> : *) on Linux systems, passing an open fd is actually fairly easy and fairly
> : compatible; old (or perhaps just all) interpreters could be started with
> : '/proc/fd/3' as the script. There was talk about this on the linux kernel
> : list a year or so ago (and probably more often before and after that) but
> : the problem was, and probably still is, that it would break interpreters
> : that rely on opening a filedescriptor right after starting and assuming it
> : will be fd 3.
> 
> What interpreter is so braindead as to assume that particular file
> descriptors are available when it is invoked?  And even if there are
> any, catering to them is worse than all of the Microsoft backward-
> compatibility-to-1985 hacks put together, IMNSHO.
> 

stdin = fd 0
stdout = fd 1
stderr = fd2
next fd you ask for = fd3

Not braindead at all.  Just a nice tradition, which is observed on
unices & their descendents, but not on M$.

<oh-poop-where-did-fd-1-go?>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From tim_one at email.msn.com  Wed Jan 26 01:08:05 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 26 Jan 2000 01:08:05 -0500
Subject: Module to access file attributes for Windows?
In-Reply-To: <388E84C1.703E3DEE@es.co.nz>
Message-ID: <000401bf67c3$b5d85300$592d153f@tim>

[Matthew Miller]
> Is their a module to access file attributes like:- date
> created/modified,etc for Windows? I've had a reaonable look
> around but with no success.

See the docs for os.stat, os.path.getatime, os.path.getmtime,
os.path.getsize, etc.  These are (mostly) platform-independent.  To get at
Windows-specific stuff (basically the whole Win32 API, plus everything else
<wink>), you need to get the PythonWin extensions.





From jbauer at rubic.com  Tue Jan  4 17:19:37 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 04 Jan 2000 16:19:37 -0600
Subject: Python & Apache
References: <A4uc4.411$hJ1.3104@news.rdc1.tx.home.com>
Message-ID: <387271F9.83A2261E@rubic.com>

Paul,

I've documented how to get Apache and Python
working on NT.

  http://starship.python.net/crew/jbauer/apachenotes/

You may also want to check out Zope, which runs
fine on Win32 platforms:

  http://www.zope.org

Jeff Bauer
Rubicon Research


"Paul E. Bible" wrote:
> Has anyone had any success getting Python CGI 
> scripts to run under the following configuration?
> 
> OS:                Windows/NT
> Web Server:   Apache 1.3.9
> Python:           v1.5.2



From oded_sha at netvision.net.il  Sat Jan 29 15:12:09 2000
From: oded_sha at netvision.net.il (Oded Shaham)
Date: Sat, 29 Jan 2000 22:12:09 +0200
Subject: Tkinter and threads
Message-ID: <38934999.932BE3FF@netvision.net.il>

Hi,
I want to use a Tk GUI for display, that is, have other threads run and
display various results using a Tk window. the problem is that whenever
Tk starts its mainloop, either on the main thread or on a seperate
thread, all other threads are immediately stopped (or to be more
precise, suspended until the mainloop exits). As far as I can see, the
only way to make things happen, once Tk is up, is through actions of Tk
components (buttons, for example).

Any ideas of how to solve this ? (will be most welcome !)

Thanks
Oded



From tim_one at email.msn.com  Sun Jan  2 02:34:48 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 2 Jan 2000 02:34:48 -0500
Subject: Sorting large files
In-Reply-To: <14433.8282.321495.794332@buffalo.fnal.gov>
Message-ID: <000d01bf54f3$d950cac0$1f2d153f@tim>

[Edward C. Jones]
> A Python program of mine generates a large number of tuples
> of ten integers each. I need to sort this collection of tuples.
>
> If there weren't so many tuples, I could make a Python list
> of these tuples and use sort(). Currently I write the
> tuples out to a file in ASCII and use the UNIX sort command.
> Is there a faster way to sort these tuples? ...

Can only echo previous requests for more info (how many is "large", is there
anything special about the distribution of ints, or the order in which they
appear, etc).

[Charles G Waldman]
> The list.sort method in Python uses quicksort (AFAIR) so
> it's just as efficient as using UNIX sort.

The 1.5.2 list.sort() is an optimized hybrid of binary insertion sort and
samplesort.  It does fewer compares than quicksort, and is much more robust
against pathological input (one data-dependent reason a Unix quicksort may
be unreasonably slow).  OTOH, the Unix sort knows it's comparing strings,
while list.sort() must redispatch to the type-approriate comparison routine
on each compare (higher overhead per comparison).  Also possible that the
Unix sort does some form of radix sort (which may-- or may not --be much
quicker than a quicksort, depending on a billion things we haven't been told
...).

no-info-in-no-info-out!-ly y'rs  - tim





From apederse at siving.hia.no  Sun Jan 16 15:11:01 2000
From: apederse at siving.hia.no (Asle Pedersen)
Date: Sun, 16 Jan 2000 21:11:01 +0100
Subject: IMAP - fetch using imaplib
References: <85ssc4$8jk$1@snipp.uninett.no> <20000116141223.C26923@quark.emich.edu>
Message-ID: <85t8jn$ah4$1@snipp.uninett.no>

hmmm!
for some strange reason, runing my code once again just now it was
working!!. I have no good explanation for this. (although it could be due to
some problems that we had in our mail-server recently). Anyway here is the
code that IS working:

M = imaplib.IMAP4('xx.xxx.xx.xxx')
M.LOGIN('xxxx','xxx')
print M.SELECT('INBOX/xxxx')
typ, data = M.FETCH(1, '(RFC822.SIZE BODY[HEADER.FIELDS (FROM SUBJECT)])')
print data[0][1]
M.LOGOUT()

(sorry to have bothered the group with this non-exisisting????
problem........)

regards,
Asle

Jeff <jam at quark.emich.edu> wrote in message
news:20000116141223.C26923 at quark.emich.edu...
> On Sun, Jan 16, 2000 at 05:42:06PM +0100, Asle Pedersen wrote:
> > I have adressed this question to the group previoussly without any
success
> > so I try once more :
> > I need help with the following IMAP command:
> > DF56 FETCH 1:3 (RFC822.SIZE BODY[HEADER.FIELDS (FROM SUBJECT)]).
> > I have tried using the imaplib for other fetch commands without trouble,
but
> > this one I can't seem to get working utilising the imaplib.
> >
> > -Asle Pedersen
> >
>
> if you could perhaps post code and/or the specific error messages you are
> seeing, it may help to diagnose the problem.
>
> regards,
> J
> --
> || visit gfd <http://quark.emich.edu/>
> || psa member -- <http://www.python.org/psa/>
> || New Image Systems & Services, Inc. <http://www.newimage.com/>
>




From lobozc at my-deja.com  Sun Jan  2 02:53:28 2000
From: lobozc at my-deja.com (lobozc at my-deja.com)
Date: Sun, 02 Jan 2000 07:53:28 GMT
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg>
Message-ID: <84mvaa$8mp$1@nnrp1.deja.com>

1. I fully support a thesis that Python is _not_ a very high level
language. There are many ways in which the level of the langauge can be
defined. Halstead's for example - but not only. Looking from the point
of view of many popular languages [C++, Java, JavaScript, Pascal, Perl
etc.] it is obvious that Python shares most of the (primitive)
mechanisms and approach of such languages.

2. Compared to higher level languages, like ICON, Mathematica etc. it
is obvious that some higher level concepts, like goal-directed
evaluation, coexpressions and many others are missing. Most of the mind-
share of improving Python seems to be going towards clarifying it's
type system and making it more OO. This direction is not bad in itself,
it's just that it hits the law of diminishing returns: makes Python
more of a me-too language in that space.

3. Unlike Java [and a couple of others] Python has a _potential_ to
became much higher level language than its competition. Functions as
first class objects, list/tuple handling by Python as well as lambda
definitions open a way to move Python quickly to another level. One
could write a lot here about it, but enough to say this: in a high
level language Haskell the quicksort algorithm takes one short (and
readable!) line - and it is mathematically provable that it does what
it should. In other words: using higher level concepts than looping and
jumping [a staple of contemporary languages] we can express much more
in much less code - as well as making it safer and less error prone.

4. Some work [by Greg Ewing? apologies if the name is spelled wrongly]
aims to extend Python with list comprehension mechanisms, thus allowing
big jump in expressivity of Python. From language level point of view
this is definitely the most important work. [Still, it seems to be that
this direction is also very underappreciated by many Pythoners.]

5. It must be noted that I'm not trying here to talk everybody into
using Haskell, as 99.9% of programmers will never accept functional
languages. But Python has an opportunity to acquire some of the
powerful mechanisms well tested in functional languages - and stay the
traditional, imperative programming language. Thus allowing
easy 'natural' startup for entry-level (VB-trained :-)) programmers but
enabling also some powerful coding by professionals.

6. I'm not sure I understand the relationship of the XML/DTD proposal
to the 'language level'. It seems to me that it may make for 'better
ide' but not necessarily for a better language. It's one thing to have
some additional notation for comments, it's another to have a language
which needs less coding, therefore less comments. I'm not talking here
about APL-like conciseness (unreadable two days later...), more about
Haskell-like expressiveness - originating from higher level abstraction
mechanisms of formal computing.

[For more information on details I can recommend the book by Simon
Thompson Haskell: the craft of functional programming; as well as Ralph
Griswold's The Icon programming language and books by R. Meader about
the Mathematica programming language].

In article <84i4bn$rqh$1 at clematis.singnet.com.sg>,
  "Ajith Prasad" <aprasad at magix.com.sg> wrote:
> http://www.oreilly.com/news/vhll_1299.html is an article by Greg
Wilson
> casting doubts on the effectiveness/value of Python and other very
high
> level scripting languages. Wilson comments that:
> "Over the past few years, I have done several medium-sized projects
using
> both Perl and Python. At first, I was very excited by what these Very-
High
> Level Languages (VHLLs) let me do, and how quickly. The more I played
with
> them, however, the less satisfied I was. In particular, I no longer
believe
> that they deserve the 'V' in their name. This article explores why,
and
> suggests some ways in which they could evolve." Worth responding to
as it
> includes detailed criticisms of Python in particular.
>
>


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


From pinard at iro.umontreal.ca  Mon Jan 17 13:56:04 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Jan 2000 13:56:04 -0500
Subject: Proposal: i-String literals in Future Version of Python?
In-Reply-To: pf@artcom-gmbh.de's message of "Mon, 17 Jan 2000 11:49:35 +0100 (MET)"
References: <m12A9j5-000CxTC@artcom0.artcom-gmbh.de>
Message-ID: <oqya9o4jej.fsf@titan.progiciels-bpi.ca>

pf at artcom-gmbh.de (Peter Funk) writes:

> So that some magic in the python interpreter expands
> 	print i"This is just a message"
> into a call to a gettext function internally, if such a function is
> available.

That might be nice, but it would not be sufficient, as the textual
domain for the translation is not specified.  It should be in some way.
The textual domain cannot be specified once for a whole program, as a
program can be made up of libraries pertaining to different translation
domains, and execution control might fly like hell between libraries :-).
If i"Text" was to call gettext automatically, the Python interpreter would
also need to know, on a per file basis, which textual domain to use.

Our best bet, right now, is sticking to _("text") for immediate translation,
and to ''"text" for deferred translation.  The value of `_' should be a
bound function into an instance of a Translation class, in which the textual
domain has been frozen (as I sketched in a recent message).  Each module
or file would initialise its global `_' by calling a convenience function
accepting a textual domain name, and retrieving the bound function for
the proper textual domain name, creating on the fly a new instance of the
Translation class, as necessary.

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





From shouldbe at message.com  Thu Jan 20 15:22:45 2000
From: shouldbe at message.com (Arinte)
Date: Thu, 20 Jan 2000 20:22:45 GMT
Subject: Now it is getting wacky was( Re: A modest Error (I mean Proposal))
References: <XFMail.000120170754.mikael@isy.liu.se>
Message-ID: <p8Kh4.2347$L4.74646@newsread2.prod.itd.earthlink.net>

I change my code to look like this

 def IOCTL(self, command, dargs):
  command1 = long(0)
  if type(command)==type(2):
   poss.write("IOCTL was int")
   command1 = long(command)
   poss.write("IOCTL Set to Long");
  else:
   if type(command) == type("  "):
    poss.write("IOCTL was a string")
    command1 = command
   else:
    if type(command)!=type(2L):
     poss.write("IOCTL fail")
     return -1
  self.dump();
  poss.IOCtl(self.devname,command1,dargs) #make this pass the ioctl command
as an PossArg with a special name
  print len("1234567")  #ADDED THIS LINE
  return 0

Just to test len.  And now my program works???? If I remove this line or
even just remove the print it doesn't work???
Now what ??? If anything.  Another not here that critical print statement
doesn't print the len of the string either...



"Mikael Olofsson" <mikael at isy.liu.se> wrote in message
news:XFMail.000120170754.mikael at isy.liu.se...

On 20-Jan-00 Arinte wrote:
 >
 >  "Mikael Olofsson" <mikael at isy.liu.se> wrote in message
 >  news:XFMail.000120142944.mikael at isy.liu.se...
 >
 >  On 20-Jan-00 Arinte wrote:
 >   >  I am getting this weird error???
 >   >  Traceback (innermost last):
 >   >    File "Beer", line 17, in ?
 >   >    File "D:\POSSApps\anotherposs app\PossDevice.py", line 37, in
IOCTL
 >   >      if type(command)==type(2):
 >   >  TypeError: len() of unsized object
 >
 >  Maybe you have accidentally redefined type to len...
 >
 >  I would entertain that thought, but this problem only occurs on the 2nd
 >  call to this function???

OK, so we've narrowed it down some. Maybe that's what happens between
those two calls.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    17:03:57

This message was sent by XF-Mail.
-----------------------------------------------------------------------




From gerrit.holl at pobox.com  Fri Jan 21 08:16:03 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 14:16:03 +0100
Subject: Suggest: .get() method on lists.
In-Reply-To: <p6qvh4oc1q4.fsf@informatik.hu-berlin.de>; from loewis@informatik.hu-berlin.de on Thu, Jan 20, 2000 at 02:27:31PM +0100
References: <20000118212229.A542@stopcontact.palga.uucp> <p6qvh4oc1q4.fsf@informatik.hu-berlin.de>
Message-ID: <20000121141603.B2701@stopcontact.palga.uucp>

Martin von Loewis wrote on 948374851:
> file = sys.stdin
> if len(sys.argv) > 1:
>     file = open(sys.argv[1])
> 
> right now, to open the file. I can't see how you could get the
> equivalent of that code with .get.

f=sys.argv.get(1) # default = None
if not f: f = sys.stdin

But what about a sys.argv like this (untested)
class argv(UserList.UserList):
    def has_key(self, num):
        return len(self.data) > num:

    def get(self, num, default=None):
        if self.has_key(num): return self.data[num]
        else: return default

Wouldn't this be a good idea? I see too much code like above.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From paul at prescod.net  Fri Jan  7 13:35:43 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 07 Jan 2000 10:35:43 -0800
Subject: XML to dictionary?
References: <FGALACFJFLICBBAA@my-deja.com> <01b901bf58b8$fadac010$0100a8c0@rochester.rr.com>
Message-ID: <387631FF.5B10CDB8@prescod.net>

> From: "John Jensen" <johnjensen at my-deja.com>
> I'm thinking that the best thing to do in this situation is just write a
> quick customized script, and not worry about using "fancy" libraries like
> xmllib, pyxie, etc.

I don't see what is "fancy" about xmllib and I can't imagine that
writing your own parser (is that what you mean by "quick customized
script") would be easier than using one that already exists.

For your task I think that you just have to override a couple of
methods. Writing a parser will be more work and it will be more fragile.

> From: "Darrell" <darrell at dorb.com>
> Here's my own hack at tag parsing http://www.dorb.com/darrell/xmlP/
> With wxPython it gives you a tree view of your tags. Replace the branches
> with dictionaries and away you go.
> 
> --Darrell

I'm a little surprised to see this comment in your code:

"""Convert this function to 'C' for real speed."""

Too masochistic to write it yourself only once?? :)

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
When I'm gone, boxing will be nothing again.  The fans with the cigars
and the hats turned down'll be there, but no more housewives and
little men in the street and foreign presidents.  .... I was the
onliest boxer in history people asked questions like a senator.  
            -- Muhammad Ali





From robin at jessikat.demon.co.uk  Wed Jan  5 12:55:19 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 5 Jan 2000 17:55:19 +0000
Subject: How can I use tcl/tk 8.2.2 in Python
References: <82qf4h$goo$1@news2.nctu.edu.tw>
 <38736F93.24A867B7@gssec.bt.co.uk>
Message-ID: <V96YSLAHW4c4Ew4+@jessikat.demon.co.uk>

In article <38736F93.24A867B7 at gssec.bt.co.uk>, Alan Gauld
<alan.gauld at gssec.bt.co.uk> writes
>> I try to use Tkinter with tcl/tk 8.1~8.2.
>> And I get an error that tell me It can not find tcl80.dll .....
>
>I don't really understand why this happens but I suspect you 
>have some Tcl/Tk DLLs in your Windows\System32 directory.
>
>Delete(or rename if your nervous) those DLLs and the error 
>should go away.... Its worked (several times) for me! :-)
>
>Alan G.
>

I think base Python 1.5.2 assumed tcl/tk 8.0.x and there fore needs
tcl80; you have to patch _tkinter.c and rebuild it to make it work with
8.1/8.2.
-- 
Robin Becker


From ullrich at math.okstate.edu  Sat Jan 29 10:40:39 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Sat, 29 Jan 2000 09:40:39 -0600
Subject: Dynamic class construction?
Message-ID: <389309F7.4A35371F@math.okstate.edu>

    Say I want to make a function that returns a new _class_.
I can do this:

def NewClass(whatever):
    class new(baseClass):
        pass
    new.SomeAttribute = whatever.whatever
    return new

and it seems to do the job. Is there another way to
do this that seems much better for some reason?

    I guess a related question would be whether
having distinct classes with the same __name__
is going to be trouble - does __name__ get used
in significant ways internally or is it just there
for the programmer's convenience?

DU

    (In case someone feels the answer might depend
on the context: it's mathematical structures. For
example baseClass might be Group; a subclass
of Group "is" a group, and an instance of the
subclass "is" an element of the group. So I might
want to create a group using

def NewGroup(groupOp):
    class new(Group):
        pass
    new.__mul__ = groupOp
    return new

so then

G = NewGroup(SomeGroupOp)
g1 = G(someparams)

makes g1 an element of the group G.)



From aa8vb at yahoo.com  Tue Jan 11 14:43:44 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 11 Jan 2000 14:43:44 -0500
Subject: [ANN] Multimethod.py -- multimethods for Python
In-Reply-To: <87zouclb8m.fsf@baal.winnegan.de>
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov> <20000111114923.B42270@vislab.epa.gov> <87zouclb8m.fsf@baal.winnegan.de>
Message-ID: <20000111144343.A44514@vislab.epa.gov>

Siggy Brentrup:
 |Can you point us to a formal definition? Seems like you initially got into
 |trouble explaining it by example :)

I don't know where to find a formal definition.  Of those definitions I've
seen, I like this one best:

 http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=464888387:
 |
 | Multimethods support runtime polymorphism on any number of arguments.

The sentence from the paper I posted a link to says the same with more verbage:

 | In object-oriented languages with multimethods (such as Common Lisp,
 | Dylan, and Cecil), the appropriate method to invoke for a message send
 | can depend on the run-time class of any subset of the message arguments,
 | rather than a distinguished receiver argument.



Aahz Maruch:
 |IOW, similar to C++-style multiple dispatch?

No, not quite.

 http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=464888387:
 >
 > The difference between the C++ dispatch on multiple arguments and
 > languages that support 'multiple dispatch' (like Dylan) is that the
 > latter languages dispatch on the run time type of the object, whilst
 > the former dispatches on the compile time type of the object (except
 > for the object the method is attached to).

--
Randall Hopper
aa8vb at yahoo.com



From python-list at teleo.net  Mon Jan  3 11:27:50 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Mon, 3 Jan 2000 08:27:50 -0800
Subject: Python not a Very High-Level Language?
In-Reply-To: <84plci$ojt$1@news.fsr.net>
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl> <84plci$ojt$1@news.fsr.net>
Message-ID: <00010308464602.04465@quadra.teleo.net>

[Mark 'Kamikaze' Hughes, on Mon, 03 Jan 2000]

::   Someone else mentioned a folding editor - the old GFA Basic (it was a
:: structured variant with editor-ENFORCED indentation-as-control but also
:: keywords) editor/IDE (originally for the Atari ST, but it was ported to
:: PCs and Amiga) had function folding that operated on a single keypress
:: in that function...  This made reading and studying big programs MUCH
:: easier, because you could see the overview of all functions available,
:: then look inside any one you wanted quickly.  There have been
:: threatening noises in the function folding direction from vim lately,
:: too.


For Linux-based Python development teams willing to consider a
commercial product with strong project management features
($150/commercial seat, $50 for private use) there's C-Forge, a nicely
done IDE with support for 20 languages, including Python. It has a
folding editor and class browser. There's a 30-day trial version at:

http://www.codeforge.com



From R.Brodie at rl.ac.uk  Fri Jan  7 12:24:52 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Fri, 7 Jan 2000 17:24:52 -0000
Subject: UDPserver bug / documentation error?
Message-ID: <8557h8$1536@newton.cc.rl.ac.uk>

I was attempting to kludge together a quick UDP server program in Python (1.5.2).
In the handler, h.request is documented as a string containg the received data
but it apparently is a tuple of the data and the socket object.

Any thoughts? My guess is at some point someone decided it might be handy to
have the socket object available but didn't fix the documentation.


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
                            ????????





From PClaerhout at CREO.BE  Thu Jan 20 03:36:28 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 20 Jan 2000 09:36:28 +0100
Subject: Python -> EXE?
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA706D@MSGEURO1>

Make a copy of the pywintypes15.dll file, and rename it to
pywintypes.dll and then it should work just fine.

Pieter

-----Original Message-----
From: tiddlerdeja at my-deja.com [mailto:tiddlerdeja at my-deja.com]
Sent: Wednesday, January 19, 2000 8:14 PM
To: python-list at python.org
Subject: Re: Python -> EXE?


I tried to use the Standalone.py from this distribution, but I got an
error.

 found OLEAUT32.dll
 found python15.dll
 added filter <TypeFilter {'d': 1}>
Creating createFruitCategories.exe
  support\Run.exe -> createFruitCategories.exe
  C:\TEMP\~535-1 -> createFruitCategories.exe
Gathering components of dist_createFruitCategories
 added filter <TypeFilter {'d': 1}>
Creating dist_createFruitCategories
Traceback (innermost last):
  File "Builder.py", line 502, in ?
    main(opts, args)
  File "Builder.py", line 484, in main
    target.build()
  File "Builder.py", line 87, in build
    self.assemble()
  File "Builder.py", line 234, in assemble
    shutil.copy2(path, self.name)
  File "d:\Program Files\Python\Lib\shutil.py", line 63, in copy2
    copyfile(src, dst)
  File "d:\Program Files\Python\Lib\shutil.py", line 17, in copyfile
    fsrc = open(src, 'rb')
IOError: [Errno 2] No such file or directory: 'c:\\winnt\\system32
\\pywintypes.dll'



I have pywintypes15.dll. Could someone tell me what I'm doing wrong?

Thanks.


In article <002801bf60fa$80b4b440$f29b12c2 at secret.pythonware.com>,
  "Fredrik Lundh" <fredrik at pythonware.com> wrote:
> Christian Schaller <Christian.Schaller at mchp.siemens.de> wrote:
> >   Is there a tool similar to Perl2Exe that transforms a Python
program
> > to a MS .EXE programc that can be executed without installing
Python =
> on
> > a machine?
>
>     http://www.python.org/doc/FAQ.html#4.28
>     "How can I create a stand-alone binary from a Python script?"
>
> </F>
>
> <!-- (the eff-bot guide to) the standard python library:
> http://www.pythonware.com/people/fredrik/librarybook.htm
> &mdash; see it live on the python conference! -->
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.
-- 
http://www.python.org/mailman/listinfo/python-list



From zawrotny at triangle.sb.fsu.edu  Mon Jan 31 09:11:09 2000
From: zawrotny at triangle.sb.fsu.edu (Michael Zawrotny)
Date: 31 Jan 2000 14:11:09 GMT
Subject: Choosing a language
References: <5650A1190E4FD111BC7E0000F8034D26A0F2A3@huina.oceanic.com>
Message-ID: <slrn89b60g.kp.zawrotny@triangle.sb.fsu.edu>

On Sun, 30 Jan 2000 08:48:02 -1000, Doug Stanfield <DOUGS at oceanic.com> wrote:
> Python is an excellent first language (or second or fourteenth for that
> matter).

[snip]

I agree entirely.  I came to Python after about ten other languages
and find it to be the most generally useful.  As much as possible,
I write everything in Python now and find that it does encourage
good practices (note, encourage, not force), and the code is still
readable a month or two later when the original problem is no longer
fresh in my head.  This despite my being too lazy to adequately
document the code in the first place.  The programs are mostly short
utilities, so the burden of reading the whole thing to figure it out
is small, but I should (and do know better), but too much to do, so
little time.  For a suggestion on how to avoid this trap, see below.

> I find the greatest strength of Python to be the interactive interpreter.
> In essence it is a sandbox to play in to try things out.  There is no other
> way to learn to program than to write programs, but writing them can be
> frustrating if you can't get into the middle of the execution and find out
> what those commands you wrote are really doing.  The interpreter is the tool
> for that.  Almost without fail if I can't make a program work I'll try a
> sequence of commands in the interpreter that show me the problem.  While
> you're learning to program, this instant feedback can make all the
> difference in speeding you to comprehension.

[snip]

Heartily seconded.  I also think the interpreter is a great tool.

Patrick Cooper [mailto:future at dotstar.net]
> > I want to start learning computer programming and from what I 
> > have found so
> > far I have chosen to start with Python.  Before I go in 
> > any more depth
> > does anyone think I should start with a different language?  I want to
> > eventually learn C, but think Python is the best place to start.

As others have noted, when you start to learn C (or some other
language), you will probably feel the pain of it not being Python.
This is the flip side to those of us who suffered first then found
Python and the fun that it brings back to programming.

My advice is that after you get comfortable with Python and start in
on C or one of its derivatives (C++, Java, etc.), pick up and read
Kernighan and Ritchie's "The Practice of Programming".  It is IMO
one of, if not the best, general programming books I've ever read.
There is multiple language coverage of nearly the entire process
program development.  Others may disagree, but I think it is a great
book.


Mike

-- 
Michael Zawrotny
411 Molecular Biophysics Building
Florida State University		| email:  zawrotny at sb.fsu.edu
Tallahassee, FL 32306-4380		| phone:  (850) 644-0069


From lutz at rmi.net  Tue Jan 18 22:56:51 2000
From: lutz at rmi.net (Mark Lutz)
Date: Tue, 18 Jan 2000 20:56:51 -0700 (MST)
Subject: ANN: Python training in Boulder
Message-ID: <200001190356.UAA25652@shell.rmi.net>

I'm pleased to announce an upcoming Python training session, to 
be held in Boulder, Colorado, February 15-17.  This class is open
to individuals, and will be filled on a first-come first-served basis.
Come learn about an exciting object-oriented scripting language, and
enjoy the beautiful Rocky Mountain region during your stay.  More 
information can be found at this page:

    http://rmi.net/~lutz/boulder-classes.html

I am the instructor for this class, but all administrative support
is being provided by Tom Christiansen's training organization.
A second session is tentatively set for May.

--Mark Lutz

----
PS: Apologies if this appears too commercial to some, but I have
gotten numerous requests for such a class in the past, and a 
general post is more efficient than wading through the 12,000+ 
messages sitting in my email box.



From cgw at fnal.gov  Fri Jan 14 13:42:08 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: Fri, 14 Jan 2000 12:42:08 -0600 (CST)
Subject: Stubbing function
In-Reply-To: <387EFCA3.94C67567@exceptionalminds.com>
References: <387EFCA3.94C67567@exceptionalminds.com>
Message-ID: <14463.28160.874963.826550@buffalo.fnal.gov>

Timothy Grant writes:
 > Is there a "standard" method of stubbing functions in Tkinter?
 > 
 > I would like to be able to use the "command=lambda:????" construct to
 > stub my functions as I don't like creating functions with empty bodies
 > just for stubbing purposes. I tried "command=lambda: pass" and that
 > generated syntax errors, so I know that I'm missing something basic
 > about how lambdas work, I just don't quite know what!


Note that 

     lambda: None

 works just fine.

lambdas may not contain any statements, just functions and
expressions.  For instance:

     lambda x: print x

is illegal, wheras

     lambda x: sys.stdout.write("%s\n"%x)

is allowed.

Hope this helps,

		-cgw




From wclodius at lanl.gov  Fri Jan 21 22:22:08 2000
From: wclodius at lanl.gov (William B. Clodius)
Date: Fri, 21 Jan 2000 20:22:08 -0700
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net> <8631bt$1nu@news.or.intel.com>  <slrn88d63f.13f.wtanksle@hawking.armored.net> <slrn88fs4o.2d0.warlock@raven.myth>
Message-ID: <38892254.A4D4E552@lanl.gov>


Jim Richardson wrote:
> 
> On 20 Jan 2000 04:58:00 GMT,
>  William Tanksley, in the persona of <wtanksle at hawking.armored.net>,
>  brought forth the following words...:
> 
> >On Tue, 18 Jan 2000 16:41:06 -0800, tye4 wrote:
> >
> >>I have offered you many reasons why we need 'end' block support..
> >
> >No, you haven't.  All you've done is claim that other languages don't use
> >indentation.  So what?  Python uses assignment.  ML doesn't.  Does that
> >mean ML is bad and Python is good, or is it the other way around?
> 
> what does ML do then? this has me puzzled...
> <snip>

As a pure functional language every entity in ML has a single definition
of its value that cannot be changed. A language with assignment allows
entities whose value can change, e.g., A in
...
A=1
X=A
A=2*X
...
A pure functional language would not allow the redefinition of A. Intead
you would have to do something similar to
...
A=1
X=A
B=2
...
and any subsequent point in the first text where you would refer to A
would have to be rewritten to refer to B, unless assignment is again
applied to A, in which case new entities  would have to be defined. In
most cases the fact that there is a single definition of an entity makes
the code easier to follow, but it potentially increase the number of
"definitions" within a scope, (such languages reduce this problem by
having additional flexibilities that encourage short functions), it
makes the compiler work harder to generate reasonable code (e.g., it
should recognize that A is not needed after the definition of X so its
"storage" can be reused for B, it cannot rely on you to do this
optimization for it), and there are some things that we think of as
changing, e.g., life, for which an imperative model with assignment is
perhaps more natural than a functional model. 

The need for an optimizing compiler for reasonable efficiency often
brings up the loudest arguments against functional languages, but,
except perhaps for arrays, the optimizations are the obvious ones you
would find in a not highly advanced compiler for an imperative staticaly
typed language. Indeed some optimizing compilers translate imperative
language into a form with single assignment (i.e., a form with
functional semantics) and then perform their optimizations on that as
the reasoning is usually simpler in that form.


From hanche at math.ntnu.no  Sat Jan 15 00:35:48 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 15 Jan 2000 00:35:48 -0500
Subject: Nested while-s
References: <8EBBEE15Ejoshtinamecom@209.208.0.7>
Message-ID: <pcopuv36gnf.fsf@math.ntnu.no>

+ josht at iname.com (Josh Tompkins):

| Is it not possible to nest while's?

It sure is.

| Consider this code:
| 
| def listOfRadius(radius, xpos, ypos):
| 	list = []
| 	currentX = (xpos - radius)
| 	currentY = (ypos - radius)
| 	while currentX <= (xpos + radius):
| 	    while currentY <= (ypos + radius):
| 		    str1 = `currentX` + " , " + `currentY`
| 		    list.append(str1)
| 		    currentY = currentY + 1
| 	    currentX = currentX + 1
| 	return list
| 
| It will list only the values from the nested while.

Move the initialization of currentY into the top of the outer while
loop.

-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From News at Titanic.co.uk  Sat Jan 15 09:32:04 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Sat, 15 Jan 2000 14:32:04 GMT
Subject: Database to Web
References: <jmsu7sk17q2hg0dbrf3qi8lobednf72m4a@4ax.com> <ejuu7s4r94q6a3scu6qm52grn5tq8h8k73@4ax.com>
Message-ID: <27118sssqtranhf5m9m4vulctblmdt8e90@4ax.com>

This % business is smart isn't it!

On Fri, 14 Jan 2000 19:35:30 GMT, Simon Faulkner <News at Titanic.co.uk>
wrote:

>What I needed was:
>
>Name = tuple[record]
> for Field in Name
>	print "<td>" + Field + "</td>"
>
>
>Why is Python so easy?
>
>Simon
>Simon Faulkner

Simon Faulkner


From pbible at home.com  Mon Jan 17 19:21:32 2000
From: pbible at home.com (Paul E. Bible)
Date: Tue, 18 Jan 2000 00:21:32 GMT
Subject: Python and MS SQL (Microsoft SQL server)
References: <38835B8F.CA022D69@psychosis.com>
Message-ID: <gmOg4.837$iL4.7073@news.rdc1.tx.home.com>

You can use mxODBC and access MS SQL via ODBC.  See the following link:
http://starship.python.net/crew/lemburg/mxODBC.html

Good Luck,
Paul

"Dave Cinege" <dcinege at psychosis.com> wrote in message
news:38835B8F.CA022D69 at psychosis.com...
> Is there a python module available to access MS SQL (Microsoft SQL server)
> My client really wants to use this over mySQL. (Which I know there is a
module
> for)
>
> Dave
>




From josht at iname.com  Fri Jan 14 21:19:40 2000
From: josht at iname.com (Josh Tompkins)
Date: Sat, 15 Jan 2000 02:19:40 GMT
Subject: Python Pre-processor (like php/asp)
References: <6D8A17398E28D3119F860090274DD7DB4B3DD7@pces.cadlab.it>
Message-ID: <8EBBD05FBjoshtinamecom@209.208.0.7>

Ummm... I'm not sure how useful this is, but there was an article in a recent 
Dr. Dobbs Journal (January, I think) about something called Python Server Pages 
(PSP).  It is supposed to be mostly compatable with ASP and is written in 100% 
python (for now, at least).

-- 
Josh "Josue" Tompkins
_______________________________________________________________________________

E-Mail:  josht at iname.com ICQ:  21219667 Web:  http://www.crosswinds.net/~josht
_______________________________________________________________________________


From wiktor1325 at my-deja.com  Wed Jan  5 21:49:22 2000
From: wiktor1325 at my-deja.com (wiktor1325 at my-deja.com)
Date: Thu, 06 Jan 2000 02:49:22 GMT
Subject: How to with **kw
Message-ID: <850vre$v3h$1@nnrp1.deja.com>

Is there a way to pass **kw args from
one function to another,like:

def func1(**kw):
  for key in kw.keys():
     print key

def func2(**kw):
  a_way_to_call_func1_with(kw)

thanks in advance

Wiktor Sadowski


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


From ivanlan at callware.com  Tue Jan 18 21:37:40 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 18 Jan 2000 19:37:40 -0700
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com>
Message-ID: <38852374.8292D5CE@callware.com>

Hi All--

Ivan Van Laningham wrote:
[snip]
>  and today we have such blights as the "Sci-Fi Channel"
> because of her ignorance, arrogance and insensitivity.
> 

I should make it clear that I'm not objecting to the Sci-Fi _Channel_,
only to its _name_.

<life-in-the-slow-lane>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From gawron at obop.com.pl  Wed Jan 26 06:46:24 2000
From: gawron at obop.com.pl (=?iso-8859-2?Q?Przemys=B3aw?= G. =?iso-8859-2?Q?Gawro=F1ski?=)
Date: Wed, 26 Jan 2000 12:46:24 +0100
Subject: Binary Numerical module
Message-ID: <388EDE90.661499FF@obop.com.pl>

Is there a binary distribution of Numerical module ?
I don't have VC++ installed and actually hope I can avoid doing so.

Thanks Przemek
--
Przemyslaw G. Gawronski UIN:8358522
mailto:gawronskip at usa.net
mailto:gawron at obop.com.pl





From bitbucket at isomedia.com  Mon Jan 10 18:26:55 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Mon, 10 Jan 2000 23:26:55 GMT
Subject: Python is much faster than Java?
References: <387A492A.CBB58B86@iqsoft.hu>
Message-ID: <387a6a5a.780742127@news.isomedia.com>

It may be too late for you to have an accurate measurement on this,
but did you also time how long it took to _write_ the sorting
algorithms?  No one seems to have anything but anecdotal evidence of
the RAD utility of Python (though I certainly don't refute it!).

     -Eugene

On Mon, 10 Jan 2000 22:03:38 +0100, Nemeth Miklos <nemeth at iqsoft.hu>
wrote:

>A colleague implemented a sorting algorithm in both Java and Python
>(CPython on NT) just to test the capabilities of Python.
>He is (or may be only was) a Java fan, but I spoke to him about Python
>so much, that he decided to make some comparisons.
>After the tests it turned out that the Python implementation was 5 times
>faster?
>We could not believe our eyes! How is it possible?
>A knew that Java is very poor in managing object (creating them and
>garbage collecting them), but I thought it is fast enough for procedural
>programming.
>
>NM
>
>

import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From nobody at nowhere.nohow  Sat Jan 22 21:13:32 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 23 Jan 2000 02:13:32 GMT
Subject: Python online docs are great!
Message-ID: <slrn88kouc.pej.nobody@isis.visi.com>

I just want to compliment GvR on the Python online documentation
files.  They are amazingly well done. They are without a doubt the
best online documentation I've used for anything, anywhere, ever.
Often online documentation has been chopped up into pages that are way
too small, and there's no useful index or navigation aids.  The result
is you spend all your time thrashing around because the needed
information is spread over a half-dozen different pages.  I've found
the Python documents very well organized, comprehensive, and easy to
use.

Python itself is very impressive, but I'm convinced that there are way
more people on the planet who can design a language and build a
compiler/interpreter than there are who can write decent documentation.

Seeing somebody do such a good job at both is humbling.

-- 
Grant Edwards                   grante             Yow!  I know how to do
                                  at               SPECIAL EFFECTS!!
                               visi.com            


From tye4 at yahoo.com  Tue Jan 18 15:50:55 2000
From: tye4 at yahoo.com (tye4)
Date: Tue, 18 Jan 2000 12:50:55 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net>
Message-ID: <862jqd$l9s@news.or.intel.com>

Tom Culliton <culliton at clark.net> wrote in message
news:YZRg4.1292$Ve7.28796 at iad-read.news.verio.net...
> In article <860ftk$bas at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
> Have you got extensive experience at designing and implementing
> computer lanaguages like Guido?  Did you spend years working on a
> major language design/development project (ABC) including lots of
> usability studies like Guido?  Have you had the benefit of Tim Peters,
> a veritable font of language implementation experience and good taste,
> whispering (or occasionally shouting) in your ear?  I rather doubt it.
>
Are you insinuating that this is not a problem since I'm not as experienced
as Guido? I don't need to design a language to detect flaws in it.
This problem needs to be fixed. That's because indentation is
highly dependent on what text editor you use. Some editors convert
tabs to spaces some don't. Only way around that would be not use
tabs... which is quite ridiculous (I don't want to type 4 spaces for every
line.)
Besides, some editors auto-indent which is a very nice feature.. but one
that might cause problems with Python.

> Part of the reason that Python is so clean and elegant is because
> Guido, with that background, discarded everything that didn't improve
> clarity, conprehensibility, and ease of learning, and only kept things
> that helped.

Python is clean and elegant.. yes but it could me much better.
Unfortunately, it's hard to change basic syntax of a language.

-tye4




From tye4 at yahoo.com  Tue Jan 18 19:48:59 2000
From: tye4 at yahoo.com (tye4)
Date: Tue, 18 Jan 2000 16:48:59 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <8631qm$26p@news.or.intel.com>

Remco Gerlich <scarblac-spamtrap at pino.selwerd.cx> wrote in message > If you
have problems with that, try indenting it more, and/or use
> blank lines. The following looks rather clear to me:
>
> if x == 10:
>       print 'x is 10'
>
> print 'Hello world'
>
> And of course, get used to it. It's pretty easy to do.
>

Actually, I use something like:

if x == 10:
    print 'x is 10'
#end if
print 'Hello world'

I'm more concerned with how many maintenance headaches will result if I were
to accidently add or delete a space on a line. Also, I would to be very
careful and choosy about what editor to use.. Some editors expand tabs while
others don't.. What indentation you see in one editor is sometimes different
to what is seen in another...

How hard can it be to add 'end' indicators to the language.. If these end
indicators are added, the language would still be backward compatible. I
can't believe nobody has fixed this for so long.

-tye4




From shawnball at uswest.net  Sun Jan  9 13:04:39 2000
From: shawnball at uswest.net (Andrew Ferk)
Date: Sun, 9 Jan 2000 12:04:39 -0600
Subject: Python Host
Message-ID: <bJ3e4.8$78.764@news.uswest.net>

Are there any web hosts that support Python?




From reic0024 at ub.d.umn.edu  Wed Jan 19 12:17:55 2000
From: reic0024 at ub.d.umn.edu (Aaron J Reichow)
Date: Wed, 19 Jan 2000 11:17:55 -0600
Subject: Newbie Q: Zope with static content?
Message-ID: <Pine.SO4.4.02.10001180010450.27409-100000@ub.d.umn.edu>

Good day,
	I hope this is an appropriate newsgroup to discuss this -- I read
this newsgroup offen, but do not subscribe to the Zope mailing list(s).  
	I am currently using Zope to develop some web-based applications,
and am really liking it.  Also, I am needing to have a webcam for this
site.  I had it working fine under Apache simply updating the jpeg on the
disk every 30 seconds.  But Zope, as far as I can see, cannot access
"static" content straight off the disk -- like I was doing with the webcam
jpegs with Apache.  
	Is there anyway Zope can read a file straight off the disk,
presumably within it's own directory structure? (I've Zope installed at
/var/zope)  I could run an Apache process just to server the webcam, but I
would like to avoid doing that.  The other option I see is to have the
jpeg uploaded to the Zope FTP server.  I'm having problems getting xtp to
work.  
	Also, I think it'd be handy, for other reasons, to know how one
can get Zope running zserver to access static content saved as flat files
on the filesystem.


Thanks in advance!
Aaron





From S.I.Reynolds at cs.bham.ac.uk  Sun Jan  9 09:00:20 2000
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Sun, 09 Jan 2000 14:00:20 +0000
Subject: urllib.urlopen returns 'connection refused' on Solaris
References: <3875CBCF.72E26CE7@ina.fr>
Message-ID: <38789474.6887@cs.bham.ac.uk>

Richard Gruet wrote:
> 
> Hi everybody,
> 
> On Solaris 2.6 (Sun OS 5.6), the fct urllib.urlopen() returns an error
> 'Connection refused' for all URLs:

I had the same problem. Locally, we have some network security features
installed that prevent direct access to http and ftp address by any
program. We can run Netscape as it uses a proxy server.

Try running Netscape on your machine without the proxy server to see if
you can download off site web pages. If you can't then you'll need to
see your sysop about it.

Stuart


From jae at ilk.de  Wed Jan 19 21:04:53 2000
From: jae at ilk.de (Juergen A. Erhard)
Date: Thu, 20 Jan 2000 03:04:53 +0100
Subject: Mapping Column Names to results using DBI
In-Reply-To: <3886458C.3256ED8A@exceptionalminds.com> (message from Timothy
	Grant on Wed, 19 Jan 2000 15:15:24 -0800)
References: <8825686B.007D977F.00@avnet.com> <3886458C.3256ED8A@exceptionalminds.com>
Message-ID: <20012000.5@sanctum.jae.ddns.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>>>>> "Timothy" == Timothy Grant <tjg at avalongroup.net> writes:

[SNIP]

    Timothy> Lance,

    Timothy> Thanks so much for your prompt reply.

    Timothy> I guess the above means that the DBI spec *doesn't* call
    Timothy> for this behaviour. I actually find that quite
    Timothy> surprising, [...]

<aol>Me too</aol>.

That's why I prefer PyGreSQL to the DB-Sig API.  PyGreSQL does it just
the way one would expect a language with dictionaries to behave:

   db.query("select * from table").dictresult()

returns a list of dictionaries keyed on column names.

The DB-Sig API seems, at least to me, to be doing things in a very
traditional way... I guess that's because most of the DB people are
used to this ugliness <0.3 wink>.

Bye, J

- -- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
              My WebHome: http://Juergen_Erhard.tripod.com
  Mesa - Free OpenGL API (http://www.ssec.wisc.edu/~brianp/Mesa.html)
         Win32 has many known work arounds. For example, Linux.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.1 (GNU/Linux)
Comment: Use Mailcrypt and GnuPG <http://www.gnupg.org/>

iEYEARECAAYFAjiGbUQACgkQN0B+CS56qs1FLACeNL0ucsSodW8nPTadW2TNVkIs
95kAn0TOLg5d64+QFjNqOOtLylmWTx+r
=aPza
-----END PGP SIGNATURE-----



From sameerc1 at yahoo.com  Wed Jan  5 22:41:28 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Wed, 5 Jan 2000 19:41:28 -0800
Subject: python with ASP
Message-ID: <007c01bf57f7$eb3941c0$db75fea9@desktop>

how could the following work.  It's supposed to either display a form or
display "Got Title".

<%@ LANGUAGE = Python %>
<%
if REQUEST.Form("title"):
%>
 <html>
 <b> Got Title</b>
 </html>
<% else: %>
 <html>
 <FORM ACTION="/pytest.asp" method="POST">
<b>Title:&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</b><BR>
 <input type="text" value="" size=60 name="title"><P>
 <input type=submit value="Submit">
 </form>
 </html>


__________________________________________
NetZero - Defenders of the Free World
Get your FREE Internet Access and Email at
http://www.netzero.net/download/index.html



From nielsenjf at my-deja.com  Thu Jan  6 17:38:10 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Thu, 06 Jan 2000 22:38:10 GMT
Subject: (wierd situation) got the answer :-)
References: <853064$ei7$1@nnrp1.deja.com>
Message-ID: <8535gi$ii8$1@nnrp1.deja.com>

Thanks to another poster Andrew Csillag, I know now that you need to do
the following:

exec('filename').read

It works!

john


In article <853064$ei7$1 at nnrp1.deja.com>,
  John Nielsen <nielsenjf at my-deja.com> wrote:
> I already tried from file import.
> That doesn't help me :-(
>
> Let me explain the situation further:
>
> I have code running in an ASP environment, but I don't want it to know
> that it is doing so. So I localize the ASP specific stuff in another
> file.
>
> The problem is the module in the other file does not have access to
> ASP's objects. And I don't want to send it those objects,
> since it is not supposed to know about it.
>
> Any ideas on the best way to deal with this?
>
> john
>
> --
> nielsenjf at my-Deja.com
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

--
nielsenjf at my-Deja.com


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


From JamesL at Lugoj.Com  Sat Jan 22 00:43:43 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Fri, 21 Jan 2000 21:43:43 -0800
Subject: Using tabs effectively. Was: Why not use tabs?
References: <1263658531-14676592@hypernet.com> <vndk8l3zbr1.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <3889438F.7A56990@Lugoj.Com>

Justin Sheehy wrote:
> 
> "Gordon McMillan" <gmcm at hypernet.com> writes:
> 
> > Um, there's no tabsetting that will produce that result, but with
> > the wonders of email, maybe that's not what you wrote.
> 
> I believe that James was thinking in terms of typewriter-style "tab
> settings", where a tab is not a fixed width, but rather moves to the
> next column that is defined as a tab stop.  Such things are common in
> word processors, but absent in most text editors.
> 
> However, I have no idea how such things are relevant to Python.  :-)

I cut and pasted from vi. It is an obscure editor one might use on an
ancient operating system known as Unix. I still have no idea what my
original message looked like to Gordon McMillan or what he was using to view
it (I'm too lazy to go back and check the Usenet headers to see if it tells
me). I was merely trying to point out that one must also worry about
embedded tabs as well as leading tabs and that ... oh never mind!

P.S. In what editors does a tab NOT designate a move to the next defined tab
stop? I've never come across an editor that treated a tab as being directly
translated into N spaces regardless of column position.


From alex at magenta.com  Sun Jan  2 17:32:57 2000
From: alex at magenta.com (Alex Martelli)
Date: Sun, 2 Jan 2000 23:32:57 +0100
Subject: assert (was Re: Python not a Very High-Level Language?)
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84mh65$2uv$1@harvee.billerica.ma.us>
Message-ID: <032401bf5571$c7c67240$c02b2bc1@martelli>

Eric S. Johansson writes:
> Of all the things he suggested, I personally want to see programming
> by contract support.  I've done it in C/C++ using assert.h and it's
> wonderful for picking out bugs in your thinking as well as your
> implementation.

Can't you, already today, use assert in Python, as well?

Surely not as nice as calling things "pre", "post", etc, but
should be a start...


Alex





From fredrik at pythonware.com  Fri Jan 21 03:48:37 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 09:48:37 +0100
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu>
Message-ID: <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com>

Roy Smith <roy at popmail.med.nyu.edu> wrote:
> When I was first learning programming (a quarter of a century ago!), 
> there was the idea of the "telephone test", i.e. your code should be 
> simple enough that if you read it to somebody on the phone they would 
> understand it.

hey, who the heck is reading code over the phone
these days?

> Python not only fails the phone test, it fails the printed page
> test; I can print two programs on a piece of paper, which 
> differ only in tabs and spaces, and they will appear identical
> yet which do different things.  Barf.

doesn't that mean that if you type them in again,
they may or may not work? ;-)

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! &smiley; -->




From grant at nowhere.  Wed Jan 19 12:17:35 2000
From: grant at nowhere. (Grant Edwards)
Date: Wed, 19 Jan 2000 17:17:35 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com>
Message-ID: <slrn88bsde.4pp.grant@grante.comtrol.com>

In article <3885D3E0.4F10F28D at netscape.com>, Michel Orengo wrote:

>As I told you, I am still a beginner wrt MAPI. It appears that
>my code worked but not always! In fact, I ran myself into
>several problems I don't understand. The strangest one is that
>some code run in Interactive window, but will not run
>otherwise! Moreover, I am using the lotus cc-mail version of
>MAPI. You are using the one provided by Outlook.

I guess so.  

I'm still pretty fuzzy on MAPI.  At first I thougt it was the
protocl used to talk to Exchange servers.  Then I thought it
was a library of routines provided by MS to talk to Exchange
Server.  Now it seems to be an API provided by e-mail programs
to let other applications invoke some of their functionality. I
think I'm getting closer.

I never realized how much simpler things are in the Unix world.

>In fact, you are lucky because your version is 1.2 I know that
>because I have access to the MSDN Library, and that's where you
>can find all the command/properties of objects of the CDO
>(Collaboration Data Object) the OLE wrapping of the MAPI.

I've got boxes and boxes of MSDN CDs that I inheritted from my
predicessor (and more arrive regularly), but I've never done
anything with them.  Perhaps it's time...

>If you do not have access to the MSDN Library, you can use a
>COM browser: it will give you the properties/methods of the
>objects. 

That's what I was trying to do yesterday afternoon.  The com
browser written in python only shows methods/attributes _after_
they've been used.  (Due to lazy dispatching, I presume.)

>Another solution is to make Python aware of the CDO
>object by using makepy (see QuickStartClientCom.html) In the
>file generated by makepy, you will find all the classes and
>attributes.

That sounds useful.

>What you want to do though should be not very complex. Here is
>a sample code (I didn't test it!), just to provide some idea:

[...]

Thanks!

>Note that you can also take a look at testExchange.py under the
>'test' directory.

I tried that, and got some sort of COM errors.  Since I had
absolutely no clue, I tried your code next.  It's probably time
to figure out how to find things in MSDN.

P.S. Don't tell anybody about this.  It would be just fine if
     everybody continued to think I knew nothing about win32
     programming.  ;)


-- 
Grant Edwards                   grante             Yow!  I've got to get
                                  at               these SNACK CAKES to NEWARK
                               visi.com            by DAWN!!


From neelk at brick.cswv.com  Wed Jan 12 19:44:25 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 13 Jan 2000 00:44:25 GMT
Subject: Python not a Very High-Level Language?
References: <000801bf5802$dcca7d60$0aa0143f@tim> <is66x6vng7.fsf@localhost.ps.uni-sb.de> <87u2kljrmz.fsf@den.home.net>  <isbt6rhudp.fsf@grizzly.ps.uni-sb.de>
Message-ID: <slrn87q817.2cl.neelk@brick.cswv.com>

Denys Duchier <Denys.Duchier at ps.uni-sb.de> wrote:
> Frank Sergeant <frank.sergeant at redneck.net> writes:
> 
> > I think I saw that you have an interface to Tk.  Could you discuss
> > GUI building in Mozart a little?
> 
> The Tk interface follows a fairly simple design: the Oz process starts
> a Tk subprocess and communicates with it via a socket.  On the Oz
> side, Tk widgets are represented by Oz objects.  The Oz/Tk interface
> is actually a very thin OO flavored layer.

I have played a (very) little with constraint-based GUI toolkits, and
am quite impressed with that style. Are there any plans to build such
a beast with Oz?

I've downloaded and built Oz, but haven't yet had a chance to study
it. I like Python and I like Prolog, so I am expecting to like Oz. :)


Neel


From darrell at dorb.com  Thu Jan 13 09:51:38 2000
From: darrell at dorb.com (Darrell)
Date: Thu, 13 Jan 2000 09:51:38 -0500
Subject: Ping in Python
References: <SKd8OPFYifsAUCG=19+7XNNTjqDx@4ax.com> <007001bf5d9d$55e2d180$f29b12c2@secret.pythonware.com>
Message-ID: <003f01bf5dd5$b291f3b0$0100a8c0@rochester.rr.com>

From: "Fredrik Lundh"
> and if you insist, Jeremy Hylton has ping and
> traceroute code at:
>     http://the-tech.mit.edu/~jeremy/python/
>

I didn't start this thread, but sure had a lot of fun with it :)
Jeremy's ping needed just a little touch up and it worked great.

Messing with raw sockets, I was able to watch a hacker attack on my system.
My defensive software was ringing bells and whistles. The .py I was
debugging started printing some interesting stuff. It was an RPC port probe
coming in.

--Darrell





From fredrik at pythonware.com  Fri Jan 14 05:19:09 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 14 Jan 2000 11:19:09 +0100
Subject: Python native compiler on NT
References: <84cdo5$9u0$1@nnrp1.deja.com> <3869EA26.9AE31E6F@matavnet.hu> <85mjrp$ci5$1@nnrp1.deja.com>
Message-ID: <001301bf5e78$cca1de70$f29b12c2@secret.pythonware.com>

tpchang at my-deja.com wrote:
> Thanks for the info. I tried the "Installer". However, to run
> the "compiled" .exe file, say xyz.exe, you have to copy xyz.exe and
> zlib.pyd to current directory before you can run it. Some how putting
> xyz.exe and zlib.pyd in a directory in the PATH, and run xyz.exe from
> another directory will have a error message something like:
> 
>    cannot load zlib.pyd
> 
> Do you know what's the problem?

Python scans the PYTHONPATH when looking
for PYD files, not the standard program path
(PYD's are extension modules, not executables).

try printing the contents of sys.path inside
your "compiled" script.  in worst case, you
can always add tweak sys.path yourself; e.g:

    import sys
    sys.path.insert(0, os.path.dirname(sys.executable))

or something like that.

</F>




From mlh at vier.idi.ntnu.no  Tue Jan  4 17:56:18 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 04 Jan 2000 23:56:18 +0100
Subject: Simple cartesian product
Message-ID: <iacpuvhqwdp.fsf@vier.idi.ntnu.no>

Just wondered if anyone had a simple and elegant little snippet that
returns the cartesian products of two lists... for-loops and append is
a simple solution, but not very "cool" ;)

I have a *nearly* finished function:

def product(a,b):
    a = 
    b = b * len(a)
    return map(None, a, b)

What I'm missing here is a simple expression which will make a list
like [1, 2, 3, 4] into [1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4] (if
len(b)==3)...

(Of course, all of this is a breeze with list comprehension ;)

--

  Magnus
  Lie
  Hetland


From sanderson at ttm.com  Mon Jan  3 12:14:51 2000
From: sanderson at ttm.com (Scott Anderson)
Date: Mon, 03 Jan 2000 12:14:51 -0500
Subject: Considering using python for something, and need a bit of info :)
References: <F652410E815802F88525685B005923FA.0059248B8525685B@ttm.com>
Message-ID: <3870D90A.693122D5@ttm.com>

> I'm involved in writing a MUD server from scratch with a small group of
> other people.  For you who do no know, a MUD is a multi-user text game.

As someone who's written both libs and muds, I'd say that Python would
be quite suitable. However, I think you'd want to write the whole thing
in Python to make it easier (I've been thinking of doing such a thing
myself for a while).

In particular, you'll want to look at the rexec and Bastion packages.

Regards,

-scott



From mhammond at skippinet.com.au  Tue Jan  4 19:36:13 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Wed, 05 Jan 2000 00:36:13 GMT
Subject: IDLE breakpoints and Go debugging command
References: <3871EF67.961E8E4E@ina.fr>
Message-ID: <1mwc4.9758$lv5.46455@news-server.bigpond.net.au>

"Richard Gruet" <rgruet at ina.fr> wrote in message
news:3871EF67.961E8E4E at ina.fr...

> And when I launch the execution (run script), the debugger stops on
1st
> line but subsequent 'go' only step one line (like a step over)
instead
> of running until the breakpoint is reached.
>
> Is this behaviour known ? is it a bug ? I searched the archives but

It is a known problem.  It will exhibit this behaviour until you step
into a function.  It is really a problem with pdb, not IDLE as such.
In Pythonwin I went to some pains to avoid this particular problem...

Mark.





From gvwilson at nevex.com  Sun Jan 30 15:17:26 2000
From: gvwilson at nevex.com (gvwilson at nevex.com)
Date: Sun, 30 Jan 2000 20:17:26 GMT
Subject: Eight suggestions for Python books (long)
Message-ID: <87268i$786$1@nnrp1.deja.com>

One of the most exciting things for me about the 8th Int'l Python
Conference in Washington last week was the growing number of books on
Python.  When I had to choose a language to use in a software
engineering course two years ago, I felt I had to choose Perl, because
there was so much more documentation for it than for any other Open
Source scripting language.  Last October, I finally felt that I could
switch to Python, and feedback since then has been pretty positive.

I heard several people at the conference say that they were also
excited and that they were thinking about doing a book.  That got me
thinking about what books the community really needs, and that got me
thinking about the list of unwritten books that I wrote for "Doctor
Dobb's Journal".  I started doing book reviews for DDJ a few years
ago.  After a while, I decided that the best way to get the books I
really wanted was to write the reviews first, and hope that someone
would come along and fill in the book later.  The result was posted at
http://www.ercb.com/feature/feature.0012.html, and got a very (very)
enthusiastic response from publishers, most of whom stay awake at
night trying to figure out what we're all going to want to buy a year
from now.

So, here's my first cut at what books Python needs, and why.  My main
interests are books that will help new Python programmers be
productive in a hurry, and help make Python look like a credible
solution in both academia and industry.  Your mileage may vary...

1. The Python Cookbook

The most frequent question in my classes is, "How do I do XYZ in
Python?"  Most of the time, the people asking the questions know how
to do what they wants in C or Perl; we need to make it easy for them
to find out what the answer is in Python.  The easiest way to write
this book would be to go through O'Reilly's "Perl Cookbook", throw
away those examples that exist only to help people get around Perl's
syntax and execution model, and re-write the rest in Python.

2. Algorithms and Data Structures in Python

This book complements "The Python Cookbook" by showing how classic
algorithms and data structures are implemented in Python, rather than
how to open a secure socket.  Again, the quickest way to tackle this
would be to take inspiration from books such as Orwant et al's
"Mastering Algorithms with Perl", or any of Mark Weiss's DS&A books.
One caution: remember that CPython and JPython manage memory
differently.  Try to show the best solution for each, rather than
trying to accommodate weaknesses that are specific to one or the
other.  Also try to include more theoretical analysis of algorithm
performance than Orwant et al did --- the long-term goal for this book
is to be usable in a second-year DS&A course.

3. Computer Science Python Style

Logo is to children's computing what C is to systems programming, and
Brian Harvey's three-volume series "Computer Science Logo Style" is to
Logo what Kernighan and Ritchie's "The C Programming Language" is to
C.  Anyone who wants to teach children to program, in any language,
will find "CSLS" a rich source of ideas, and a good guide to what
children (and their teachers) can actually understand.  In particular,
if anything is longer or harder to explain in Python than in Logo,
it's a good sign that an extension library is needed...

4. Software Tools in Python

Unix might not be the most widely-used operating system today, but it
has certainly been the most influential.  One reason for its success
is the ease with which different tools can be combined, and one reason
for that is that most Unix tools use a common data format: tokens
separated by whitespace, terminated by newline characters.  Another
reason was Kernighan's book "Software Tools" (and later "Software
Tools in Pascal"), which explained why and how to build small things
that did one job well.

The advent of XML means that there is now another universal data
format that a wide variety of tools could operate on.  Unlike flat
text, this format is hierarchical, but the principle of connecting
many tools together using pipes is still the same.  That raises three
questions: what are the equivalents of 'wc', 'head', 'grep', and 'sed'
for tree-structured data, what language will those tools be written
in, and who is going to write the best-selling book that explains all
of this?  As with (3) above, a lot of work would have to be done on
software to make this book possible, but anyone thinking about writing
usable XML filters in Python would do well to think about how to
explain them before laying down any code.

5. Case Studies in User Interface Design

Some philosophers distinguish between knowing that, and knowing
how. The former is mostly facts, such as the airspeed of an unladen
African swallow. The latter consists of techniques, like how to ride a
bicycle, or how to design a user interface. Unfortunately, knowledge
of the second type is hard to put into books, since every general rule
has exceptions, and none of the important ideas can be given exact
definitions.

This book on user interfaces would try to teach the "how" by working
through a series of examples. The starting point for each would be
either an existing interface that needs improvement, or a set of
requirements and a blank canvas.  Each design would be iterated
several times, or try to satisfy different constraints (e.g. hand-held
vs. desktop platforms).  The focus would not be on Python per se, but
implicit in the book would be the message that Python facilitates
rapid experimentation, prototyping, and evolution.

6. C for Python Programmers

As many people have pointed out, Python is an excellent teaching
language.  It runs almost everywhere, is used in industry, is
object-oriented, and checks for a lot of simple errors, including bad
type casts, dangling pointers, and out-of-bounds array indices.  If
Python becomes widely used in grade schools, and if highbrow features
like generic programming and type inference are added, it will start
to look very attractive as a college-level language.

The problem, however, is what to do with students in their second
year. As a sophomore, I took two courses: one on data structures and
algorithms, and one on machine architecture and assembly-language
programming. The first is easy to translate into Python, but what
about the second? Going from Pascal to PDP-11 assembler was bad
enough; going from an interpreted, garbage-collected language to
today's pipelined RISC architectures would be impossible.

This book's premise would be that C has always really been a
high-level assembly language, and that we should teach it to students
in relation to Python as we taught assembler in relation to Pascal. It
would show readers how to build a simple Python interpreter using C,
and would cover pointer arithmetic and array indexing, allocating and
freeing memory, garbage collection, signed vs. unsigned arithmetic,
and machine-dependent data sizes. Along the way, students would be
introduced to the quirks of C, including the use of preprocessor
directives to handle platform-dependent code.  By the time they were
done, students would be ready to tackle higher-level courses in
networking, operating systems, computer graphics, and other topics.
They would also have learned how to structure complex software systems
in layers and modules, and how to deal with a variety of real-world
software engineering issues.

(Note that this book would teach C, rather than C++, so that students
would learn what computers really do --- C++ sweeps far too much under
the rug for this purpose.)

7. Object-Oriented Analysis and Design Using Python and UML

The aim of this book is to teach analysis, design, and patterns as if
they were as important as language syntax or basic data structures.
The book would be aimed at sophomores who have learned an
object-oriented language (not necessarily Python) in a first-year
course. It would introduce both the idea of design patterns, and the
basic elements of the Unified Modeling Language (UML), and then show
how to apply both to a variety of web-enabled applications.  The
emphasis of the book would be how to analyze a problem methodically,
and how to design a program, or set of programs, to meet users'
needs.  Liberty's "Beginning Object-Oriented Analysis and Design", and
Rosenberg and Scott's "Use-Case Driven Object Modeling with UML",
would both be good inspirations, but in order to really be successful,
this book would have to include many more examples, sample questions,
and back-of-the-book answers than either.

8. Software Tools for Scientists and Engineers

I have a confession to make.  I sat down to start writing this book
four years ago, and found that I couldn't, because existing software
engineering tools were so hard to learn.  I found out the hard way
that most scientists and engineers have better things to do with their
time (like science and engineering) than learning how to write
non-trivial makefiles, or re-installing Gnats.  They're not stupid ---
many of them really are rocket scientists --- but they can tell the
difference between intrinsic and accidental complexity, and don't have
a lot of respect for how much of the latter there is in modern
computing.

The aim of the Software Carpentry project (www.software-carpentry.com)
is to create tools that are worth describing.  If it succeeds, I hope
that someone will take those tools, and Python itself, and use them to
show computational scientist and engineers how to capture software
requirements, write up designs, create test plans, accommodate special
features of different platforms, and everything else that makes up the
"other 90%" of programming.

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

Greg Wilson is an independent software developer and author.  He is
currently managing the Software Carpentry project on behalf of Los
Alamos National Laboratory and CodeSourcery, LLC.  You can find out
more about the project at http://www.software-carpentry.com, or reach
Greg at gvwilson at nevex.com.


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


From s.schwarzer at ndh.net  Wed Jan 26 03:44:08 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Wed, 26 Jan 2000 09:44:08 +0100
Subject: read from binary file
References: <38877de8.0@sia.uibk.ac.at>
Message-ID: <388EB3D8.99F07093@ndh.net>

Hello Stefan ;-)

Stefan Andreatta schrieb:
> [...]
> But the read() function always returns just some 1000 bytes of data and the
> stops.

This is probably due to some "end-of-file character" (Ctrl-Z on DOS,
OS/2, Windows, Ctrl-D on Unix, ? on Mac<-this is an embedded question
(on the character, not the Mac) ;-) ) in the file.

> Does anybody know, what could be the reason for that, and how I could read
> the whole file into a string?

As Ivan pointed out :-)  Additionally, it might be a good thing to
close the file when you don't need it open anymore. (It wasn't clear
from your posting if you finally closed the file.)

Stefan


From aa8vb at yahoo.com  Fri Jan 21 16:29:25 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 21 Jan 2000 16:29:25 -0500
Subject: Python - SIGPIPE handled bug
Message-ID: <20000121162925.A189033@vislab.epa.gov>

...or so it appears anyway.

  ---------------------------------------------------------------------------
    > python -c "for i in xrange(1000): print 'Some output'" | head -2
    Some output
    Some output
    Traceback (innermost last):
      File "<string>", line 1, in ?
    IOError: [Errno 32] Broken pipe
  ---------------------------------------------------------------------------

python getting a SIGPIPE is appropriate behavior.  "Whether" it gets a
SIGPIPE is up to the shell, in UNIX:

    http://www.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=567584780

That's all fine.  I have two questions:

  ---------------------------------------------------------------------------
  1) Why does SIGPIPE cause Python to throw an error exception, and call
     atexit functions?

     According to signal(2), the default C behavior should be to just _exit(2).
     So, by default, shouldn't Python just exit quietly with status 2, 
     without calling atexit functions?

  ---------------------------------------------------------------------------
  2) If SIGPIPE throwing an IOError is decided to be correct,
     what's the right way to suppress the error in Python code?  

     a) Resetting Python's SIGPIPE handling behavior back to what the C library
        defines (_exit(2)) is one option:

            from signal import signal, SIGPIPE, SIG_DFL
            signal( SIGPIPE, SIG_DFL )

        _exit(2) is now called, and no atexit functions are called.
        But we bypass exception handlers, which might not be a good thing.
        
     b) Install an SIGPIPE error handler, reset sys.exitfunc inside it, and
        then manually throw an IOError from inside the signal handler?

        Does Python have any atexits internally?  Should those be called?

     c) Or, we let the IOError bubble to the top, let all the exception
        handlers fire, and then invoke this verbosity:

            try:
              my_program()
            except IOError, x:
              if x.errno == errno.EPIPE:
                os._exit(2)
              else:
                raise

        Is this best?  Is it portable?  And is there a simpler way?

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

Opinions appreciated.

Thanks,

-- 
Randall Hopper
aa8vb at yahoo.com
-------------- next part --------------
# write.py - Demonstrates the hoops you have to jump through
#            to deal with SIGPIPE in Python.  This isn't right because
#            sys.exitfunc (and Python's atexits) are called.


import errno, sys

def at_exit():
  sys.stderr.write( "atexit called (a bug)\n" )

def my_program():
  for i in xrange(10000): 
    print 'Some output'

sys.exitfunc = at_exit 
try:
  my_program()
except IOError, x:
  sys.stderr.write( "Ok, now lets do some cleanup\n" )
  if x.errno == errno.EPIPE:
    sys.stderr.write( "Got a SIGPIPE!\n" )
    raise SystemExit, 2
  else:
    raise

From samschul at pacbell.net  Sat Jan 15 18:00:42 2000
From: samschul at pacbell.net (sam schulenburg)
Date: Sat, 15 Jan 2000 15:00:42 -0800
Subject: How to use Pythonwin's ability to access windows dll's???
Message-ID: <XX6g4.24$vz.3707@nnrp3-w.snfc21.pbi.net>

I can access windows dll's using Sam Rushing's calldll.pyd and windll.py. My
question is are thes features built into Pythonwin? I find that I can use
win32ui.LoadLibrary('MyDll') to obtain a PyDll object, but I can not find an
example on how to use this Pydll object to access a function within the dll.
For example:
MyPydllobj = win32ui.LoadLibrary('kernel32.dll') will obtain the object, but
how do I access the Beep() function within the dll?

Sam Schulenburg




From aahz at netcom.com  Sun Jan 16 23:38:52 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jan 2000 04:38:52 GMT
Subject: Pardon my ignorance, data conversion
References: <38829357.ED633932@home.com>
Message-ID: <85u6cs$85d$1@nntp9.atl.mindspring.net>

In article <38829357.ED633932 at home.com>,
Robert W. McGwier <rwmcgwier at home.com> wrote:
>
>I need to do the simple task of taking a nonnegative floating point
>variable and taking its integer part and storing it in an integer.
>Damned if I can find how to do it.

import math
i = int(math.floor(real))
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From tim_one at email.msn.com  Sat Jan 22 16:16:23 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 22 Jan 2000 16:16:23 -0500
Subject: Python dumps core with this.. how come?
In-Reply-To: <m3ln5i3x74.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <000001bf651d$f05ac240$132d153f@tim>

[Interrupting Michael Hudson's fine advice to inject a nit]

[Roey Katz]
> this brings up another concern. Given
> the following definition:
>
>    class A:
>
>       def __getattr__( self, name ):
>         return -1
>       def __setattr__( self, name, val ):
>          self.__dict__[ name ] = val
>
> I see that self.__dict__ must be a special variable
> in that setting it somehow avoids a recursive call
> to __setattr__()?

[Michael]
> Yup. Look in Objects/classobject.c around about line 580
> if you want the gory details.

While true, it doesn't apply to Roey's example:  he's not setting __dict__.
He's binding a key *of* self.__dict__; self.__dict__ itself is merely
referenced.  So Roey should really be wondering why the reference to
self.__dict__ isn't invoking __getattr__ (&, of course, that's also due to
special-case __dict__ magic, but in __getattr__'s implementation).

I don't mean to be overly picky; it's just that the excruciating details are
vital when mucking with __getattr__/__setattr__, and *any* small
misconception about them will eventually bite hard.

> ...
> Bizarrely,
>
> filter(dest.append,src)
>
> is probably more memory efficient (the map form conses up a
> list of return values, but as append always returns None -
> which is false - filter `rejects' every element) - but if
> you write code like this, you deserve to lose.

Yes, pointing him to the most bizarre endcases of Python behavior will
surely aid him in "resetting his brain" <wink>.

Guido gave some excellent semi-exasperated advice (similar in spirit to
parts of Michael's advice that I've snipped) many moons ago, when the whole
newsgroup was temporarily filled with msgs of this character:  he invited
people to just *try* writing a Python program without using any __xxx__
hooks.  I expect some people were amazed at how much easier their lives got!
If he were to get that exasperated again today, I bet he'd expand his
invitation to include avoiding map, filter, reduce and lambda too.

you're-not-ready-to-use-them-in-python-before-you're-
    happy-without-them-ly y'rs  - tim





From phil at geog.ubc.ca  Mon Jan 10 19:20:55 2000
From: phil at geog.ubc.ca (Phil Austin)
Date: 10 Jan 2000 16:20:55 -0800
Subject: Did you all see this??
References: <387A4D4C.3D1F8330@zora.nl>
Message-ID: <m3zoudxxug.fsf@brant.geog.ubc.ca>

jaap groot <jaap at zora.nl> writes:

> lwn.net/daily shows a link to a site
> (http://www.software-carpentry.com/)where an implementation 
> contest with big prices is announced.
> Four projects to replace for instance Make.
> And all have to be implemented primarily in Python.
> 

Yes, for such interesting news it doesn't seem to have made much of a
splash.  Perhaps something more official in the way of an announcement
is planned for the Python conference?


-- 
------------------------------------------------------------
Phil Austin  (phil at geog.ubc.ca)
Department of Geography,                 Tel: (604) 822-2663
University of British Columbia, B.C.     Fax: (604) 822-6150


From darcy at vex.net  Fri Jan  7 09:07:09 2000
From: darcy at vex.net (D'Arcy J.M. Cain)
Date: 7 Jan 2000 14:07:09 GMT
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net> <20000106143130.A4003@stopcontact.palga.uucp>
Message-ID: <854rud$q3q$2@news.tht.net>

Gerrit Holl <gerrit.holl at pobox.com> wrote:
> Chuck Esterbrook wrote on 947100067:
>> Seems like many web host providers, such as MindSpring, offer CGI with
>> Perl, but not other languages, notably Python.
>> 
>> Does anyone know who would host a CGI Python site?

> I think vex.net's doing that... I don't think parnassus runs on perl ;-)

Yep.  We got Python.  And given that the owner (me) maintains PyGreSQL
and our entire billing system is written in Python, that has about a zero
chance of changing.  :-)

-- 
D'Arcy J.M. Cain <darcy at vex.net>           |  Democracy is three wolves
http://www.vex.net/                        |  and a sheep voting on         
+1 416 425 1212     (DoD#0082)    (eNTP)   |  what's for dinner.


From rjroy at takingcontrol.com  Sat Jan 15 09:45:56 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Sat, 15 Jan 2000 14:45:56 GMT
Subject: Help me improve my Python Skills, please
References: <m2g0vzsrte.fsf@excalibur.peernet>
Message-ID: <38808726.751144390@news1.on.sympatico.ca>

On 15 Jan 2000 00:43:41 -0700, tarude at whitestar.localdomain (Trebor A.
Rude) wrote:

>
>I just started picking up Python a few days ago, and am rather
>enjoying it.  To help me learn it, I converted one of my old TCL
>scripts into Python, which I've attached below. However, this being my 
>first serious python script, I'm pretty sure I'm doing some things
>less efficiently than necessary, through lack of familiarity with the
>library and language in general. So I'd like to enlist the group's
>help in improving the code. Logic, style, better ways of doing things, 
>whatever.  Thanks in advance.
>
>A little explanation before the code.  The script recursively descends 
>a directory structure, building a .jdb (Jukebox database) file for
>each directory based on the mp3 tag info of the mp3s in that
>directory and all its subdirectories. It also repairs damaged comment
>fields caused by a slightly buggy encoder program.
>
The os.path.walk method provides an elegant way of recursing a
directory structure. It might be useful here.


From jmagave at yahoo.com  Thu Jan 20 22:20:51 2000
From: jmagave at yahoo.com (Jayr Magave)
Date: Fri, 21 Jan 2000 03:20:51 GMT
Subject: What about REBOL (www.rebol.com)
Message-ID: <ngQh4.17853$pb2.1435429@tw11.nn.bcandid.com>

Sorry to introduce a strange subject to this group. Please, no flames.
In fact, it isn't really strange -- I woul like to know if someone evaluated
the new script language REBOL. It looked very powerful and high level to me.
But lacks library support (specially database support).

I am about to take my time to learn an useful multiplataform script language
and Python looked very fine,  until I took a look at REBOL.

I would to like to hear from you all.

Thanks

Jayr Magave




From s.schwarzer at ndh.net  Tue Jan 25 08:26:53 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Tue, 25 Jan 2000 14:26:53 +0100
Subject: And there was the whitespace topic again ;-) (was Re: ( 2.31.New 
 operators: 'eq', 'ne', 'last', '..' ) ?)
References: <388670D2.2FA842F6@starmedia.net> <Pine.SOL.3.91.1000120001532.29235B-100000@sunse> <tIeGOHk6cMDxz41yCwn96Iy45Dgx@4ax.com> <388CF7C4.D6247285@ndh.net> <oqln5ex5kn.fsf@titan.progiciels-bpi.ca>
Message-ID: <388DA49D.C9B2A17@ndh.net>

Hello Fran?ois,

Fran?ois Pinard schrieb:
> To be honest, I just do not understand the big fuss some people do about
> whitespace in Python.  They are all given a very nice gift, and it makes
> their misery.  I shudder to think what their sources would look like, if
> they really consider that indentation is a curse.  I like the idea that
> Python programs have readability on their side from the start, by design.

The concerns are (as far as I understood) not against whitespace to
be semantically relevant, but to the fact, that the same looking code
could perform different actions if you use another tab/space scheme
for indentation.

Because this _follows_ from the "whitespace (here: indentation) is
relevant" issue, some would like to abondon the whole relevancy of
indentation to get rid of their problems, - which could IHMO also be
solved with a good editor. As a side effect, this solution would (not
would but actually does :) ) keep the _advantage_ to spare the verbal
block delimiters (e. g. "endbla") ;-) .

Stefan


From phd at phd.russ.ru  Tue Jan 18 10:48:26 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Tue, 18 Jan 2000 15:48:26 +0000 (GMT)
Subject: DCOracle Large Objects
Message-ID: <Pine.LNX.4.21.0001181545370.15178-100000@fep132.fep.ru>

Hello!

   Anyone can help with Oracle LOBs and stored procs?
   I want to create a LOB in a database. DB programmer created a stored
procedure for me, but when I call it, passing data as Python string:

   cursor.execute("add_image", [1028, "This is the body of my Large Object"])

I got the error "invalid SQL". How do I pass the body of LOB to stored proc?

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From emwpeso at emw.ericsson.se  Fri Jan 28 05:19:36 2000
From: emwpeso at emw.ericsson.se (Per Soderstam)
Date: Fri, 28 Jan 2000 11:19:36 +0100
Subject: Installing quirk...
References: <389007A5.6B22105C@emw.ericsson.se>
Message-ID: <38916D38.2C438F11@emw.ericsson.se>

Thanks everyone! It's working!

/Per

Per S?derstam, Ericsson Microwave Systems AB, SE-431 84 Molndal, SWEDEN, 
tele: (+46) (0)31 7470195, e-mail: per.soderstam at emw.ericsson.se
********************************************************************
The views expressed above is my own and does not nessecarily
reflect the policy of my employer. http://www.ericsson.se/microwave/


From teep at danaicorp.com  Tue Jan 25 07:38:50 2000
From: teep at danaicorp.com (Prateep Siamwalla)
Date: Tue, 25 Jan 2000 19:38:50 +0700
Subject: newbie question about self and eval
References: <86jjkj$jmu$1@news.inet.co.th>
Message-ID: <86k58e$kqc$1@news.inet.co.th>

OK, this one works.....

self.__dict__[token[0]].append(token[1])

Prateep Siamwalla wrote in message <86jjkj$jmu$1 at news.inet.co.th>...
>Hello o' wise pythoners,
>
>I have a simple question to ask any generous pythoners out there:
>
>I have a simple tab delim text file:
>test.txt
>---------
>fee    harry
>fi    joe
>fo    jack
>fum    jim
>fee    john
>fo    sam
>
>I trying to make a class to read in these values and stuff them into lists
>which this class keeps
>
>(warning, spacing migth be off)
>-------
>class myclass:
>
>    def __init__(self,filename='c:\\test.txt'):
>        self.fee = []
>        self.fi = []
>        self.fo = []
>        self.fum = []
>        f = open(filename,'r')
>        rl = f.readlines()
>        f.close()
>        for lines in rl:
>            token = string.split(lines,'\011')
># do something more efficient than this !!!
>            bucket = token[0]
>            if bucket =='fee':
>                self.fee.append(token[1])
>            elif bucket == 'fi':
>                self.fi.append(token[1])
>            elif bucket == 'fo':
>                self.fo.append(token[1])
>            else:
>                self.fum.append(token[1])
>
>---------
>I know there must be a more effective way of appending items into the
>various class lists, (which probably involves eval or something similar,
but
>I've tried some feeble ideas, none of which seem to fly.
>
>I've tried self.eval(token[0]).append(token[1])  but that obviously failed
>in a big way.  Could someone please show me the way?
>
>Thanks in Advance,
>teep
>
>
>
>




From mal at lemburg.com  Wed Jan 19 13:18:20 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 19 Jan 2000 19:18:20 +0100
Subject: procmail-like spamfilter ?
References: <3884EA56.4A3C29C8@lemburg.com> <001301bf6270$9bc08d50$f29b12c2@secret.pythonware.com>
Message-ID: <3885FFEC.DE6D1748@lemburg.com>

Fredrik Lundh wrote:
> 
> M.-A. Lemburg <mal at lemburg.com> wrote:
> > Does anyone know of a procmail like spam filter that I could
> > use to prevent all these spam mails arriving at my mail box ?
> >
> > Ideally it should be written in Python, of course :-) Snakes
> > eating spam... you know.
> 
> here's one:
> http://www.nightmare.com/software.html => SpamWall

Thanks :-)

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




From e.e.sutton at cummins.com  Mon Jan 31 10:52:26 2000
From: e.e.sutton at cummins.com (e.e.sutton at cummins.com)
Date: Mon, 31 Jan 2000 15:52:26 GMT
Subject: Newbie: How to create a global or static variable?
Message-ID: <874b3n$mhd$1@nnrp1.deja.com>

I am trying to convert a simple Microsoft VBScript example script which
manipulates an Excel spreadsheet via COM (excel.vbs) to Python for an
apples-to-apples comparison of this particular script and how it looks
in each language. I know can I have classes in Pyhton but I do not want
to do that yet.

1 - How do I create a global variable in a script that is in scope for
all the functions defined in the script?

2 - Can I create a static variable with an initial assignment in a
function?

I have not yet found the documentation I need to explain the scope of
variables in Python. I am coming from a C/C++ background.


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


From news at titanic.co.uk  Wed Jan 12 17:37:59 2000
From: news at titanic.co.uk (Simon Faulkner)
Date: Wed, 12 Jan 2000 22:37:59 GMT
Subject: Text File
Message-ID: <cc0q7s8ggiqj62j27mcrtgg5a374qbm94h@4ax.com>

Hi All,

I would like to write a CGI program that diskplays a small (<1k) text
file on a web page and saves any changes made by the browser back to
the same file.

Any sugestions to help me get started?

Specifically

Which data type should I use and how do I read the text  file into it?

Cheers  


Simon



From johnjensen at my-deja.com  Thu Jan 27 01:13:38 2000
From: johnjensen at my-deja.com (johnjensen at my-deja.com)
Date: Thu, 27 Jan 2000 06:13:38 GMT
Subject: Does Monty Python know?
Message-ID: <86onmg$l40$1@nnrp1.deja.com>

I was just wondering whether anyone has ever formally or informally
told the former members of the Monty Python troupe that a popular
computer programming language has been named in their honour.  Does
anyone have any anecdotes in this regard?  Will John Cleese be speaking
at the next Python Convention?

John


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


From hei at adtranzsig.de  Mon Jan 31 09:32:19 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Mon, 31 Jan 2000 15:32:19 +0100
Subject: Namespace pollution
Message-ID: <949332864.553076@news.adtranzsig.de>

In the recent whitespace war,
some people said "Python is such an expressive
language, you need only short functions.", "don't indent
too much, that's bad style", even Linus Thorvalds was
quoted (who likes 1 TAB = 8 spaces indentation to not
lose control)...

I've got one problem with all this stuff:
I end up with dozens of helper functions that are
called exactly one time. They pollute the namespaces
of the module or class they're defined in. I hate polluted
namespaces.

How do others cope with this problem (that does not only
apply to Python)?

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






From mfletch at tpresence.com  Wed Jan  5 18:44:12 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Wed, 5 Jan 2000 18:44:12 -0500 
Subject: lazy (?) evaluation, request for advice
Message-ID: <CC59868A0A76D311B1F50090277B321C1F300A@VRTMAIL>

Idea:

class Hand:
	def losers( self ):
	def spades( self ):


# setup global environment
env = {"whatever":somefunction, "whenever":somevalue }
selected = []
for hand in somelistofhands:
	# setup specific environment
	for boundfunction in [ "losers", "spades", "hearts"]:
		env[ boundfunction ] = getattr( hand, boundfunction )
	if (eval( userfunction, env )):
		selected.append( hand )
env.clear() # to eliminate references to the last hand's methods

So the user then writes the following:

if (spades() > 6) or (3.0 < losers() < 4.0):
	whatever()

That is, define an environment in which the code runs, then put simple
functions in that environment to take care of the lazy evaluation explicitly
for the given algo.  Alternatively, you could poke around in the internals
of a function definition picking up names and deciding which elements to
calculate, but then you're still calculating losers for every hand in the
above example, instead of only doing it when spades() > 6 .

HTH,
Mike
		

-----Original Message-----
From: Alex Martelli [mailto:alex at magenta.com]
Sent: Wednesday, January 05, 2000 6:17 PM
To: python-list at python.org
Subject: lazy (?) evaluation, request for advice
...

    (spades>=6 or hearts>=6) and (3.0<losers<4.0)
rather than have to mention "current_hand.spades" or
other such explicit notation.
...



From quinn at pfennig.ugcs.caltech.edu  Fri Jan 14 12:40:10 2000
From: quinn at pfennig.ugcs.caltech.edu (Quinn Dunkan)
Date: 14 Jan 2000 17:40:10 GMT
Subject: best Template module
References: <6D8A17398E28D3119F860090274DD7DB4B3DD1@pces.cadlab.it>
Message-ID: <slrn87unua.2tu.quinn@pfennig.ugcs.caltech.edu>

On Fri, 14 Jan 2000 11:30:39 +0100, Alex Martelli <Alex.Martelli at think3.com>
wrote:
>Otis Gospodnetic writes:
>
>> Could anyone recommend a good 'Template module'?
>> 
>> Template module - a module that will let me create a plan text/HTML
>> file with some special markup/tags that I can then replace by some text
>> values from within my Python code
>> 
>> good - fast execution and not too complex :)
>> 
>> 
>Shorn of the 'statement-level' functionality (conditionals and loops),
>this is what my smartcopy.py boils down to:

Also, if you want simple and fast, python's built-in interpolation tags work
just fine:

% cat >template
Dear %(sucker)s,
    We're glad you've signed up for the "10,000 DISPOSABLE PLASTIC GIZMOS"
program.  Your first shipment will arrive by %(date)s, following shipments
will arrive every %(interval)d days.
    Sincerely,
    %(shark)s
^D
% python
Python 1.5.2 (#7, Dec 16 1999, 12:11:26)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> s = open('template').read()
>>> print s %{'sucker': 'John Smith', 'date': 'tommorrow', 'interval': 7,
...         'shark': 'Mr. Black'}
Dear John Smith,
    We're glad you've signed up for the "10,000 DISPOSABLE PLASTIC GIZMOS"
program.  Your first shipment will arrive by tommorrow, following shipments
will arrive every 7 days.
    Sincerely,
    Mr. Black
>>>

s % locals() is also a popular idiom.

If you want to get more elaborate, check out the DocumentTemplate module from
the zope distribution.  You don't have to use it with zope, and you don't have
to generate html with it, although it assumes you are.  The other thing which
may or may not be an advantage is that DT restricts embedded python statements
to 'safe' ones.


From torppa at polykoira.megabaud.fi  Sat Jan  8 18:57:58 2000
From: torppa at polykoira.megabaud.fi (Jarkko Torppa)
Date: 8 Jan 2000 23:57:58 GMT
Subject: pipe problem on FreeBSD....
References: <200001082039.PAA12903@acer.spvi.com>
Message-ID: <858iu6$m75$1@news.kolumbus.fi>

In article <200001082039.PAA12903 at acer.spvi.com>, Steve Spicklemire wrote:
>   Hmmm.. what I thought started out as a "Zope Bug" isn't. Let me 
>forgo all the history here and cut to the current analysis. There
>seems to be a problem on FreeBSD with python sending large amounts of 
>data through a pipe that blocks.
...
>Here is the reference to that problem report:
>http://lists.zope.org/pipermail/zope/1999-December/015874.html
...
>I think I've now simplified it to it's most basic form:

This is quite different matter that what you are seeing in Zope.
>This python script:
...
>fails on FreeBSD when invoked thusly:
>
>pluto.spvi.com> python tryit.py | head -3
...
>IOError: [Errno 32] Broken pipe
>where this c program does not:
..

Here is c program that does tries more or less match python's 
behaviour. 
----
#include <stdio.h>
#include <err.h>
#include <signal.h>

int main() {
  size_t charCount,c;
  char buffer[1024];
  FILE *f;

  signal(SIGPIPE,SIG_IGN);

  if((f = fopen("blah","r")) != NULL) {
    do {
      charCount = fread(buffer,1,sizeof(buffer),f);
      c=fwrite(buffer,1,charCount,stdout);
      if ( c != charCount )
	err(1,"WriteFailed");
    } while(charCount == sizeof(buffer));
    close(f);
  }
  else
    printf("Can't open file... blah\n");
}
---
tuhnu[2]%./tryit | head -3
...
tryit: WriteFailed: Broken pipe

>My thinking is that they should behave in the same way. (This also
>happens with 'more' and other such blocking programs.) Where in python
>should I look for the difference between tryit.c and tryit.py? Of
>course on linux both programs *do* behave the same way... I just
>created the 'c' version to see if it was deep in FreeBSD or somewhere
>in the python port. The python and c version work correctly in linux.
><dang!>

By default SIGPIPE terminates the program, also your cprogram
did not check writes return value so it could not have
caught the error anyhow. 

Behaviour you are seeing is correct. But this is different from
blocking it should not give SIGPIPE, writing thread should
just stall untill there is space in pipe to put the data.

This python piece does blocking reads on stdin

import time,sys
while 1:
    r=sys.stdin.read(1024)
    if len(r) == 0:
        break
    sys.stdout.write(r)
    time.sleep(30)

-- 
 Jarkko Torppa                torppa at staff.megabaud.fi
  Megabaud Internet-palvelut


From markus.oberhumer at jk.uni-linz.ac.at  Fri Jan 21 14:08:21 2000
From: markus.oberhumer at jk.uni-linz.ac.at (Markus F.X.J. Oberhumer)
Date: Fri, 21 Jan 2000 20:08:21 +0100
Subject: [ANNOUNCE] PySol 3.21 - a solitaire game collection
Message-ID: <20000121200821.A8582@laetitia.oberhumer.com>

-----BEGIN PGP SIGNED MESSAGE-----


                    PySol - a Solitaire Game Collection
                              Version 3.21

                          http://pysol.tsx.org

           Copyright (C) 1998, 1999, 2000 Markus F.X.J. Oberhumer
                   <markus.oberhumer at jk.uni-linz.ac.at>


What is PySol ?
===============

PySol is an exciting collection of 153 solitaire card games.

Among the supported games are classics like Aces Up, Baker's Game,
Canfield, FreeCell, Forty Thieves, Golf, Klondike, Monte Carlo,
Osmosis, Pyramid, Scorpion, Spider, Yukon, and many more...

PySol may be freely distributed under the terms of the GNU GPL.

PySol aims to be Commercial Quality Freeware.


Why yet another solitaire game ?
================================

Here are some highlights of PySol:

  - freely available
  - distributed under the GNU GPL with full source code
  - currently supports 151 (!) distinct solitaire variants
  - based upon an extensible solitaire engine
  - very nice look and feel including multiple cardsets
    and background table tiles
  - unlimited undo & redo
  - load & save games
  - player statistics and log files
  - hint system
  - demo games
  - support for user written plug-ins - add your own solitaire variants
  - integrated HTML help browser
  - lots of documentation
  - fully portable across Unix/X11, Windows 95/98/NT and MacOS
  - written in 100% pure Python
  - just run it - no need to compile anything


Nice. And what's new ?
=======================

  * Implemented 2 new games: Kingdom and Maze.

  * Added a missing DLL to the Windows setup file.

  * Updated the pysol-sound-server.


Cool. Where can I get it ?
==========================

Point your browser to http://pysol.tsx.org
The PySol Gallery is awaiting your visit as well.


What do I need to start playing ?
=================================

PySol requires Python 1.5.2 and Tcl/Tk 8.0.5. Both packages are
freely available for Unix, Windows 95/98/NT and Macintosh.

The Windows version ships as a completely self-contained setup file.

BTW, there is no need to compile anything since the whole program is just
a Python script. Just run it, and that's all.


License terms
=============

PySol is Copyright (C) 1998, 1999, 2000 Markus Franz Xaver Johannes Oberhumer

PySol is distributed under the terms of the GNU General Public License (GPL).
See the file COPYING.


Have fun,
Markus

http://pysol.tsx.org



-----BEGIN PGP SIGNATURE-----
Version: 2.6.3ia
Charset: noconv

iQCVAwUBOIipZm10fyLu8beJAQGIFgP/XqYdEtPlngAdB21+77IqLibxtzPPvcvU
k2M4fA/tcCsxiOrKcjeKr70WLEOBZwSgrxgP5HLI/j/4uwW9WO3vc0PgO+oyaPlW
fbGrQGA2V1X5YeADj0Z7zD3qBPMQn5Y7yXYvmjrCFUjDh2rFqR/d5dDYVq2/5FGz
vxU+a0pwHn8=
=ksJw
-----END PGP SIGNATURE-----



From Denys.Duchier at ps.uni-sb.de  Thu Jan  6 17:34:48 2000
From: Denys.Duchier at ps.uni-sb.de (Denys Duchier)
Date: 06 Jan 2000 23:34:48 +0100
Subject: Python not a Very High-Level Language?
References: <000801bf5802$dcca7d60$0aa0143f@tim>
Message-ID: <is66x6vng7.fsf@localhost.ps.uni-sb.de>

"Tim Peters" <tim_one at email.msn.com> writes:

> If you've got the time, investigate Oz/Mozart for us!
> 
>     http://www.mozart-oz.org/
> 
> If I had the time, that's the one I'd check out next, because it's the one
> I'm most prejudiced against <no wink>:  it seeks to combine all of logic,
> constraint-based, functional, and OO approaches on a highly concurrent
> substrate.  When I picture those all together, I can only conjure up visions
> of extreme pain; but browsing thru the intro material didn't hurt at all,
> and someone here last year (Boris?) spoke highly of it.  If not truly
> innovative, it's at least bold <wink>.

Hi Tim! Thanks for the publicity (your check is in the mail, minus the
penalty for that "extreme pain" bit :-) Mozart is getting quite a bit
of attention these days (finally!), in part because it is the most
advanced language for concurrent constraint programming, but
increasingly also for its support for concurrent as well as
transparently distributed computations.

As one of mozart's designers and implementors, as well as a python fan
since the early days, I'd be happy to field questions, if there is any
interest.

Cheers,

-- 
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 boud at rempt.xs4all.nl  Fri Jan 21 01:55:13 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 21 Jan 2000 06:55:13 GMT
Subject: HTML Widget
References: <38876AB2.58FF3EA5@hks.com> <slrn88f5n6.q2r.zeitlin@seth.lpthe.jussieu.fr>
Message-ID: <868vsh$rf2$1@news1.xs4all.nl>

Vadim Zeitlin <zeitlin at seth.lpthe.jussieu.fr> wrote:
> On Thu, 20 Jan 2000 15:06:11 -0500, Chris Swepson <swepson at hks.com> wrote:
>>    I've been searching far and wide for an HTML widget which can be
>>used in Python/Motif.

>  Hello,

>  I don't know if it works under Motif (it should, but I didn't test it), but
> wxPython has interface to wxHTML library which you could check out.

Most modern widget sets would have a html widget - the PyKDE one is
well up to scratch, too.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From jfzhou at 188.net  Sun Jan 30 19:48:09 2000
From: jfzhou at 188.net (Öܹ«)
Date: Mon, 31 Jan 2000 08:48:09 +0800
Subject: Why Perl do But Python can't?
Message-ID: <872m6v$2pdg$1@news.cz.js.cn>

I write a CGI useing Perl to modify passwd,set the CGI mod to 4755,and it
really do.
But When I change it into Python with the same function,set mod to 4755,but
it not use the Setuid bit.

Someone know the difference of Perl and Python in the CGI running mode?

Luckey at soim.net
2000.1.30





From jussij at zeusedit.com  Mon Jan 31 19:09:38 2000
From: jussij at zeusedit.com (Jussi Jumppanen)
Date: Tue, 01 Feb 2000 11:09:38 +1100
Subject: Zeus Programmer's Text Editor
Message-ID: <38962442.21C3@zeusedit.com>

Zeus for Windows Version 3.20 Now Released
------------------------------------------
This is an announcement to indicate that the Zeus for Windows 
Version 3.20 has now been released. Zeus is a programmer's is 
designed to run on the Windows 95, Windows 98 or NT platforms.

What is new in Version 3.20:
    - FTP file editing
    - Full DDE support added
    - Supports for the wheel mouse added
    - Minor bug fixes

Other features include:
    - Builtin scripting support for the Python, LUA or 
        Small C languages 
    - Fully configurable syntax highlighting comes 
        pre-configured for: 
         C/C++, Java, Perl, HTML, Python, Pascal, etc
    - Fully configurable compiler, linker and tool support
        will run almost any compiler, linker or tool.
        
For some screen shots of Zeus in action goto:
  http://www.zeusedit.com/z300/look.htm

You can download the shareware version from:
  http://www.zeusedit.com/z300/download.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 skip at mojam.com  Mon Jan 24 11:35:40 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 24 Jan 2000 10:35:40 -0600 (CST)
Subject: Living without static member functions
In-Reply-To: <86hbmt$g8s@newton.cc.rl.ac.uk>
References: <86hbmt$g8s@newton.cc.rl.ac.uk>
Message-ID: <14476.32604.435090.245040@beluga.mojam.com>

    Richard> There's a programming quirk of mine that I haven't seen
    Richard> elsewhere, although probably well-known to some of you. Here's
    Richard> a rough outline of the code:

    Richard> class Milkman:

    Richard>     allMilkmen = {}

    Richard>     def __init__(self, name):
    Richard>         ...
    Richard>         allMilkmen[name] = self

    Richard>     def find(self, name):
    Richard>         return(allMilkmen[name])

I think what you are looking for here is

    class Milkman:

        allMilkmen = {}
    
        def __init__(self, name):
            ...
            MilkMan.allMilkmen[name] = self
    
        def find(self, name):
            return(Milkman.allMilkmen[name])

(That is, you need to reference the allMilkmen attribute via the class.  It
doesn't stand alone.)

    Richard> Firstly, I'm curious if this is well-known, useful or
    Richard> boneheaded design on my part.

It's useful in certain circumstances but does create circular references
between the class and its instances that you will have to break explicitly.
Instead of a Milkman instance just disappearing when it goes out of scope,
you'll have to define (and call) a destroy method that executes:

    del Milkman.allMilkmen[name]

    Richard> Secondly, it doesn't work in Python, of course, because I need
    Richard> a Milkman to find all the others (i.e. Milkman.find['Phil']
    Richard> doesn't work).  I can do something similar by taking allMilkmen
    Richard> and find to module scope; that seems kind of ugly, and worries
    Richard> me that I'm abusing the language. Can I do better?

How about:

    Milkman().find("Bob")

?  If you do that, modify __init__ to be

    def __init__(self, name=None):
	...
	if name is not None:
	    MilkMan.allMilkmen[name] = self

to avoid creating and having to destroy the circular reference.

    Richard> Thirdly, is there some neat way that I can find all members of
    Richard> a class by inquiring of the Python internals?

That would require classes to maintain references to all their instances
behind the scenes.  That, in turn, would generate the same circular
references that you have in your code, but not give the user any way to
delete it.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098





From wolf at one.net  Thu Jan 20 19:49:26 2000
From: wolf at one.net (wolf at one.net)
Date: Thu, 20 Jan 2000 19:49:26 -0500
Subject: In A Pickle Over Pickle()
Message-ID: <sgaf8s4dtfq5mplbvuem1pgiscmmk9pp5n@4ax.com>

I'm using Python to create a text-adventure writing system, and need
to use pickle to save/restore the program state. The problem is, I
seem to have a mental block when it comes to Pickle--I've read all the
documentation I can find on pickle() and can't make heads or tails of
it. I suspect I'm missing some obvious "first clue" that would make it
come together.

Does anyone know where there's a good pickle() tutorial? Or, better
still, has somebody already written a module to do what I need?

The program structure is very simple, 3 library modules imported by a
"skeleton" that has no variables of its own. Basically what I want to
do is pickle/unpickle the *entire* mess, variables, dictionaries, etc,
of all three modules, and store them in a single file, preferably
calling wrapper code that resides in one of the 3 libraries.

Any help appreciated!

Respectfully,

Wolf

"The world is my home, it's just that some rooms are draftier than
others". -- Wolf


From gmcm at hypernet.com  Thu Jan 13 12:59:57 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 13 Jan 2000 12:59:57 -0500
Subject: Dynamic module/package import ???
In-Reply-To: <387E035F.69F91BFE@kw.igs.net>
Message-ID: <1264336097-5019135@hypernet.com>

jim wrote:

> I'm trying to import a module from a package dynamically. The layout is
> like so...
> 
> /DataHandler
>   |- DataRow.py
>   |- MyReader.py
>   |- __init__.py
> 
> In class DataRow.DataRow I try...
> 
>   def setReader( self, reader )
>     mod = __import__( 'DataHandler.' + reader + 'Reader' )
> 
> ... but this just loads the __init__.py, so it seems that the __import__
> doesn't understand packages or something.
> 
> Any help?

You need to use the fromlist parameter, as "print 
__import__.__doc__" will tell you.

mod = __import__( 'DataHandler.' + reader + 'Reader' , 
globals(), {}, [reader+'Reader'])

Actually, "nonempty" is the keyword. [1] works just as well.

- Gordon



From lemiss at my-deja.com  Tue Jan 11 04:35:56 2000
From: lemiss at my-deja.com (G. Lewis)
Date: Tue, 11 Jan 2000 09:35:56 GMT
Subject: starting a cgi with params
References: <iQp6OHyH9j2y64kYNrpGPcKoXKu0@4ax.com> <20000110133338.A15914@quark.emich.edu> <387a686e.3028554@nntpserver.swip.net>
Message-ID: <85ethr$ojj$1@nnrp1.deja.com>

In article <387a686e.3028554 at nntpserver.swip.net>,
  ame at swipnet.se (Anders Eriksson) wrote:
> I have looked in the 'manual' and found the cgi, but somehow I don't
> understand... I want to be able to call my script from a link, i.e.
> http://yourhostname/cgi-bin/cgi.py?xxx.zip&get

It is true, most CGI tutorials (not just in python) seem to focus on Form
examples, and completely skip explaining the basics of the mechanism.

Likewise i will also skip explaining, but will give an example without using
forms, based on your sample above; changed slightly to:

    http://yourhostname/cgi-bin/cgi.py?get=xxxx.zip

Now a fake interactive python session (obviously you can't really do an
interactive session submitting to a CGI script)!

>>>import cgi
>>>fs = cgi.FieldStorage()
>>>fs.has_key("get")
1
>>>fs['get'].value
'xxxx.zip'

If you want the "get" and "xxxx.zip" as seperate arguments you will have to
put in some bogus key values for them to equal... perhaps emulate argument
numbers:

    http://yourhostname/cgi-bin/cgi.py?1=get&2=xxxx.zip

>>>fs['1'].value
'get'
>>>fs['2'].value
'xxxx.zip'

Note that, as you already seem to know, you must seperate the key=value items
with & characters. There are also other rules for escaping characters in URLs
that you may want to look up. Notably space characters are encoded as +, and
+ and = and others are escaped with %xx hex codes.











The link format above is (almost) the same as an HTML form would send using
the GET method.


>
> The cgi class seems to be form getting info from a form(?)
>
> What am I missing??
>
> >
> >by the way, what platform are you using? linux + apache?
>
> To make everything easy. The developing platform is WinNT+Xitami and
> the target platform (where the script will run) is Solaris + Apache
>
> // Anders
> PS! This is my home account thats why the user changed!
>
>

--
...


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


From akuchlin at mems-exchange.org  Mon Jan 24 15:16:30 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 24 Jan 2000 15:16:30 -0500
Subject: OFFTOPIC: *-ly y'rs,
References: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com>
Message-ID: <3dvh4jz0m9.fsf@amarok.cnri.reston.va.us>

Agent Drek <drek at MonsterByMistake.Com> writes:
> Where did they come from and are they patented :)
> How can one join the signoff-ly club?

The first occurrence, at least in the python-list archives, was quite
a while ago:

  in-any-case-the-best-christmas-present-i-got-today!-ly y'rs - tim
	-- Tim Peters, 29 Dec 91 

I imagine Tim probably used this on other mailing lists before
python-list.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Why is a mouse when it spins?
    -- The Doctor, in "Robot"



 


From gerrit.holl at pobox.com  Wed Jan  5 07:51:16 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 5 Jan 2000 13:51:16 +0100
Subject: SUGGEST: os.path.validate()?
Message-ID: <20000105135116.A2974@stopcontact.palga.uucp>

Hello,

I've got a suggestion: what about a function os.path.validate()?
For posix, it's easy:
def validate(p):
    return not '\0' in p

but for Dos/Windows, it's harder, and I've no idea what's a valid
path on Mac.

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From robin at jessikat.demon.co.uk  Mon Jan 24 07:03:26 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 24 Jan 2000 12:03:26 +0000
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com>
Message-ID: <$WXiIAAO+Dj4Ew2W@jessikat.demon.co.uk>

In article <388BE68E.B26985CA at tismer.com>, Christian Tismer <tismer at tismer.com> writes

I'm very impressed. here are the pybench timing differences (absolute and expressed as % of
original python) for my 333Mhz win32 machine. here negative is bad for stackless, positive is
good. My only disappointment is that Zope wouldn't start up.

Tests:                   per run  per op.     oh  per run  per op.  oh  
BuiltinFunctionCalls:    -2.03ms  -0.02us 0.38ms  -0.45%ms -0.56%us 31.40%ms
BuiltinMethodLookup:     92.94ms  0.18us  0.91ms  18.43%ms 18.75%us 32.85%ms
ConcatStrings:           45.03ms  0.3us   0.66ms  6.74%ms  6.73%us  7.23%ms
CreateInstances:         -56.32ms -1.34us 0.74ms  -7.84%ms -7.84%us 30.08%ms
CreateStringsWithConcat: 7.93ms   0.04us  1.04ms  1.70%ms  1.72%us  32.91%ms
DictCreation:            30.65ms  0.2us   0.77ms  5.31%ms  5.19%us  32.35%ms
ForLoops:                193.48ms 19.35us 0.1 ms  30.67%ms 30.67%us 16.39%ms
IfThenElse:              98.93ms  0.14us  1.93ms  16.94%ms 16.09%us 32.66%ms
ListSlicing:             2.75ms   0.79us  -0.03ms 0.50%ms  0.50%us  -3.26%ms
NestedForLoops:          92.3ms   0.26us  0ms     24.95%ms 24.53%us 0.00%ms
NormalClassAttribute:    114.04ms 0.19us  1.28ms  22.48%ms 22.35%us 32.24%ms
NormalInstanceAttribute: 117.29ms 0.2us   1.29ms  23.72%ms 24.39%us 32.33%ms
PythonFunctionCalls:     -5.96ms  -0.04us 0.72ms  -1.11%ms -1.23%us 29.51%ms
PythonMethodCalls:       26.22ms  0.35us  0.26ms  6.08%ms  6.09%us  30.59%ms
Recursion:               -34.18ms -2.74us 0.64ms  -8.00%ms -8.01%us 32.16%ms
SecondImport:            -54.01ms -2.16us 0.28ms  -7.68%ms -7.68%us 34.57%ms
SecondPackageImport:     -45.16ms -1.81us 0.25ms  -6.35%ms -6.36%us 30.86%ms
SecondSubmoduleImport:   -53.17ms -2.13us 0.26ms  -6.47%ms -6.48%us 32.10%ms
SimpleComplexArithmetic: 13.27ms  0.06us  0.5 ms  2.36%ms  2.35%us  31.25%ms
SimpleDictManipulation:  36.49ms  0.12us  0.64ms  8.03%ms  7.89%us  32.16%ms
SimpleFloatArithmetic:   25.87ms  0.05us  1.28ms  5.86%ms  6.25%us  32.41%ms
SimpleIntFloatArithmetic:45.92ms  0.07us  1.55ms  10.55%ms 10.61%us 32.77%ms
SimpleIntegerArithmetic: 46.71ms  0.07us  1.53ms  10.74%ms 10.61%us 32.28%ms
SimpleListManipulation:  47.47ms  0.17us  0.75ms  9.37%ms  9.04%us  31.25%ms
SimpleLongArithmetic:    9.55ms   0.06us  0.37ms  2.20%ms  2.29%us  30.58%ms
SmallLists:              4.68ms   0.02us  1.14ms  0.57%ms  0.62%us  17.09%ms
SmallTuples:             -10.95ms -0.05us 1.02ms  -1.85%ms -2.02%us 32.28%ms
SpecialClassAttribute:   122.76ms 0.21us  1.28ms  24.24%ms 25.00%us 32.24%ms
SpecialInstanceAttribute:120.81ms 0.2us   1.29ms  21.62%ms 21.51%us 32.49%ms
StringSlicing:           -32.27ms -0.19us 1.33ms  -6.11%ms -6.29%us 28.60%ms
TryExcept:               166.99ms 0.11us  2.6 ms  19.57%ms 19.30%us 33.16%ms
TryRaiseExcept:          -34.71ms -2.32us 0.74ms  -5.89%ms -5.91%us 30.71%ms
TupleSlicing:            -124.35ms-1.18us 0.04ms  -21.64%ms-21.57%us4.04%ms
Average Round Time       1580.24ms                   7.73%ms             



Zope failure traceback
Traceback (innermost last):
  File "C:\Python\devel\Zope\z2.py", line 479, in ?
    exec "import "+MODULE in {}
  File "<string>", line 1, in ?
  File "C:\Python\devel\Zope\lib\python\Zope\__init__.py", line 135, in ?
    Globals.VersionNameName)
  File "C:\Python\devel\Zope\lib\python\ZODB\ZApplication.py", line 101, in __init__
    root=conn.root()
AttributeError: 'None' object has no attribute 'root'
C:\Python\devel\Zope>


>"Why should I care about this, isn't my Python fine"?
>
>If you know what Generators, Coroutines and Continuations
>are about and want to have them at your fingertips,
>this is the real thing for you.
>
>If not, how about speed?
>About 5 percent faster, binary compatible as a drop-in
>replacement for your python15.dll, unlimited recursion,
>why not give it a try? Just click on the installer :-)
>
>Stackless Python 1.01 with Continuations 0.6
>can now be downloaded from a web page:
>
>http://www.tismer.com/research/stackless/
>
>Fact sheet, links to documentation, source and
>binaries can be found there.
>
>Source and binary distribution have been split.
>The Windows binaries come with a Wise installer.
>
>SLP has been successfully compiled on Solaris (Thanks to Jeff
>Collins). It will be used on PalmIII, to solve the 4k stack problem.
>
>Version 1.01 is a small clean-up for non-Windows platforms.
>
><P><A HREF="http://www.tismer.com/research/stackless/">Stackless
>Python 1.01 + Continuations 0.6</A> - a version of Python 1.5.2 that
>does not need space on the C stack, and first-class callable
>continuation objects for Python.  (20-Jan-2000)
>
>Christian Tismer
>Mission Impossible 5oftware Team
>

-- 
Robin Becker


From news at dorb.com  Wed Jan 19 23:12:52 2000
From: news at dorb.com (Darrell)
Date: Wed, 19 Jan 2000 23:12:52 -0500
Subject: How to copy a binary file?
References: <3886876F.20FFEA27@erols.com>
Message-ID: <qUvh4.973$VD4.26969@typhoon.nyroc.rr.com>

> What is the preferred way of copying a binary file?

shutil.copyfile(src,dest)
or
open(dest,'w+b').write(open(src,'r+b').read())

--
--Darrell




From Alex.Martelli at think3.com  Fri Jan  7 12:33:54 2000
From: Alex.Martelli at think3.com (Alex Martelli)
Date: Fri, 7 Jan 2000 18:33:54 +0100 
Subject: lazy (?) evaluation, request for advice
Message-ID: <6D8A17398E28D3119F860090274DD7DB4B3DB1@pces.cadlab.it>

Tim Peters writes:

> >> Against that, you invented ad-hoc semantics as you went along;
> >> the only connection to Python appeared to be syntactic.
> 
> [Alex]
> > Did I?  I used, e.g., spades to mean "the value of a
> > variable named `spades' in the current scope[s]".  Isn't
> > that normal Python semantics? ...
> > I would appreciate understanding how you define "syntactic" in
> > this context to justify this extremely peculiar assertion.
> 
> The answer to "what's in a name?" is fundamental semantics for any
> language.
> Your language's answer is not Python's answer, hence my objection to
> calling
> the former Python.
> 
Here we go again.  Let's see if we can pinpoint this huge
difference of opinion -- I claim I'm looking for absolutely
nothing that is outside of the "fundamental semantics" of
Python, you claim there is a "my language" which differs.

Suppose, then, that my program accepts from the user some
text string, e.g. from a GUI or whatever -- say the user enters:
		
	(foo + bar) > baz and 3 < fee < 22

My program's role is to take this string (possibly compile
it), and eval it in a suitable dictionary (possibly more than
once, for different "suitable dictionaries"); doing one of
two different actions depending on whether eval returns
a true value in the Python sense.

Here, I may have a limited vocabulary, made up of the
variablenames: fee, fie, foo, fum, bar, baz; so my
life is easy:

	# code to obtain and possibly compile
	# user_expression omitted
	while 1:
		proceed_to_next_case()
		if no_more_alas():
			do_failure_sadly()
			break
		dict = {
			'fee': recompute_fee(),
			'fie': recompute_fie(),
			'foo': recompute_foo(),
			'fum': recompute_fum(),
			'bar': recompute_bar(),
			'baz': recompute_baz()
			   }
		if eval(user_expression, dict):
			do_success_triumph(dict)
			break

OK so far, I hope?  I'm doing absolutely nothing
non-Pythonic, am I?  This kind of thing IS just
about exactly why eval takes dictionary args, is
it not?  If you disagree, would you please point
out how this violates the "what's in a name" choices
of Python, and how eval and its dictionary args
are supposed to be used instead...?

Please suppose, in case it matters, that the
various recompute_fee(), etc, have no side
effects -- they do in fact nothing except
re-compute some things on the basis of the
state as changed by proceed_to_next_case();
they are 'pure' functions.

As I see it (and I'm a newbie, so I guess I could 
easily be wrong), "what's in a name" in Python
is: what is currently attached to that name in the
local scope, or failing that in the global scope,
or failing that in the builtins' scope; where such
attachments can and often do change when the
same code is evaluated over and over again in
some kind of loop.  When I eval or exec, I get
to control the scopes by passing dictionaries to
be used as local and global scope (or the same
dictionary for both) -- and this doesn't violate
Python's semantics, it's PART of Python's
semantics.


OK, so, suppose we DO agree that eval, and its
dictionary argument[s], ARE part of Python's
semantics (after all, they _are_ well documented
as part of the standard Python docs, aren't
they now...?); so that, in the above program,
what the user gets to do in that expression string
is use Python semantics (and syntax) to tell the
program what conditions, exactly, need to be
checked for [of course, an expression has
limited power, compared to other Python pieces
the user could be writing, such as a suite of
statements; but, very often such power will be
sufficient in this application domain, and for
cases where it isn't the program's interface
will provide alternative ways for the user to
enter statement suites or whatever -- this is not
the issue here, and has nothing to do with the
"what's in a name" question, does it?].


Now, all I'm asking is how I can do an optimization
(reduction of running time) that will change
NOTHING about how the program behaves,
from the point of view of the user who enters
his or her nice little string to be evaluated as
an expression -- except, hopefully, it may make
things faster, by avoiding useless computations.

If nothing changes except for speed, how can
what-used-to-be-Python's semantics suddenly
become "my own private and badly designed
language"'s...?


Specifically, the issue is that the vocabulary
gets larger, or maybe it doesn't but I notice
(thanks to a profiler) that recompute_fee()
takes a LONG time to run, so I would much
rather only call it when it IS needed -- which
is not all that often, as it happens, e.g.
because the "(foo + bar) > baz" portion of
the user-entered *Python* expression is
most often false, and thanks to *Python*'s
shortcut semantics this means that the value
of 'fee' is mostly not necessary.

I.e., I'd like to have lazy evaluation, to "coin"
a name.


I do not want to change the interface that the
user sees -- please assume that the users
like that interface and do not desire that it
changes... is it such a difficult assumption to
make?  I want the users to keep the exact
semantics and syntax seen above; I would
just like the program to be a bit (or a lot)
faster by avoiding needless calculations.

Now, please _explain_, at long last, why this
desire for optimization suddenly makes the
semantics non-Pythonic... assuming of
course that the premises were in fact agreed
(fat chance, I guess).
	

> > Where more than one hand is of interest, object notation will
> > be used; and where more power is needed, statements will be
> > used rather than just an expression.
> 
> The last wasn't (isn't) clear.  You described it as a "deal-selection
> function", but the example fragments showed only expressions.  If you are
> in
> fact going to treat isolated expressions as if they were functions, it's
> 
Sorry for expressing myself imprecisely, if I did.  "Deal-selection"
*expression* rather than *function* is no doubt what I should be
calling used-entered expressions such as the above example.

When the user feels that "just an expression" is not appropriate
(e.g., because he or she wants to compute partial results and
save them somewhere, etc), alternate ways for him or her to
supply Python code to be executed will also be provided.  But,
for now, I wanted to focus on the frequent case where pretty
simple Python expressions are in fact sufficient.


> again a sense in which your language is not (although may be related to)
> Python.  This means you're in the language-design business, as you went to
> 
Designing user interfaces, libraries, protocols for communication
between programs, etc, are ALL "language-design businesses" in
some quite reasonable sense.  If I have to design libraries, UI's,
protocols, and so on, I therefore have to design languages.

> great lengths in this reply to prove; I can only repeat my advice that
> Python is already well-designed and your users will be better served in
> the
> end by letting them use the language as-is; you don't agree, so that's
> that.
> 
I fully agree that Python is very well designed (or I would not be
here).  If by "letting them use the language as-is" you mean that
I should not provide any further interface or functionality besides
what is already IN Python, i.e. that I should not be writing any
further program at all, yes, then, I do disagree (and it would seem
to be a peculiar claim on your part).

I do not plan to hide the internals of the libraries/modules/classes/
call them as you will from users, if they do want to go that deep; I
do, however, plan to also offer the users other interfaces besides
a naked Python interpreter with some 'import mystuff'.

As soon as I do offer other interfaces, there are other 'languages'.
"Clic on that pushbutton does X" is not intrinsically part of Python's
semantics, so I'm "in the language design business" just as soon
as I have the simplest GUI with a pushbutton on it.

So, I'll have a textfield where (as one of the different optional ways
to express what it is they're looking for) users can enter a Python
expression (an expression in the language which you loudly claim
is not Python but rather "my own specially purpose-designed
language") -- if they do that then clic a certain pushbutton, the
expression gets (perhaps computed, then) evaluated as above.

Since all I do is a Python eval (after perhaps a Python compile),
with suitable argument[s] (as Python lets me give to eval, to be
used as scope[s] for it), I think your claim that this is "my own
special designed language" is absolutely preposterous.


> From what you've said so far, it doesn't *look* like anything fancier than
> a
> regexp replace is needed; e.g.,
	[snip]
Looks to me like that could be slow if the vocabulary is
pretty large.  E.g., hundreds of possible words, of which 
maybe 4 to 6 will typically be used in one expressions.

> transforming e.g. biglosers or heartsaches.  Generating a whole parse tree
> wouldn't work any better, *provided that* you don't have string literals
> in
> your language, and that the mapping to Python isn't hairier than simple
> token substitution.
> 
I have no intention of somehow forbidding users from having string
literals in the expression they can enter, although offhand I'm not
sure what use they could have for them in such an expression.

As for the "mapping to Python", none at all is needed, unless you
claim that eval is not part of Python, or is only part of Python if
its optional arguments are not used, or other such silliness; as I
hope I showed, were it not for the lazy evaluation idea, I could use
a simple eval after populating a dictionary appropriately.

Keep claiming to your heart's desire that this is "my own little
language" -- I fail to see what purpose such claims may fulfil,
but, hey, suit yourself.  Me, I'll just as relentlessly claim it is
Python, which I guess will be just as useless (except that I have
a strong aversion to see false assertions being uttered, and just
let them go, in a discussion I'm taking part in; so, the [limited]
use of my counter-claim is in easing my disgust at seeing such
falsehoods spewed -- with a small but finite probability that I will
learn something during this unpleasant process, such as, that
"eval" is NOT truly part of Python, it was just snuck in by a bad
witch's spell, or something).


> > I will appreciate further advice about this, and tolerate the
> > sneers and name-calling that seem to go with the advice
> > (although rarely have I seen discussion quite as sneering as
> > this on this group
> 
> Alex!  That wasn't sneering -- it was brazen ridicule <wink>.  Seriously,
> the exaggerated relentlessness of it all was supposed to nudge you away
> from
> taking at all soooooo seriously.  Didn't work, eh?
> 
Apparently not.  Do _you_ happen to know many "people" who
take well to being "brazenly ridiculed" in a public forum, and
react by jolliness and mirth?  You must introduce me to some
of them sometimes: being somewhat of a Martian myself in
several respects, I am always interested in acquaintance with
other aliens who happen to be visiting this planet.


> > -- reminds me a lot of the ambiance on the Perl group -- this
> > one had so far seemed friendlier, though I guess that may have
> > been an artefact due to a frequentation started more recently).
> 
> Oh, this *is* much friendlier.  Over on c.l.p.m., they would have called
> *you* an idiot.  I consistently insulted your *users*.
> 
I guess that "accusing" me of using non-Python semantics
does not count as an "insult", strictly speaking.  Maybe I'm
just being more newbye-y than usual, rather than really thick,
in failing to see how eval-with-dictionary-arguments can mean
"non-Python semantics".  So, please do elucidate.


Alex







From pinard at iro.umontreal.ca  Wed Jan 19 21:25:20 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 19 Jan 2000 21:25:20 -0500
Subject: Python Rocks!
In-Reply-To: talon@coursen.ifu.net's message of "Thu, 20 Jan 2000 00:56:45 GMT"
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com> <slrn88cnb6.4i6.talon@coursen.ifu.net>
Message-ID: <oqzou1v5rj.fsf@titan.progiciels-bpi.ca>

talon at coursen.ifu.net (Stephen Coursen) writes:

> On Wed, 19 Jan 2000 16:43:43 -0800, tye4 <tye4 at yahoo.com> wrote:
> >
> >Ivan Van Laningham <ivanlan at callware.com> wrote in message >
> >
> >> Don't code that way to start with.  It's dumb.  Treat yourself to a good
> >> editor.
> >>
> >Better yet. Fix the dumb language.
> >
> As a confused newbie, what gives you the experience to declare Python to be
> a "dumb" language?

Oh, you just misread `tye4'.  He was referring to the very language he
uses when he wrote his messages to the list, and hopefully beginning to
ponder he should fix himself, before fixing his editor :-) :-).

P.S. - I begin to think he is trolling us, and rather successfully! :-)
OK, enough said.  Let's switch to something more productive...

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





From news at dorb.com  Sat Jan  1 17:58:09 2000
From: news at dorb.com (Darrell)
Date: Sat, 1 Jan 2000 17:58:09 -0500
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg>
Message-ID: <FBvb4.1013$Sr1.114800@typhoon.nyroc.rr.com>

[Ajith Prasad]
> http://www.oreilly.com/news/vhll_1299.html is an article by
> Greg Wilson

>From the article, "I want to see a VHLL defined by an XML DTD."

Does Zope's DTML qualify as an XML language ?

I just finished a project that defined a text processing language with XML
tags. It works very well and everyone is happy. Of course now I need a new
Python job :)

On the down side it gets pretty verbose to write all those tags with out a
GUI. Although the users don't seem to mind.

Parsing was cake and being able to add defaultable attributes later that
didn't break existing code was very nice. Forgiving of the amateur language
designer.

A macro facility was a must and popular demand had macros create new tags on
the fly. So we couldn't parse against a DTD. Not sure this was a great loss
compared with how well the macros worked out.

--
--Darrell







From garabik at melkor.dnp.fmph.uniba.sk.spam  Wed Jan 12 11:36:22 2000
From: garabik at melkor.dnp.fmph.uniba.sk.spam (Radovan Garabik)
Date: 12 Jan 2000 16:36:22 GMT
Subject: PhotoImage.zoom problem
Message-ID: <947694914.694045@cdwork.cvt.stuba.sk>

I am trying to create some canvas image items, using zoom method.
All goes well, until I try to zoom pixmap containing transparent 
color. Pixmap is zoomed properly, but the transparent color is
turned into black. The same goes for xpm and gifs, and for subsample
method too.

This is sample of the code:
pixmap = PhotoImage(file="b.xpm").zoom(2)
item = canvas.create_image(x, y, image=pixmap)

Am I missing something or is this a bug in TkInter?

-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From akuchlin at mems-exchange.org  Sun Jan  2 13:59:15 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 02 Jan 2000 13:59:15 -0500
Subject: Finding where Python is on a Unix server through Telnet?
References: <386EA0EF.F7A72441@jslove.net> <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com> <84ncfb$e5q$0@216.39.151.169>
Message-ID: <3dn1qol2po.fsf@amarok.cnri.reston.va.us>

Donn Cave <donn at oz.net> writes:
> "locate" is a new one on me, not available on Digital UNIX.

It's a GNU program, so probably most common on the free Unixes.  A
cron job runs a script that builds an index of all the files on the
machine, and the "locate" command then looks through this index.
Source at ftp.mkintraweb.com/pub/linux/slocate/.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Guards! Guards! Stop this madman! He's turning everyone into monkeys!
    -- A sudden intrusion, in ZOT! #1


From dnichols at sourcesignal.com  Tue Jan 18 21:43:59 2000
From: dnichols at sourcesignal.com (David Nichols)
Date: Tue, 18 Jan 2000 18:43:59 -0800
Subject: Tcl/Tk Path problems in Python 1.5.2 installation
References: <387F80E5.A0AF82A5@sourcesignal.com>
Message-ID: <388524EF.241C037E@sourcesignal.com>

Thanks to those who replied ... the problem turned out to be old (1997!)
version of
tcl80.dll and tk80.dll in the windows/system directory.  Once those were
removed
everything went well.  I was initially deceived because the wish shell
launched
correctly, but that turned out to be an artifact of its initial directory -
wish found
the correct dlls but programs launched from arbitrary places linked to the
old ones
and then died.  So ... Beware that old junk hanging out in windows/system.

Now, back to the task of Python-ing.

David Nichols wrote:

> I just installed python 1.5.2 (running py152.exe and answering 'yes - I
> want to install Tcl/Tk') and have been unable to get the path set
> up correctly for graphical applications.  For instance, if I go to the
> idle directory (//E/Programs/Py152/Tools/idle) and try to launch it I
> get the following result.
>
> ============================================================
> BASH.EXE-2.02$ python idle.py
> Traceback (innermost last):
>   File "idle.py", line 3, in ?
>     PyShell.main()
>   File "E:\Programs\Py152\Tools\idle\PyShell.py", line 611, in main
>     root = Tk()
>   File "E:\Programs\Py152\Lib\lib-tk\Tkinter.py", line 886, in __init__
>     self.tk = _tkinter.create(screenName, baseName, className)
> TclError: Can't find a usable init.tcl in the following directories:
>     {} ./lib/tcl8.0 . E:/Programs/Py152/tcl8.0/library
> E:/Programs/Py152/Tools/library
>
> This probably means that Tcl wasn't installed properly.
> ============================================================
>
> The directories in sys.path are as follows ...
> E:\PROGRAMS\PY152
> E:\PROGRAMS\PY152\DLLs
> E:\PROGRAMS\PY152\PIL
> E:\PROGRAMS\PY152\lib
> E:\PROGRAMS\PY152\lib\lib-tk
> E:\PROGRAMS\PY152\lib\plat-win
> E:\Programs\Py152\DLLs
> E:\Programs\Py152\Lib
> E:\Programs\Py152\Lib\lib-tk
> E:\Programs\Py152\Lib\plat-win
>
> Tcl in installed in E:\PROGRAMS\Py152\\Tcl, and init.tcl exists in
> E:\PROGRAMS\Py152\Tcl\lib\tcl8.0 ... but this directory does not
> seem to be searched.
>
> Do I need to set additional environmental variables or make changes
> to the registry beyond those made by the installer?
>
> Thanks in advance for any help
>
> David Nichols
> Source Signal Imaging
> dnichols at sourcesignal.com





From amused at webamused.com  Thu Jan 20 22:48:17 2000
From: amused at webamused.com (Joshua Macy)
Date: Fri, 21 Jan 2000 03:48:17 GMT
Subject: regex question
References: <roy-835F11.21172620012000@netnews.nyu.edu>
Message-ID: <3887D4DB.20C06C75@webamused.com>

Roy Smith wrote:
> 
> This isn't really a python question per-se, but it came up while using
> the re module, so there is a connection :-)
> 
> Let's say I've got a pattern "[0-9]", i.e. a 1-digit number.  I want to
> search a string for that pattern, and only return a match if it's found
> preceeded by either a non-digit or the beginning of the string, and
> likewise followed by either a non-digit or the end of the string.  Is
> there a single compact regex I can write to cover that?
> 
> 

A literal translation of your 1 digit problem might look like the
following (using the \d shorthand for [0-9] and \D for [^0-9]:

import re
myRe = re.compile(r'(\D|^)(\d)(\D|$)')

testList = ["This is the number 1, isn't it?", "2 ought to be found",
"We ought to find 3", "But shouldn't find 12",  "12 should not be
found"]
for test in testList:
    m = myRe.search(test)
    if m:
	print m.group(2)
    else:
	print "Not found"

Which produces:
1
2
3
Not found
Not found

If you know you're looking for exactly 12 digits, you can use \d{12} in
place of \d above, e.g.
print "Test2"
testList2 = ["This is the number 123456789012, isn't it?", "123456789012
ought to be found", "We ought to find 123456789012", "But shouldn't find
12",  "12 should not be found"]
myRe2 = re.compile(r'(^|\D)(\d{12})(\D|$)')
for test in testList2:
    m = myRe2.search(test)
    if m:
	print m.group(2)
    else:
	print "Not found"

which prints:
123456789012
123456789012
123456789012
Not found
Not found

It's helpful to know as much as possible about the actual text that
you're going to run the regex against, since there are often shortcuts
you can take (like the \d \D thing) if you know that you'll never
actually see a-fA-F, or looking for whitespace or word boundaries
instead of the negation of the character set.

Joshua


From moshez at math.huji.ac.il  Mon Jan 31 19:58:16 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Tue, 1 Feb 2000 02:58:16 +0200 (IST)
Subject: nesting class definitions?
In-Reply-To: <slrn891nqh.bt0.neelk@brick.cswv.com>
Message-ID: <Pine.SOL.3.96.1000201024451.23158A-100000@sundial>

On 28 Jan 2000, Neel Krishnaswami wrote:

<about inner classes>
> (Bar, Thud) is just an ordinary Python tuple -- there's nothing magic
> about them, and as long as it is tuple of classes Python is perfectly
> happy. Now, in languages with multiple inheritance, there is a style
> of defining classes called "mixins" -- you create a set of classes,
> each containing a separate capability, and then to create a class with
> all of those capabilities you define a new class that inherits from
> each mixin. (These classes are "mixed in", hence the name.)

As an idionsyncristic side note, allow me to note that I feel using mixin
classes to be a bit unclean OO design methodology. I'd prefer doing the
object method calling mechanism myself, by using composition rather then
inheritance (for example, through proxying). The only solution to using
mixin classes are places when the API is already designed, and is closed
enough so the only place one can "get through" is through inheritance. 
Even then, I'd prefer a single mixin which opened up the API through
composition.

Here's a simple way to use composition rather then inheritance:

class spam:

	def __init__(self, helpers):
		self.helpers = helpers
		self.num = num

	def use_helpers(self):
		for helper in self.helpers:
			f = getattr(helper, 'do', None)
			if f is not None:
				return f(self)

class helper1:

	def do(self, spam):
		print spam.num

class helper2:

	def do(self, spam):
		print spam.num*2

Now you can have

spam(5, helper1)
spam(6, helper2)

and even (though pointless in this example and valuable if there was
more then one method):

spam(7, helper1, helper2)

That way the method stacking is up to you, not to rather obscure (though
well defined) Python language rules.
--
Moshe Zadka <mzadka at geocities.com>. 
INTERNET: Learn what you know.
Share what you don't.




From rickhigh at aol.com  Tue Jan  4 04:41:27 2000
From: rickhigh at aol.com (RICKHIGH)
Date: 04 Jan 2000 09:41:27 GMT
Subject: Quick Python v.s. Learning Python
References: <387152EB.10DD0E17@Lextek.com>
Message-ID: <20000104044127.13218.00000286@ng-fk1.aol.com>

There is another Python book coming out in  March 2000. The book focuses on
JPython, but has 8 chapters geared towards Python newbies.

It covers, RE, functional programming, OO in Python, working with Python files
with Pickle, etc.

It also has some chapters on Java Applets, Java Servlets, JDBC/SQL., etc.

The book is going to be published by Addison Wesley.




From gmcm at hypernet.com  Fri Jan 21 10:03:01 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 21 Jan 2000 10:03:01 -0500
Subject: Whitespace delimiters suck
In-Reply-To: <roy-3D90FF.09510421012000@netnews.nyu.edu>
Message-ID: <1263655511-14858205@hypernet.com>

Roy Smith wrote:

> That's a far cry from confusion about how many spaces a tab is equal to leading 
> to syntacticaly correct but semanticaly incorrect programs.

There is no confusion. Python always treats a tab as 8, (or 
aligning to the next n*8th column). tabnanny will tell you if 
there's any other value of tab which would yield an 
inconsistent interpretation.
 
> Even SQL, which has to be the most wart-infested language syntax I've ever seen, 
> gets this right :-)

I've seen SQL implementations which give you a syntax error if 
the last statement in a chain ends with a ";". I don't believe the 
SQL grammar specifies whether this is correct or incorrect.


- Gordon



From mfletch at tpresence.com  Thu Jan  6 06:53:40 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Thu, 6 Jan 2000 06:53:40 -0500 
Subject: lazy (?) evaluation, request for advice
Message-ID: <CC59868A0A76D311B1F50090277B321C1F300E@VRTMAIL>

Minor note: Tim is actually one of the most helpful people on the group.  He
has a quirky sense of humour sometimes.  Instead of reading "sneer", read
"playful banter among assumed friends and colleagues".  Tim tends to play
the "crotchety old man" role quite a lot for its entertainment value. It's
not a slight on you, it's just part of his persona in this group.

Enjoy yourself,
Mike(y)




From darrell at dorb.com  Thu Jan  6 21:43:05 2000
From: darrell at dorb.com (Darrell)
Date: Thu, 6 Jan 2000 21:43:05 -0500
Subject: XML to dictionary?
References: <FGALACFJFLICBBAA@my-deja.com>
Message-ID: <01b901bf58b8$fadac010$0100a8c0@rochester.rr.com>

Here's my own hack at tag parsing http://www.dorb.com/darrell/xmlP/
With wxPython it gives you a tree view of your tags. Replace the branches
with dictionaries and away you go.

--Darrell

From: "John Jensen" <johnjensen at my-deja.com>
> I'm thinking that the best thing to do in this situation is just write a
quick customized script, and not worry about using "fancy" libraries like
xmllib, pyxie, etc.





From pinard at iro.umontreal.ca  Mon Jan 17 14:06:13 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Jan 2000 14:06:13 -0500
Subject: Pardon my ignorance, data conversion
In-Reply-To: aahz@netcom.com's message of "17 Jan 2000 04:38:52 GMT"
References: <38829357.ED633932@home.com> <85u6cs$85d$1@nntp9.atl.mindspring.net>
Message-ID: <oqpuv04ixm.fsf@titan.progiciels-bpi.ca>

aahz at netcom.com (Aahz Maruch) writes:

> Robert W. McGwier <rwmcgwier at home.com> wrote:

> >I need to do the simple task of taking a nonnegative floating point
> >variable and taking its integer part and storing it in an integer.
> >Damned if I can find how to do it.

> import math
> i = int(math.floor(real))

Would not a mere:

   integer_value = int(real_value)

be sufficient?

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





From shang.remove_edu at st.jyu.fi.edu  Mon Jan 24 09:24:33 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Mon, 24 Jan 2000 16:24:33 +0200
Subject: Living without static member functions
References: <XFMail.000124133458.mikael@isy.liu.se>
Message-ID: <86hnac$r83$1@mordred.cc.jyu.fi>

On 24-Jan-00 Richard Brodie wrote:

<snip>

>  Secondly, it doesn't work in Python, of course, because I need a
>  Milkman to find all the others (i.e. Milkman.find['Phil'] doesn't
>  work). I can do something similar by taking allMilkmen and find to
>  module scope; that seems kind of ugly, and worries me that I'm
>  abusing the language. Can I do better?

Well, you don't have to take allMilken out of the class scope, just
the find function. i.e. I would write your class like this:

class Milkman:
    allMilkmen = {}
    def __init__(self,name):
        Milkman.allMilkmen[name] = self

def Milkman_find(name):
    return Milkman.allMilkmen[name]

>  Thirdly, is there some neat way that I can find all members of a
>  class by inquiring of the Python internals?

>>> from test import Milkman
>>> dir(Milkman)
['__doc__', '__init__', '__module__', 'allMilkmen']


Mikael Olofsson <mikael at isy.liu.se> wrote in message
news:XFMail.000124133458.mikael at isy.liu.se...

<snip>

> a=Milkman('Mikael')
> b=Milkman('David')
> a=Milkman('Mattias')
>
> then allMilkmen would be
>
> {'Mikael': <__main__.Milkman instance at e0f68>,
> 'David': <__main__.Milkman instance at e0de8>,
> 'Mattias': <__main__.Milkman instance at e0aa0>}
>
> Hence, allMilkmen still keeps a reference to the first instance
> (Mikael), which has been replace by the third (Mattias).

Because objects use reference semantics, Mattias does NOT replace
Mikael. The only thing changed is the local variable 'a' which now
referes to another object. If you do not intend to keep the objects in
the dictionary after they are not referenced from anywhere outside the
dictionary, you can add the following function:

def Milkman_clear():
    from sys import getrefcount
    dict = Milkman.allMilkmen
    for k in dict.keys():
        # if the number of references is two (one from
        # Milkman.allMilkmen and the other because the
        # object is the parameter to the getrefcount
        # function), remove the object from the dictionary
        if getrefcount(dict[k]) == 2:
            del dict[k]

> Perhaps you could make it more obvious by declaring it as global
> inside the class:

NOTE: The below code does not declare allMilkmen as global inside the
class, but as global inside the module.

> class Milkman:
>     global allMilkmen
>     allMilkmen = {}
>     def __init__(self, name):
>         allMilkmen[name] = self
>     def find(self, name):
 >         return(allMilkmen[name])

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi

Get paid to surf the Web!
http://www.alladvantage.com/home.asp?refid=BOT021




From anders.eriksson at morateknikutveckling.se  Thu Jan 13 11:20:14 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 13 Jan 2000 17:20:14 +0100
Subject: Cookies
References: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com> <Pine.LNX.4.21.0001131516440.18609-100000@fep132.fep.ru>
Message-ID: <I=p9OJtYnt3u+UxxIZ0WBCMmSX4=@4ax.com>

On Thu, 13 Jan 2000 15:17:35 +0000 (GMT), Oleg Broytmann
<phd at phd.russ.ru> wrote:

>> import os, Cookie
>> 
>> kaka = Cookie.Cookie(os.environ["HTTP_COOKIE"])

Then I get a KeyError: HTTP_COOKIE

Which probably means that the HTTP_COOKIE environment variable doesn't
exist!?

Any more ideas?

// Anders

PS! I have tried the script on a WinNT+Xitami and a Solaris+Apache
Both failed

---
Enlish isn't my first or second language, 
so anything that you find offending is not there on 
purpose it's just due to the translation ;-)


From mwh21 at cam.ac.uk  Sat Jan 22 17:26:01 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jan 2000 22:26:01 +0000
Subject: Suggestions for python 2
References: <000501bf623e$18b56a20$9a2d153f@tim> <3.0.5.32.20000122162257.009af7f0@mail.dicksonstreet.com>
Message-ID: <m3ya9hvj46.fsf@atrus.jesus.cam.ac.uk>

Felix Thibault <felixt at dicksonstreet.com> writes:

> >In article <000501bf623e$18b56a20$9a2d153f at tim>,
> >Tim Peters <tim_one at email.msn.com> wrote:
> >>[Evan Simpson]
> >>>
> >>> I rather like the idea of having to explicitly bring names
> >>> into an inner scope,
> >>
> >>Me too -- except in the cases that bite most often, where a trivial little
> >>lambda wants to refer to one or two measly little names from its context.
> >>People stumble on that so often it's a FAQ.
> >
> 
> How hard would it be to implement something like:
> 
> def like_a_list(*now_args):
>     def return_func(!now_args, *later_args, **show):
>         if show:
>             return list(now_args+later_args)
>         else:
>             return apply(like_a_list, now_args + later_args)
>     return return_func
> 
> where the ! (or whatever) means something like this happens:
> 
> def like_a_list(*now_args):
>     """like_a_list(value0, value1,...) store some values in a function"""
>     fun_str = """def return_func(*later_args, **show):
> 	now_args = %s
> 	if show:
> 		return list(now_args+later_args)
> 	else:
> 		return apply(like_a_list, now_args + later_args)\n"""\
> 		 % (now_args,)
>     exec(fun_str)
>     return return_func
> 
> when the inner function is defined ? (Except that it works when
> eval(repr(object)) doesn't and it's not as ugly). 

I don't really understand what you are trying to do, but you are
invited to check out xapply.py in my bytecodehacks package which can
be found at

http://starship.python.net/crew/mwh/

and is documented here:

http://starship.python.net/crew/mwh/bch/module-bytecodehacks.xapply.html

If that isn't what you're after, can you post an example of what you
might want to do with this hypothetical function? I can probably whip
something up for you.

Cheers,
Michael


From kc5tja at garnet.armored.net  Mon Jan 10 16:24:48 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 21:24:48 GMT
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu> <slrn87fol9.15t.kc5tja@garnet.armored.net> <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk> <slrn87imad.thc.wtanksle@hawking.armored.net> <slrn87j11s.vc0.kc5tja@garnet.armored.net> <Y$u7SGAdYae4EwJv@jessikat.demon.co.uk> <387A04AC.886DD88C@bellatlantic.net> <T2FFKFAaCie4EwJn@jessikat.demon.co.uk>
Message-ID: <slrn87kjtf.930.kc5tja@garnet.armored.net>

In article <T2FFKFAaCie4EwJn at jessikat.demon.co.uk>, Robin Becker wrote:
>I'm sorry if I sounded too defensive. Perhaps I should regard Sam's post
>as being enthusiastic rather than aggressive. Telling us we're in the
>gutter of standard OO won't endear though.

Of course not -- the momentum behind the current OO model is just too great.
And it DOES have a large applicability.  It's just not suitable for use in a
distributed object/component environment composed of heterogeneous
programming languages. :)

I will say that, upon re-reading my message, I did tend to use some pretty
strong verbiage.  English, despite being my native (and only) language, is
NOT one of my strong points.  However, C and assembly are.  :)

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From gerrit.holl at pobox.com  Fri Jan 14 07:14:06 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 14 Jan 2000 13:14:06 +0100
Subject: Wanneer is nu?
In-Reply-To: <20000113155245.D1928@stopcontact.palga.uucp>; from gerrit.holl@pobox.com on Thu, Jan 13, 2000 at 03:52:45PM +0100
References: <20000111104436.A4538@dnd.utwente.nl> <200001112348.AAA00993@esperanto.hetnet.nl> <20000113155245.D1928@stopcontact.palga.uucp>
Message-ID: <20000114131406.A341@stopcontact.palga.uucp>

Gerrit Holl wrote on 947775165:
> Johan.Swenker at hetnet.NL wrote on 947634539:
> > werkdag: maandag t/m vrijdag
> 
> Heet dat in Onze Taal niet een heel klein beetje een "werkweek"?
>                                                           ~~~~

Sorry, wrong list!

-- 
Please correct any bad English you encounter in my mail!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From ttnttn at toad.scripps  Thu Jan 13 06:42:13 2000
From: ttnttn at toad.scripps (Thien-Thi Nguyen)
Date: 13 Jan 2000 03:42:13 -0800
Subject: inverse of __repr__ ?
Message-ID: <yqtn1qa8aga.fsf@toad.scripps>

i've looked through _programming python_ for the inverse of __repr__, but
could not find anything.  perhaps i'm missing some clue?

	read : write
	??   : __repr__

or is there no way to extend the reader in a python repl?

thanks,
thi


From a.eyre at optichrome.com  Mon Jan 24 13:05:26 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 24 Jan 2000 18:05:26 -0000
Subject: Embedding -- getting error/traceback information not to stderr?
In-Reply-To: <slrn88oujk.ebi.erw@erw-pc.dde.dk>
Message-ID: <000101bf6695$977d6140$3acbd9c2@peridot.optichrome.com>

>  1) copy PyErr_Print(), modify it to append to a file (yuck!)
>  2) Modify sys.stderr to append to an internal buffer instead of a file
3) Set stderr to be stdout

#!/usr/bin/env python
import sys
sys.stderr = sys.stdout

If you need to set it back, use:

sys.stderr = sys.__stderr__

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From w_keller at gmx.de  Sun Jan  9 12:13:03 2000
From: w_keller at gmx.de (Wolfgang Keller)
Date: Sun, 9 Jan 2000 18:13:03 +0100
Subject: French references?
References: <853agi$2hh7$1@news5.isdnet.net>
Message-ID: <01HW.B49E802E0008C7D80304709C@personalnews.de.uu.net>

On Fri, 7 Jan 2000 1:02:50 +0100, Yann SAMAMA wrote
(in message <853agi$2hh7$1 at news5.isdnet.net>):

> unless I give him some well-known french references... 

There's a french-speaking Python mailing list. Some information should be 
available at http://www.aful.org/mailman/listinfo/python. If you gave more 
information about what the company you work for is doing/wants to do with 
Python, it might even possible to get some more application-specific 
references from the corresponding SIGs.

Regards,


-- 
Wolfgang Keller               

Zu Risiken und Nebenwirkungen von Junkmail lesen Sie 
de.admin.net-abuse.mail und fragen sie Ihren Postmaster oder Provider



From thor at localhost.localdomain  Sun Jan 30 06:05:01 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 30 Jan 2000 11:05:01 GMT
Subject: Python Ring Info
Message-ID: <slrn898a7p.mg.thor@localhost.localdomain>

Python ring Info
-------------------------

This is a piece of information about the existing python ring in:
http://www.ctv.es/USERS/irmina/pythonring.html
(now with Statistics)

What is it?
>From a functional point of view, it's a list of linked web sites
that share a common interest or information, in this case python
related stuff.

How does it work?
The HTML code included in each node of the ring calls a CGI that
performs the movement thru the ring.

How can I join?
In the above web site , fill in your data and that's all.

Why I should ? 
If you've done a web site, I think it's worth you spend 15 minutes
promoting it, after your web site has been included in it you'll get
more hits than ever.
Moreover, your information may be interesting for other people, or
you can be interested in other people's work, a big Webring provides
the community a lot of unusual but somehow interesting data.

How can I get more hits? 
Because the Webring is a democratic way of organization, that is,
all its members tend to receive the same ammount of visits, the
big ones as much as the small ones.

Well, I don't have a Web page.., so?
So  you must (  :) ) start doing one right now. Seriously, we
( at least me,MGA) are interested in people's work, perhaps you've
been collecting data or even your libraries or methods of work are
clearly defined, the ammount of time required for a fast compilation
of that info is not too much. And I remark that you can learn only
if all ( or many) want to show or teach what they know.

Of course we admit commercial Web sites also, *any* Web site about
python.

Thanks for your time and patience reading this.
MGA




From mlh at vier.idi.ntnu.no  Wed Jan  5 08:08:45 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 05 Jan 2000 14:08:45 +0100
Subject: Python ring propaganda
References: <slrn876erg.18q.thor@localhost.localdomain>
Message-ID: <iacbt70r7he.fsf@vier.idi.ntnu.no>

thor at localhost.localdomain (Manuel Gutierrez Algaba) writes:

> Python ring propaganda
> -------------------------
> 
> This is a piece of information about the existing (La)TeX ring in:
> http://www.ctv.es/USERS/irmina/pythonring.html
> 
> What is it?
> From a functional point of view, it's a list of linked web sites
> that share a common interest or information, in this case TeX
> related stuff.

This seems more like a TeX ring than a Python ring... (Is this a
copy/paste error in your posting? ;)

--

  Magnus
  Lie
  Hetland


From boud at rempt.xs4all.nl  Sun Jan  2 07:13:17 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 2 Jan 2000 12:13:17 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com>
Message-ID: <84nfct$bal$1@news1.xs4all.nl>

Simon Faulkner <News at titanic.co.uk> wrote:
> Hi All,

> Where can I find the instructions for the DB API?

> I have managed to figure out the .connect() and .fetchone() methods
> but am flagging with the rest.

> The Python Database API Specification 2.0 document does not realy show
> me the methods or examples that I need to get started and the MySQLdb
> documentation is deliberatly vague to discourage use of the _mysql.
> methods (so you use the proper API?)

If you're using to MySQL, then you'd better not go to far using all
the options and stuff from DB-API II - if you use connect, execute and
fetchall, you've exhausted the capabilities of the database. No need
for commit, rollback or things like that - MySQL doesn't support them.
There's one area where MySQL has functionality beyond DB-API II, and
that's in the area of autoincrement columns, where you can get quite
easily get back the autoincrement values back after insertion.

Also, with most^H^H^H^Hall database servers, what you really need to do is
ask for a few rows, and fetch them immediately from the server. Keeping
large resultsets on the server and nibbling piecemeal from them is a
recipe for performance problems.

-- 

Boudewijn Rempt  | http://denden.conlang.org


From gmcm at hypernet.com  Mon Jan 17 12:56:37 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 17 Jan 2000 12:56:37 -0500
Subject: Global help
In-Reply-To: <85vfbp$f3t@catapult.gatech.edu>
Message-ID: <1263990696-9667177@hypernet.com>

Shaun Allen Dishman writes:
> 
> I have something like the following situation:
> 
> ----- x.py -----
> 
> from y import *
> 
> def x(filename):
> 	file = open(filename, 'w')
> 	y()
> 
> ----------------
> 
> ----- y.py -----
> 
> def y():
> 	# must be the same file object as in x.py
> 	file.write('blah blah blah \n')	
> 
> ----------------
> 
> I have fiddled around with the global settings for the file object but
> cannot seem to get things to work.  All I need is to be able to access the
> file object from one module inside of another one, without passing it as a
> parameter.  Is this even possible?  Thanks in advance.

Pass it as a parameter anyway.

But if you really insist, you've got 2 choices: put the file in y's 
globals or put the file in x's globals.

To do the former, don't use "from y import *", just use "import 
y" and then "y.file = ... "

To do the latter, add "import x" to the top of y, and 
"x.file.write(...)".

However, if either one of these is __main__, it won't work; the 
non-__main__ module will have to "import __main__" and use 
"__main__.file..."

So give yourself a break and pass it as a parameter.


- Gordon



From wjk at wjk.mv.com  Wed Jan  5 00:25:54 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Wed, 05 Jan 2000 00:25:54 -0500
Subject: Importing a dicrionary from a file
References: <3872C4EC.20ADE046@trojanslair.zzn.com>
Message-ID: <3872D5E1.E7DA06E5@wjk.mv.com>

I am a relative newbie as well - I thought about this
before and had done something along similar lines
but not necessarily the best way:

Created a tab-delimited file (dict.txt) - with no blank lines:
--actually its a table--

a    data something a
b    data something b
c    data something c
d    data something d

then wrote program:

import string                               #import string module
fp = open("dict.txt","r")              #open file
data = fp.readlines()                   #whole file as one variable
new = {}                                     #created empty dictionary
foreach entry in data:                   #entry would be "a    data"
      stuff = string.split(entry,"\t")  #split by tab (now 2 pieces)
      new[stuff[0]] = stuff[1]          #1st part is key 2nd is data

fp.close() #close file

new["a"] would print out data

1. if your keys are numbers just convert the string to a number for
    dictionary key
2. if your file is going to be huge then I would seek an alternative way
3. make sure the separator in your file is a character not used anywhere
    else... ( Field separator in awk lingo me thinks)

I hope I got this right! -- There is probably another way such as just
typing or creating a module with your dictionary in it and using that.
[So I will let the expert Pythonistas help further..... ] (c;{>*

Black Mantis wrote:

> hello
> I am a newbie, and i am working on a program that uses a really big
> dictionary. i was wondering if i could store the dictionary in a
> seperate file, let's say dict.dat. Now, at the beggining of the program
> i want to import that dictionary and get it ready, how would i do this?
> is there a way to read the whole input file into one variable?
>
> thanks for your help
> keagan



Black Mantis wrote:

> hello
> I am a newbie, and i am working on a program that uses a really big
> dictionary. i was wondering if i could store the dictionary in a
> seperate file, let's say dict.dat. Now, at the beggining of the program
> i want to import that dictionary and get it ready, how would i do this?
> is there a way to read the whole input file into one variable?
>
> thanks for your help
> keagan



From jbauer at rubic.com  Thu Jan  6 20:31:33 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 06 Jan 2000 19:31:33 -0600
Subject: Now What? :o)
References: <n2bd4.64$%Ph.18874529@news.randori.com>
Message-ID: <387541F5.1606D7C6@rubic.com>

> I'll be working with the language on NT and LINUX boxes; 
> Are there any issues with writing code to run on both 
> platforms that I should be aware of?

James,

Python Unix/Win32 portability works pretty well for
many kinds of projects.  The Library Reference is your
best guide for portability issues.  Here are a couple
of "gotchas" I've noticed when trying to use other
people's code on mixed platforms:

1. Use os.sep rather than slashes or backslashes
for your path separators.

2. Open binary files in binary mode.  Unix programmers
often neglect to use the binary switch since all Unix
files *are* binary.

It's often helpful to test portions of your code on
both platforms as the code is being developed, rather
than wait to test it when the code is completed.  If
you have portability issues, it's easier to isolate 
them early in the development process.

Jeff Bauer
Rubicon Research



From moshez at ladygrey.  Sat Jan  8 09:33:05 2000
From: moshez at ladygrey. (Moshe Zadka)
Date: 8 Jan 2000 14:33:05 GMT
Subject: Sydney Pythonistas/Pythoneers (was Re: Sydney scoreboard?)
References: <385DB48C.7A7A52D6@bby.com.au> <8On74.6414$Dh3.85256@ozemail.com.au> <YCU74.7437$NV6.31360@news-server.bigpond.net.au> <38601AF6.7D39F091@bby.com.au> <D0da4.3285$lv5.19947@news-server.bigpond.net.au> <3872CEA2.38C8E5A0@bby.com.au> <20000105233639.A1855@Ridcully.home>
Message-ID: <857hr1$rm4$1@news.huji.ac.il>

On Wed, 5 Jan 2000 23:36:39 +1100,
	Malcolm Tredinnick <malcolmt at smart.net.au> wrote:

>Let me volunteer to do something:
<organizing a Sydney pythonistas meeting>

Hmmm....sounds like a damn good idea, if I was in Sydney, rather then
Israel. So here is the Israely answer:

As per Malcolm's post, but re.sub('Sydney', 'Israel'), please e-mail
me personally, and I'll see what I can do.

-- 
Moshe Zadka -- <mzadka at geocities.com>
Hackers of the world unite! You have nothing to lose but 
software licenses!


From philh at vision25.demon.co.uk  Thu Jan 13 21:19:22 2000
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Fri, 14 Jan 00 02:19:22 GMT
Subject: os.system() and PHP?
References: <871z7lercw.fsf@ethereal.dhis.org>
Message-ID: <947816362snz@vision25.demon.co.uk>

In article <871z7lercw.fsf at ethereal.dhis.org>
           nolan at ethereal.dhis.org "Nolan Darilek" writes:
> The script is called, yet the os.system() in the Python script which
> calls mp3info doesn't work. I've inserted debugging prints into the
> script and have gotten output, and I printed the return code from
> os.system(), which is 256. The script works nicely from the command
> line; once I edit the track I can then call the script with the exact
> same command line and see the results I want.

When you run your Python script from the command line, is it running
as the same user that it runs as when it is invoked from PHP? Are
the environment variables the same?
 
> I'm thinking that perhaps PHP's passthru() prevents os.system() from
> running. Is there some sort of limitation on how many subshells a
> program can spawn?

IIRC Linux has a limit; it's quite high though. I imagine other
Unices are similar.


-- 
Phil Hunt - - - phil at comuno.com

 See Comuno's Linux coverage at <http://www.comuno.com/linux/> for:
* Links to useful sites * Linux-related news * Open source software *



From tjg at avalongroup.net  Fri Jan  7 13:02:10 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Fri, 07 Jan 2000 10:02:10 -0800
Subject: First try at Tkinter stuff
References: <3875819D.41ECB8BA@exceptionalminds.com> <m37lhm3uey.fsf@havenrock.com>
Message-ID: <38762A22.96C79903@exceptionalminds.com>

Matt Gushee wrote:

> Or you're just missing a small piece of the puzzle.
> 
> >         SupplierFrame.grid(row=0, column=0, rowspan=2)
> 
> Try this instead:
> 
>         SupplierFrame.grid(row=0, column=0, rowspan=2, sticky=N+S+E+W)
> 
> ... or in your case, 'sticky=N+S' might be enough.
> 
> If you were packing instead of gridding the SupplierFrame, you would
> use the options 'fill' and possibly 'expand':
> 
>         SupplierFrame.pack(expand=YES, fill=BOTH)


Well, it did help a bit as I now have a better understanding of what's
going on, but it didn't help in that the minute I put a child widget in
any of the frames, the frame shrinks to the width of the label or the
button I place in it, and the height of the two central frames stacked
together.

I did some more reading last night (Tkinter Life Preserver) and one of
the things Matt says is that containers shrink to the size of their
contents. Is that always the case, or is there a way around it?


-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From josht at iname.com  Fri Jan 14 22:33:07 2000
From: josht at iname.com (Josh Tompkins)
Date: Sat, 15 Jan 2000 03:33:07 GMT
Subject: Nested while-s
Message-ID: <8EBBEE15Ejoshtinamecom@209.208.0.7>

Is it not possible to nest while's?

Consider this code:

def listOfRadius(radius, xpos, ypos):
	list = []
	currentX = (xpos - radius)
	currentY = (ypos - radius)
	while currentX <= (xpos + radius):
	    while currentY <= (ypos + radius):
		    str1 = `currentX` + " , " + `currentY`
		    list.append(str1)
		    currentY = currentY + 1
	    currentX = currentX + 1
	return list

It will list only the values from the nested while.  It was supposed to 
increment the inner while until currentY <= (ypos + radius), and then 
increment the outer while (currentX), and then repeat the inner while's 
code, etc., etc., etc.  I'm not very good at explaining these things, so I 
hope you can see what I'm trying to do from the code.

Any help you can give me would be greatly appreciated.

TIA,

Josh "Josue" Tompkins
________________________________________________________________

"Destined For Great Things -- but pacing myself."
- From a t-shirt.

E-Mail:  josht at iname.com 
ICQ:  21219667 
Web:  http://www.crosswinds.net/~josht
_________________________________________________________________


From bitbucket at isomedia.com  Thu Jan  6 02:36:45 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Thu, 06 Jan 2000 07:36:45 GMT
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com>
Message-ID: <38744512.377982379@news.isomedia.com>

On Wed, 5 Jan 2000 23:52:46 -0500, "Darrell" <news at dorb.com> wrote:

>from filename import *
>
>Doesn't create another namespace. Avoid this in most cases.
>

What are the implications of your statement for Tkinter?  Is that one
of the exceptions?  I understand and respect the differences in what
happens to the namespaces, and I try to avoid 'from x import *', but
I'm wondering where the Python community buckles on this issue.

     -Eugene
import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From jjlucsy at concentric.net  Fri Jan 28 08:57:51 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Fri, 28 Jan 2000 08:57:51 -0500
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com> <86q5vv$mci$1@nnrp1.deja.com> <Qq2k4.61$RU4.2585@news4.usenetserver.com> <86qt6t$7ta$1@nnrp1.deja.com>
Message-ID: <fehk4.2029$CX6.142432@news4.usenetserver.com>

> I see where this technique will speed up the initial loading
> of Python,but you will loose time on all subsequent library calls. Is it
> not more efficient to have all the library functions available
> immediately.

It won't happen on ALL subsequent calls, just the first to use a particular
library. It saves on time when, like most programs I've been writing lately,
they don't even use those libraries. In particular, winsock and multimedia,
multimedia being the real culprit. If my program never makes a sound, then
why should I load the library?
So... I'm not saying everyone should be using delayload, or that the
standard distribution should use it. I'm justing providing an alternative
where individuals can determine it's own merit.

-
Joel Lucsy (jjlucsy at concentric.net)







From codeslave at worldnet.att.net  Sat Jan  8 10:05:11 2000
From: codeslave at worldnet.att.net (Eric Hausig)
Date: Sat, 08 Jan 2000 10:05:11 -0500
Subject: Error using string.strip
Message-ID: <38775227.2CC0D685@worldnet.att.net>

Hello there,

I'm very new to Python.  (In fact, I'm walking through the tuturial
right now!)

I'm getting the following message:

>>> string.strip('str') + 'ing'
Traceback (innermost last):
  File "<stdin>", line 1, in ?
NameError: string
>>>

I'm running Redhat 6.1 and have installed all of the dev rpms off the
CD.   Is there some library I have to manually include?  If someone
could give me a little insight, I would be thankful.  (PS - I'm
currently a VB/SQL programmer trying to make the switch to Linux.  =)

Sincerely,

Eric Hausig
aka Codeslave2k



From bsb at winnegan.de  Sat Jan  1 15:25:51 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 01 Jan 2000 21:25:51 +0100
Subject: pickle problem:cant reload object saved with binary option
In-Reply-To: "Darrell"'s message of "Sat, 1 Jan 2000 14:59:13 -0500"
References: <386e1de1.998928692@news-server> <YHpb4.914$Sr1.104202@typhoon.nyroc.rr.com> <386e31c2.1004019140@news-server> <386E4B0E.C9DD0BF@appliedbiometrics.com> <003001bf5492$ada0faf0$0100a8c0@rochester.rr.com>
Message-ID: <87aemp4jzk.fsf@baal.winnegan.de>

"Darrell" <darrell at dorb.com> writes:

> Out of curiosity I looked into this file mode thing. Sure beats cleaning out
> the garage !
> 
> Looks like the '+' allows read/write.
> >>> open('t','w').write('now is the time')
> >>> open('t','r+').write('xx')
> >>> print open('t','r').read()
> xxw is the time
> >>>
> For append you have to use 'a'
> 
> /////////////////////////////////
> gcc\libio\fileops.c function _IO_file_fopen

[...]

No need for UTSL, it's available in the python docs:

http://www.python.org/doc/current/lib/built-in-funcs.html

look for "open"

Thanks
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
****** ceterum censeo javascriptum esse restrictam *******



From ---  Tue Jan  4 10:53:00 2000
From: --- (---)
Date: Tue,  4 Jan 2000 19:53:00 +0400
Subject: Hi, its for you !
Message-ID: <37664630545567311@d203.p6.col.ru>

-- spam --



From P.J.W.S.VRIJLANDT at INT.azg.nl  Tue Jan  4 10:56:30 2000
From: P.J.W.S.VRIJLANDT at INT.azg.nl (P.J.W.S. Vrijlandt)
Date: Tue, 4 Jan 2000 16:56:30 +0100
Subject: pythonwin installation
Message-ID: <218A08577EF@nds-3.azg.nl>

Hi all,

When installing pythonwin (from win32all-125) I first saw this 
message about 3 times. But pythonwin did install and I can start it.

It seems I can't do anything like using running makepy &c.
This is what happens:

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit 
 (Intel)] on win32 Copyright 1991-1995 Stichting Mathematisch 
Centrum, Amsterdam Portions Copyright 1994-1999 Mark Hammond 
(MHammond at skippinet.com.au) 
>>> import pythoncom 
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
ImportError: DLL load failed: Een apparaat dat op het systeem is
aangesloten, werkt niet. 
>>> 

Since Dutch is your 2nd language (as in another thread), you all 
understand that the DLL does not load because "a device connected to 
the system, does not work". In the configuration screen everything 
seems fine.

Any ideas?

I'm running Win95 osr 2 on a pentium with 32 mb. Does pythonwin have 
any configuration requirements?
I used all the default installation options.
Python and Idle/Tkinter work fine.

TIA,

Patrick Vrijlandt



From python at commandprompt.com  Sun Jan  9 12:56:25 2000
From: python at commandprompt.com (python at commandprompt.com)
Date: Sun, 9 Jan 2000 09:56:25 -0800 (PST)
Subject: New Python Books
Message-ID: <Pine.LNX.4.10.10001090941020.26842-100000@crazypenguins.commandprompt.com>

Hello,

I recently posted this message to comp.os.python:

Hello,
 
I represent LinuxPorts.Com (http://www.linuxports.com/) and I have been
approached by a printer to possibly publish OpenSource books. To start
these books would just be paper versions of the actual documentation. In
this case, the Python Documentation and we were wondering if this would be
of interest to the Python Community.
 
I spoke with the printer and we both have agreed that if we do this that
we will be providing a portion of the Gross Profits back to the individual
projects. In this case the Python project.
 
Anyway, if this of interest to you please go to
http://www.linuxports.com/ and make your vote. It will help us
determine the demand for such a book.
 
-----

We received about 8 responses with feedback that we much appreciated and
would now like to solicit some more. If the people
that read this email could just think about a few of the following
questions and provide some feedback we would be very thankful.

We are considering publishing the current set of Python documentation as a
single volume versus several volumes such as toExcel does. Would this be
of interest to the Python community?

One person made mention of the "Vaults of Parnassus" if we were to include
a snapshot of this would this be of interest?

We are also considering "paying" someone to write a book about Python.
Unfortunately, unlike traditional publishers the author would not get paid
until the book was done, and then it would be on the basis of per book
sold. We are aware that this is not the normal way of doing things, but we
are not Prentice Hall and this would be the only way "currently" that we
could pull this off. Is there anyone on this list that feels confident
enough in their Python skills and writing ability to be able to perform
such a task?

Last but definitely not least, no matter what we do, if we publish a
Python book, 20% to 40% of the Gross Profits would go back to the PSA.


Thanks for your time:

Joshua Drake
Webmaster - LinuxPorts.Com
WebMaster - LinuxDoc.org

 
 





From wjk at wjk.mv.com  Sun Jan  2 00:45:07 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Sun, 02 Jan 2000 00:45:07 -0500
Subject: Converting an intergar to a string
References: <19991231140532.A8331@dark.net> <082a01bf533f$a61d3400$0100a8c0@rochester.rr.com> <19991231164828.A8614@dark.net>
Message-ID: <386EE5E2.4E9F7B38@wjk.mv.com>

Hi: As I am relatively new to Python, I had asked basically the same
question some time ago... Here is what I found:

print "Some Text " + str(num) + "Some text"

Hope this helps.

Jeremy Lunn wrote:

> That fixed that problem.  Now would it be possiable to do the opposite
> and covert an interger to a string so that it can work like this:
> print 'some text'+num+'some text' ?
>
> Thanks,
>
> Jeremy
>
> On Thu, Dec 30, 1999 at 10:32:21PM -0500, Darrell wrote:
> > Try this.
> >
> > num = string.atoi( urlargs["num"].value)
> >
> --
> Jeremy Lunn
> Melbourne, Australia
> ICQ: 19255837



From jbauer at rubic.com  Tue Jan 18 11:16:24 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 18 Jan 2000 10:16:24 -0600
Subject: indentation woes? use emacs!
Message-ID: <388491D8.ADF9E08E@rubic.com>

A number of new Python programmers have posted
comments about Python's use of indentation.

While it may be a hassle to learn a new
language *and* and new editor, Python
mode in Emacs/XEmacs is truly wonderful.
I would encourage any newbie to consider 
investigating XEmacs.  I use it daily
on Solaris, Linux, and NT.

http://www.xemacs.org

Jeff Bauer
Rubicon Research



From darrell at dorb.com  Fri Jan 14 09:03:03 2000
From: darrell at dorb.com (Darrell)
Date: Fri, 14 Jan 2000 09:03:03 -0500
Subject: file width and height
References: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com> <ubqf4.3018$hN5.92483@typhoon.nyroc.rr.com> <004b01bf5e68$92965a40$f29b12c2@secret.pythonware.com>
Message-ID: <026a01bf5e98$12edfd90$0100a8c0@rochester.rr.com>

I screwed up here.
Sorry.

--Darrell
----- Original Message ----- 
From: "Fredrik Lundh" <fredrik at pythonware.com>
To: "Python Mailing List" <python-list at cwi.nl>
Sent: Friday, January 14, 2000 3:22 AM
Subject: Re: file width and height


> Darrell <news at dorb.com> wrote:
> > Sposhua wrote:
> > >Is there a module to find gif/jpeg width and height?
> > 
> > I hacked this out of PIL and it should work for gif and jpg.
> 
> """
> hacked for PIL by Darrell Gallion
> """
> 
> where "hacked" means "copied 150 lines of source
> code, and removed the copyright notice"...
> 
> </F>
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list




From bradmarshall at fastpointcom.com  Thu Jan 27 15:34:47 2000
From: bradmarshall at fastpointcom.com (brad marshall)
Date: Thu, 27 Jan 2000 12:34:47 -0800
Subject: Cookies
References: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com> <14461.65478.595789.251070@ghost.bbn.com>
Message-ID: <3890ABE7.40DF69AD@fastpointcom.com>

Tim,

I can't seem to find anywhere to download your Cookie module.  COuld you
help me out here?

Thanks
Brad Marshall




From emile at fenx.com  Wed Jan 19 11:47:14 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 19 Jan 2000 08:47:14 -0800
Subject: def args idiom and signatures
Message-ID: <02b001bf629c$d7ce3960$01ffffc0@worldnet.att.net>

In /F's Standard Python Library I spotted:

    def _d(y, m, d, days=(0, 31, 59, ..., 334, 365)):

where the balance of the function relies on days having this value.

Trying to understand why it may be an argument vs defined within
the function, I suspect that as the def is only processed to bytecode
once, whereas an in-line assignment would be re-processed with
each execution, that the decision is performance based.

Now, in light of the various discussions pertaining to function 
signatures, it seems that this idiom presents a particular problem.

Various munging of the name or introduction of a static type keyword
may help address this, but I'm curious if this common usage and
what people's thoughts are.

Or-even-if-my-understading-of-it-is-correct?-ly y'rs

Emile van Sebille
emile at fenx.com
-------------------







From gpatri at videotron.ca  Fri Jan 14 06:39:19 2000
From: gpatri at videotron.ca (Patrice Girard)
Date: Fri, 14 Jan 2000 06:39:19 -0500
Subject: Python Pre-processor (like php/asp)
References: <387D7349.DCF6567A@sprint.ca>
Message-ID: <rSif4.15$jw3.712@wagner.videotron.net>

Yes, ASP do the job.  Just to install Pythonwin and put SCRIPT=python inside
your .ASP and it's running!



Orlando Vazquez <ovazquez at sprint.ca> wrote in message
news:387D7349.DCF6567A at sprint.ca...
> Hi I'm looking for a Python equivalent of php or asp. I like their basic
> ideas but I dislike the languages. Is there any module or set of modules
> for this purpose? (I've seen one on the snippets site but it's kind of
> bare)
>
> return "Thanks" * 100
>
> --
> Orlando Vazquez




From billtut at microsoft.com  Thu Jan 13 14:39:58 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 13 Jan 2000 11:39:58 -0800
Subject: os.popen under Win32 Using Apache
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BCC01@RED-MSG-50>

http://www.python.org/doc/FAQ.html#7.13

Should help you out.

Bill

> -----Original Message-----
> From: Nicolas L?onard [mailto:leonar_n at epita.fr]
> Sent: Thursday, January 13, 2000 4:33 AM
> To: python-list at python.org
> Subject: os.popen under Win32 Using Apache
> 
> 
> Hit there
> 
> 
> I have a problem when I try to use os.popen() in a cgi with Apache,
> under Win32.
> 
> It works in command-line mode, but don't work through Apache.
> 
> Any idea ?
> 
> Thanks.
> 
> --
> Nicolas L?onard
> T?l : (33) 06-61-40-30-31
> E-Mail : leonar_n at epita.fr
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From hanche at math.ntnu.no  Sun Jan 23 13:42:25 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 23 Jan 2000 13:42:25 -0500
Subject: Tkinter, threads, mainloops, etc
Message-ID: <pcobt6c635a.fsf@math.ntnu.no>

I have a program that does some numerical computation on a time
dependent problem.  I'd like for the program to create some graphics,
updating it as it goes along, to visualise the results.  Up to now I
have used a really crude method, but one that works:  I open a pipe to
a ghostscript interpreter and send raw postcript commands down the
pipe.

I'd like to use Tkinter instead, in order to take advantage of some of
the higher order functionality available.  However, I still want to
have the python command prompt available, which I wouldn't if I called
the root window's mainloop().

I saw a message a while back, suggesting a separate thread to do event
handling, so I can keep the interactive interpreter in the main
thread.  Can someone provide some hints on how to do that in practice?
Maybe a pointer to some code that does it?

Actually, I have nearly come to believe that mainloops aren't
necessary at all.  Perhaps naively, I thought events would not be
handled at all until I run a mainloop, but the following experiment
shows that callbacks are indeed called even when no mainloop is
running:

Python 1.5.2 (#2, Nov 19 1999, 23:01:49)  [GCC 2.7.2.3] on freebsd3
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import Tkinter, whrandom
>>> def timbot(): print '<%.2f wink>' % whrandom.random()
.. 
>>> root = Tkinter.Tk()
>>> Tkinter.Button(root, text='Spam', command = timbot).pack()
>>> Tkinter.Button(root, text='Quit', command = root.destroy).pack()
>>> # Here I start hitting the Spam button
<0.95 wink>
<0.51 wink>
<0.88 wink>
<0.26 wink>

I can resize the window too, and the buttons get rearranged
accordingly.  So what is a mainloop good for, anyway?

(Also, would I get different advice if I were to run my program from
within IDLE?  That would certainly be no great hardship at all - I
just now started playing with it, and I have already more or less
abandoned py-shell in emacs.)

Confused-about-how-tkinter-really-hangs-together-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From pbible at home.com  Sat Jan 15 11:39:30 2000
From: pbible at home.com (Paul E. Bible)
Date: Sat, 15 Jan 2000 16:39:30 GMT
Subject: CGI, Apache, and mxODBC
Message-ID: <6p1g4.35$iL4.1041@news.rdc1.tx.home.com>

Hi all,

First of all, sorry for the length of this message...

Okay, I've been successful in getting Python CGI scripts to run under Apache
1.3.9 under Windows/NT.  Now it's time to add database functionality using
mxODBC to the script, but I'm encountering some problems.  I have provided
my script and the Apache environment debug listing below.

The script, as written, gives me a Internal Server Error.  If I comment out
the lines dealing with mxODBC (i.e., db = Odbc.Win..., etc.), the script
works fine; just doesn't do much.

Thanks in advance your help,
Paul

----------------------------------------------------------------------------
---
#!c:\progra~1\python\python.exe

import os, string, sys, cgi, re, Odbc

def Main():
     sql = 'Select * from cashacct'

     db = Odbc.Windows.Connect('TestDB','testid','testpw')

     # Add the columns
     cols = ()
     c = db.cursor()
     c.execute(sql)
     cols = c.description

     # Print the HTML header info
     print "Content-type: text/html"
     print
     print "<HTML>"
     print "<HEAD>"
     print "SQL Statement Test"
     print "</HEAD>"
     print "<BODY>"

     # Add the columns to the Access table
     print '<table border="0">'
     print "<tr>"
     for idx in range(len(cols)):
          print "<td>" + cols[idx][0] + "</td>"
     print "</tr>"
     print "</table>"

     print "</BODY>"
     print "</HTML>"
     sys.exit(0)

Main()

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

Shell Environment:
COMSPEC
D:\WINNT\system32\cmd.exe
DOCUMENT_ROOT
c:/apache/htdocs
GATEWAY_INTERFACE
CGI/1.1
HTTP_ACCEPT
image/gif, image/x-xbitmap, image/jpeg, image/pjpeg,
application/vnd.ms-powerpoint, application/vnd.ms-excel, application/msword,
application/x-comet, */*
HTTP_ACCEPT_ENCODING
gzip, deflate
HTTP_ACCEPT_LANGUAGE
en-us
HTTP_CONNECTION
Keep-Alive
HTTP_HOST
XX.X.XX.XXX
HTTP_REFERER
http://XX.X.XX.XXX/blank.htm
HTTP_USER_AGENT
Mozilla/4.0 (compatible; MSIE 5.01; Windows NT)
PATH
C:\orant\bin;D:\WINNT\system32;D:\WINNT;C:\RECYCLER\S-1-5-21-862944694-15761
02806-697575874-500\DC0;C:\RECYCLER\S-1-5-21-862944694-1576102806-697575874-
500\DC0\Perl;c:\Program Files\Symantec\pcAnywhere;c:\Program Files\python
PYTHONPATH
c:\Program Files\python;c:\Program Files\Lib;c:\Program
Files\python\lib\tkinter;c:\Program Files\python\win32com;c:\Program
Files\python\win32
PYTHONUNBUFFERED
1
QUERY_STRING
sql=select+*+from+cashacct&submit-button=Submit
REMOTE_ADDR
XX.X.XX.XXX
REMOTE_PORT
1052
REQUEST_METHOD
GET
REQUEST_URI
/cgi-bin/sql.py?sql=select+*+from+cashacct&submit-button=Submit
SCRIPT_FILENAME
c:/apache/cgi-bin/sql.py
SCRIPT_NAME
/cgi-bin/sql.py
SERVER_ADDR
XX.X.XX.XXX
SERVER_ADMIN
webmaster at home.com
SERVER_NAME
XX.X.XX.XXX
SERVER_PORT
80
SERVER_PROTOCOL
HTTP/1.1
SERVER_SIGNATURE
<ADDRESS>Apache/1.3.9 Server at 24.6.21.112 Port 80</ADDRESS>
SERVER_SOFTWARE
Apache/1.3.9 (Win32)
SystemRoot
D:\WINNT
WINDIR
D:\WINNT
None
These environment variables could have been set:
AUTH_TYPE
CONTENT_LENGTH
CONTENT_TYPE
DATE_GMT
DATE_LOCAL
DOCUMENT_NAME
DOCUMENT_ROOT
DOCUMENT_URI
GATEWAY_INTERFACE
LAST_MODIFIED
PATH
PATH_INFO
PATH_TRANSLATED
QUERY_STRING
REMOTE_ADDR
REMOTE_HOST
REMOTE_IDENT
REMOTE_USER
REQUEST_METHOD
SCRIPT_NAME
SERVER_NAME
SERVER_PORT
SERVER_PROTOCOL
SERVER_ROOT
SERVER_SOFTWARE
In addition, HTTP headers sent by the server may be passed in the
environment as well. Here are some common variable names:
HTTP_ACCEPT
HTTP_CONNECTION
HTTP_HOST
HTTP_PRAGMA
HTTP_REFERER
HTTP_USER_AGENT
None





From glenebob at nwlink.com  Mon Jan 17 14:58:34 2000
From: glenebob at nwlink.com (Glen Parker)
Date: Mon, 17 Jan 2000 11:58:34 -0800
Subject: Call Python functions from C[++] as macros
Message-ID: <38837344@news.nwlink.com>

Hi all,

Just getting started with Python (just downloaded the Win32 version last
night :-).  I need to figure out how to get some reasonably advanced things
working.  I'm writing a large app in C++ (BCB4.0) and I want to use Python
as an event handler for executing business rules etc.  So far I've been able
to export a function from C++ that is called from a Python script, and been
able to execute the script from C++, based on the examples.  That's a
usefull start, but I'm not sure where to look now for the next step.  I hope
somebody can give me some good pointers.

I want to have one or a small number of Python files that implement
functions I can call from C++, but I don't believe I have any need to
actually run the file(s) as scripts.  So I need to be able to pre-compile or
pre-import or whatever the file, get function references hooked up, and then
call them from arbritary spots in my C++ code.  Something like this:

void init() {
  Module = ImportPythonFile("ButtonEvents.py");
  Funcs["ButtonPress1"] = GetPythonFunction(Module, "ButtonPress1");
  Funcs["ButtonPress2"] = GetPythonFunction(Module, "ButtonPress2");
}

void Button1PressEvent()
{
    CallPythonFunction(Funcs["ButtonPress1"]);
}

void Button2PressEvent()
{
    CallPythonFunction(Funcs["ButtonPress2"]);
}

I know there are other things to be done here too, like passing args and
parsing return values, but I think I should be OK when it comes time to get
into that phase.

Any small code frags or tips on which Python library functions I'm looking
for here would be greatly appreciated.

TIA
Glen Parker
glenebob at nwlink.com




From skip at mojam.com  Tue Jan  4 16:29:15 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 4 Jan 2000 15:29:15 -0600 (CST)
Subject: String Handling
In-Reply-To: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com>
References: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com>
Message-ID: <14450.26155.589779.576625@beluga.mojam.com>

    Simon> Can anyone give me the Python line/command to change all spaces
    Simon> in a string to +s?

Taking a slightly different tack than the other respondents, I'll presume
with no real evidence that you are quoting strings for use in URLs (that's
the only other place I have seen space-to-plus conversions).

*If* that's the case, you probably want to use urllib.quote_plus:

    >>> import urllib
    >>> urllib.quote_plus("Can't we *ever* agree on anything?")
    'Can%27t+we+%2aever%2a+agree+on+anything%3f'

If you can tolerate (or want) %20 instead of +, use urllib.quote:

    >>> urllib.quote("Can't we *ever* agree on anything?")
    'Can%27t%20we%20%2aever%2a%20agree%20on%20anything%3f'

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 hanche at math.ntnu.no  Wed Jan 26 00:57:16 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 26 Jan 2000 00:57:16 -0500
Subject: Wanted: A python icon for X11
Message-ID: <pcosnzl4bpf.fsf@math.ntnu.no>

Would anybody perhaps have some pixmap lying about that might be (or
become) a suitable icon for IDLE windows under X11?  A picture of a
python (the snake) perhaps, or of a can of spam, or something like
that?  Right now, I only have the generic X icon, and that is of
course intolerable.

a-portrait-of-John-Cleese-might-also-do-nicely-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From nolan at ethereal.dhis.org  Thu Jan 13 13:52:47 2000
From: nolan at ethereal.dhis.org (Nolan Darilek)
Date: 13 Jan 2000 12:52:47 -0600
Subject: os.system() and PHP?
Message-ID: <871z7lercw.fsf@ethereal.dhis.org>

This isn't entirely a Python problem; it may very well be PHP's
fault. :) I thought I'd ask here first though before searching out PHP
newsgroups/mailing lists, since the problem is originating in a Python
script.

I'm working on an MP3 cataloging/jukebox system written in Python and
PHP. The program is controlled by a master Python script and a set of
PHP scripts. I've decided that the Python scripts would handle all
filesystem access and most table updates, thus allowing the Python
script to be called from the command line and letting the user
organize the MP3 collection and database. In keeping with Unix's small
is better philosophy, the Python script shells out to lots of
suplimental programs using os.system(). Similarly, the PHP scripts
occasionally shell out to the Python scripts to perform track editing
functions.

Here is where my problem lies. I've written a quick track editing page
which seems to work. Its goal is to read title, album and artist
information from the user, and to update that information in the
database. It then calls the Python driver script using PHP's
passthru() (For debugging only).

The script is called, yet the os.system() in the Python script which
calls mp3info doesn't work. I've inserted debugging prints into the
script and have gotten output, and I printed the return code from
os.system(), which is 256. The script works nicely from the command
line; once I edit the track I can then call the script with the exact
same command line and see the results I want.

I'm thinking that perhaps PHP's passthru() prevents os.system() from
running. Is there some sort of limitation on how many subshells a
program can spawn? I apologize if I sound confused; I'm not sure
whether this problem is with PHP, Python, or standard policies about
how os.system()-esc calls should work.

Thanks.


From ajung at server.python.net  Thu Jan 13 14:47:12 2000
From: ajung at server.python.net (Andreas Jung)
Date: Thu, 13 Jan 2000 14:47:12 -0500
Subject: ANN - PyRecode
Message-ID: <20000113144712.A666@starship.python.net>

I am happy to announce the availability of PyRecode.

PyRecode is a wrapper around the functionality of
Recode. Recode is a tool (under GPL) for converting
text from one character set to another one. Recode
is very useful when you are working with texts
in different formats, languages and character sets.

The sources are available from 
      http://www.suxers.de/PyRecode.tgz

At the moment it has been only tested under Linux,
Python 1.5.2 and recode 3.5. More platforms will follow
soon.

Cheers,
Andreas Jung



From mgushee at havenrock.com  Fri Jan  7 08:04:05 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 07 Jan 2000 08:04:05 -0500
Subject: First try at Tkinter stuff
References: <3875819D.41ECB8BA@exceptionalminds.com>
Message-ID: <m37lhm3uey.fsf@havenrock.com>

Timothy Grant <tjg at avalongroup.net> writes:

> I've got a window that displays four frames. It does it quite nicely and
> they look very nice. However, when I add a label or a button, the frame
> to which I add the label or button shrinks to the size of the button.
> This is completely undesirable behaviour.

Actually, it's completely desirable in many cases ;-). The frame is  
supposed to be a generic container for whatever you want, and is often
used as the base class for compound or customized widget. So by
default it has as few of its own characteristics/behavior as
possible; e.g., it automatically scales to fit whatever's inside it.

But I can understand your frustration.

> I have tried using both .pack
> and .grid, but I guess I don't understand how they are supposed to
> work.

Or you're just missing a small piece of the puzzle.

>         SupplierFrame.grid(row=0, column=0, rowspan=2)

Try this instead:

	SupplierFrame.grid(row=0, column=0, rowspan=2, sticky=N+S+E+W)

... or in your case, 'sticky=N+S' might be enough.

If you were packing instead of gridding the SupplierFrame, you would
use the options 'fill' and possibly 'expand':

	SupplierFrame.pack(expand=YES, fill=BOTH)

... or maybe just 'fill=Y' or 'fill="y"'. Fill affects the size of the 
packed widget, while 'expand' affects the cavity it rests in -- and
that distinction probably isn't clear, but I don't know how to make it 
any clearer in words. Just try it with and without 'expand' and see
what works. BTW, you may know this, but don't try to mix pack() and
grid() within the same container widget.

You can also pass options to an already gridded/packed widget with
grid_configure() or pack_configure().

Hope this helps a bit.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From vkolosov at unitedmedia.com  Tue Jan 25 11:20:43 2000
From: vkolosov at unitedmedia.com (Kolosov, Victor)
Date: Tue, 25 Jan 2000 11:20:43 -0500
Subject: Python online docs are great!
Message-ID: <461182F0B9B7D111B4000008C7282ADC22B0C5@nynt04.umny.scripps.com>

How do we persuade him to do the same to Zope documentation which really
needs it.

> -----Original Message-----
> From:	aahz at netcom.com [SMTP:aahz at netcom.com]
> Sent:	January 22, 2000 9:41 PM
> To:	python-list at python.org
> Subject:	Re: Python online docs are great!
> 
> In article <slrn88kouc.pej.nobody at isis.visi.com>,
> Grant Edwards <nobody at nowhere.nohow> wrote:
> >
> >I just want to compliment GvR on the Python online documentation files.
> >They are amazingly well done. They are without a doubt the best online
> >documentation I've used for anything, anywhere, ever.  Often online
> >documentation has been chopped up into pages that are way too small,
> >and there's no useful index or navigation aids.  The result is you
> >spend all your time thrashing around because the needed information
> >is spread over a half-dozen different pages.  I've found the Python
> >documents very well organized, comprehensive, and easy to use.
> 
> Fred Drake deserves much of the credit here. I can say from personal
> experience that the docs have improved greatly over just the last year.
> --
>                       --- Aahz (@netcom.com)
> 
> Androgynous poly kinky vanilla queer het    <*>
> http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
> 
> Have a *HAPPY* day!!!!!!!!!!



From skip at mojam.com  Wed Jan 26 11:19:32 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 26 Jan 2000 10:19:32 -0600 (CST)
Subject: To find a point inside a polygon
In-Reply-To: <388F0E36.6CE37135@ust.hk>
References: <388F0E36.6CE37135@ust.hk>
Message-ID: <14479.7828.793389.76253@beluga.mojam.com>

    David> Does anyone know if any python implementation for finding whether
    David> a point is inside a polygon? E.g., by giving a point(x,y) and an
    David> array of polygon's points (x0,y0,x1,y1,x2,y2,...), it returns
    David> true or false to tell if the given (x,y) is inside the polygon.

I think this has been asked before.  If not, check the comp.graphics FAQ for
a description of the algorithm and code one up in Python.  It's not
difficult.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From m.faassen at vet.uu.nl  Wed Jan 19 19:16:24 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jan 2000 00:16:24 GMT
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu> <865ij9$ia4$1@newshost.accu.uu.nl> <38865072.DBD8B4E5@callware.com>
Message-ID: <865k4o$iaa$1@newshost.accu.uu.nl>

Ivan Van Laningham <ivanlan at callware.com> wrote:
[snip]
> Well, this year I am *not* attending.

Darn, I was looking forward to meeting you! Oh well.

> Therefore, I plan to redesign
> Python so that you will all have to read and write Mayan glyphs in order
> to write any programs.

Sounds great! Should it be part of Numpy or the core language?
This is a math related extension, right?

Though-dutch-glyphs-would've-been-better-ly yours,

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


From max at neomeo.com  Wed Jan 26 13:14:42 2000
From: max at neomeo.com (David Max)
Date: Wed, 26 Jan 2000 13:14:42 -0500
Subject: PythonWin 2.0?
Message-ID: <388f3957_1@spamkiller.newsfeeds.com>

Does anybody know where I can download PythonWin 2.0? The latest I seem to
be able to find is build 128, yet I saw version 2.0 demonstrated on Monday
at the Python conference.

Thanks,

- David




From gerrit.holl at pobox.com  Thu Jan 20 03:26:44 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 20 Jan 2000 09:26:44 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <Pine.SOL.3.91.1000119233305.25615A-100000@sunse>; from sei@sunse.jinr.ru on Wed, Jan 19, 2000 at 11:37:45PM +0300
References: <Pine.SOL.3.91.1000119233305.25615A-100000@sunse>
Message-ID: <20000120092644.A1087@stopcontact.palga.uucp>

Evguenii Smogailov wrote on 948321465:
> Hi all,
> 
> Maybe someone (like PSA member) would like to submit this (below) as an 
> entry to the To-Do list (in minor language design section)?
> --------------------------------
> 
> New operators:
> 1) 'eq' (like '==');
> 2) 'ne' (like '!=','<>');
> 3) 'last' (like 'continue');
> 4) '..' for generating sequences (like '1..20' or 'a..z'),
> like in Perl.
> 
> Generally for improving readability,

Do you really think "a eq b" is more readable than "a == b"?
And 'last'? Why 'last' if we have 'continue'?

And I'd HATE 1..20

> (and Perl compatibility).

EEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEKS

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From tim_one at email.msn.com  Wed Jan 26 02:54:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 26 Jan 2000 02:54:08 -0500
Subject: using threads with python and winnt
In-Reply-To: <86lbsh$21i$1@nntp9.atl.mindspring.net>
Message-ID: <000501bf67d2$86b0f3c0$592d153f@tim>

[Tim, on the now fixed but still possible in 1.5.2 thread death
    Fatal Python error: PyThreadState_Delete: invalid tstate
]

[Aahz Maruch]
> Thanks.  That's exactly the error we're seeing, shortly
> after we made a recent increase in the number of running threads.

You're welcome -- but still a bummer.  Switching to a thread pool should
make it as unlikely as a Minnesota winter without snow.

Shortly after this got fixed, there was talk of putting together a bugfix
patch for 1.5.2 (with this & a few others).  Alas, nobody volunteered the
work to actually do it.  This is the rarely acknowledged Dark Side of the
Open Source.

this-is-not-the-bug-you're-looking-for-move-along-
     move-along-ly y'rs  - tim





From milek at task.gda.pl  Fri Jan 21 09:59:29 2000
From: milek at task.gda.pl (Robert Milkowski)
Date: 21 Jan 2000 14:59:29 GMT
Subject: os._exit(), os.kill(os.getpid(), 9)
References: <20000121145429.A4320@stopcontact.palga.uucp> <slrn88gsj5.77g.hildeb@stahlw06.stahl.bau.tu-bs.de>
Message-ID: <dt3888743di4ba0n79%milek@korweta.task.gda.pl>

Ralf Hildebrandt <hildeb at www.stahl.bau.tu-bs.de> wrote:
> On Fri, 21 Jan 2000 14:54:29 +0100, Gerrit Holl <gerrit.holl at pobox.com> wrote:
> 
>>os._exit(127)
> 
> This exits with exit code 127 and doesn't clean up...

how to exit with clean up?

-- 
							Robert Milkowski
							milek at task.gda.pl



From phd at phd.russ.ru  Mon Jan 31 05:26:14 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 31 Jan 2000 10:26:14 +0000 (GMT)
Subject: Date and Time
In-Reply-To: <000201bf6bd2$eecba190$3acbd9c2@peridot.optichrome.com>
Message-ID: <Pine.LNX.4.21.0001311023070.22111-100000@fep132.fep.ru>

On Mon, 31 Jan 2000, Adrian Eyre wrote:
> > I also have an all-python module that does a lot of date and time
> > things. It might give you some ideas.
> > See:
> >     datetime.py
> > at:
> >     http://starship.python.net/crew/jjkunce/
> 
> I'm surprised that something like this doesn't exist already in the
> standard Python library. Date and time manipulations are amongst the
> most common things done in programs. How difficult would it be to get
> it to be added in the next release?

   http://sun.med.ru/~phd/Software/Python/misc.html, file m_lib.tgz, module
opdate.py. Very simple, but useful.

   Mirrors:
http://members.xoom.com/phd2/Software/Python/
http://www.fortunecity.com/skyscraper/unix/797/Software/Python/

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From michael.stroeder at inka.de  Tue Jan  4 18:35:59 2000
From: michael.stroeder at inka.de (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 05 Jan 2000 00:35:59 +0100
Subject: http POSTing
References: <8EB19A9E7newshoesmailcom@205.147.63.94>
Message-ID: <387283DF.168A3543@inka.de>

NewShoes wrote:
> 
> i'm trying to create an offline form i can send to a website
> through "POST". being new to python and minimal knowledge of
> the http protocol, this has stumped me rather quick.

Use httplib and search examples on DejaNews...

Ciao, Michael.


From tismer at tismer.com  Mon Jan 31 14:55:36 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 31 Jan 2000 20:55:36 +0100
Subject: Ann: Stackless Python 1.02
Message-ID: <3895E8B8.995CA560@tismer.com>

Stackless Python is a Python without C stack usage.
It allows for all kinds of non-local control flow.
For info see http://www.tismer.com/research/stackless/

Update for Stackless Python V. 1.02:

- passes all standard tests
- should work with Zope now (try...finally was incorrect)
- has a smart object destructor for really deeply nested stuff

The "5 percent speedup" is no longer there currently, since
the smart object destructor needs to be optimized into
the objects. It is not done in this test phase but will come.

Please visit the Stackless Python homepage at

http://www.tismer.com/research/stackless/

Fact sheet, links to documentation, source and
binaries can be found there.

cheers - chris

<P><A HREF="http://www.tismer.com/research/stackless/">Stackless
Python 1.02 + Continuations 0.6</A> - a version of Python 1.5.2 that
does not need space on the C stack, and first-class callable
continuation objects for Python.  (31-Jan-2000)

Christian Tismer
Mission Impossible 5oftware Team



From calishar at my-deja.com  Mon Jan 24 16:29:24 2000
From: calishar at my-deja.com (calishar at my-deja.com)
Date: Mon, 24 Jan 2000 21:29:24 GMT
Subject: Importing a dll with a different name
Message-ID: <86ig7d$2pf$1@nnrp1.deja.com>

Hi Folks (who are here),

  I am running into a slight problem with imports after using installer
that I can't quite manage to figure out.

  The application I am working on import win32com.client, which imports
pythoncom this works fine as long as I am on my system, but when I
switch to a different system (after building the installable app) the
import fails.

  I have managed to figure out that it is because pythoncom is defined
in the windows registry to point to a specific dll
(c:\windows\system\pythoncom15.dll). I am also having the same problem
with pywintypes (and pywintypes15.dll).

  What I am trying to workout is a way that I can check to see if
import pythoncom will work or not, if it doesn't I will need to import
pythoncom15 from the current path

  Does this make sense? and is there someone around who can help me
figure out what I need to put in the sanity check routine?

 Calishar


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


From evan at tokenexchange.com  Wed Jan  5 11:59:29 2000
From: evan at tokenexchange.com (Evan Simpson)
Date: Wed, 5 Jan 2000 10:59:29 -0600
Subject: control structures (was "Re: Sins")
References: <Pine.LNX.4.10.9912261534560.8355-100000@palin.python> <387358BB.E71EC4E7@bellatlantic.net>
Message-ID: <s76uc120hu870@corp.supernews.com>

Steve Holden <sholden at bellatlantic.net> wrote in message
news:387358BB.E71EC4E7 at bellatlantic.net...
> Phil Jensen wrote:
> >     loop until "element found" or "search failed":
> >         ...
> >         if blah1 == blah2:
> >             "element found"
> >         ...
> >     when "element found":
> >         ...
> >     when "search failed":
> >         ...

I proposed a spelling for this a while back, and have even made sporadic
forays into the Python source to see if I can implement it myself.  It
extends 'try', and would look like this:

try:
    ... nested loops & stuff ...
           break "element found", elem
    ... more stuff ...
        break "search failed"
continue "element found", e:
    print "Found", e
continue "search failed"
    print "Not Found"

> Typically one would use enumerations rather that strings to
> represent the loop states, to combine readability with efficiency.

In this syntax, the strings are treated as labels (despite being string
literals grammatically) and are purely static, for both efficiency and
error-checking.  That is, "break STRING, exprlist" is a compile-time
SyntaxError unless the string exactly matches the string in a "continue"
clause of an enclosing "try" and the expression list is
assignment-compatible with the argument list of that same clause.  Once this
matching is done, the string is discarded and the "break" converted into an
expression evaluation followed by a Jump.  "continue" clauses compile into a
Jump to the end of the "try" followed by an assignment LHS (the target of
the "break" Jump) and the suite body.

In many ways this is quite similar to raising an exception and catching it
in an "except" clause.  The major difference is that everything is totally
resolved at compile time, and is lexically fixed.

> > Whatever the loop is, its exit or exits have some "meaning"
> > in the program, and this control structure encourages an
> > element of Literacy by forcing them to be named.

Exactly.  With no penalty in speed or (compiled) code size, in this case.

Cheers,

Evan @ 4-am




From akhar at videotron.ca  Mon Jan  3 04:31:48 2000
From: akhar at videotron.ca (Akhar)
Date: Mon, 3 Jan 2000 01:31:48 -0800
Subject: How to handle and create cookie with python??
Message-ID: <4oXb4.1127$UO3.31620@wagner.videotron.net>

I would like to know how can I create cookies and use them for various cgi
with python? Will python be fast enough for  a busy web site? Will it scale
properly? And for the cookies are there any tutorials or FAQs?




From mikael at isy.liu.se  Mon Jan 24 07:51:39 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Mon, 24 Jan 2000 13:51:39 +0100 (MET)
Subject: global variable problem
In-Reply-To: <388c496a.9657176@122.0.0.250>
Message-ID: <XFMail.000124135139.mikael@isy.liu.se>

On 24-Jan-00 volumeregeling wrote:
 >  Debugging = 0
 >  def remote_debug():
          global Debugging   # Here it is
 >        Debugging = 1
 >  
 >  For some reasong the Debuggine in the function is not the same as the
 >  global one , can you tell me how i can update the global variable from
 >  within a function ????  

See above.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    24-Jan-00
Time:    13:49:22

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From spamfranke at bigfoot.de  Thu Jan 20 18:21:22 2000
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Thu, 20 Jan 2000 23:21:22 GMT
Subject: Socket module on Mac
Message-ID: <388797dc.2416809@news.btx.dtag.de>

Are the socket module - and based on that, asynccore and asyncchat - 
fully available by Python on the Mac platform?

Sorry, I don't have a box here to check it myself.

Thx, Stefan



From scherbi at bam.com  Tue Jan  4 10:16:45 2000
From: scherbi at bam.com (Bill Scherer)
Date: Tue, 04 Jan 2000 10:16:45 -0500
Subject: String Handling
References: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com> <84t1u5$7dv$1@nntp8.atl.mindspring.net>
Message-ID: <38720EDD.32AF7323@bam.com>

Aahz Maruch wrote:

> string.translate() is faster than string.replace()

Faster to execute, slower to comprehend. ;-)

Bill


-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000104/b34d84f1/attachment.vcf>

From fredrik at pythonware.com  Fri Jan 14 03:33:01 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 14 Jan 2000 09:33:01 +0100
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com> <002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com> <jtu2kh52y6.fsf@wazor.biostat.wisc.edu>
Message-ID: <009501bf5e69$f908e530$f29b12c2@secret.pythonware.com>

William Annis wrote:
> "Fredrik Lundh" <fredrik at pythonware.com> writes:
> 
> > # -*-python-*-
> > 
> > or
> > 
> > #  -*- Mode: Python; tab-width: 4 -*-
> 
>         Those all work, but as I recall those work only on the first
> line of the file.  A quick test seems to confirm this.

well, Daryl asked for something to put "at the
top of the file", so I didn't think I had to point
that out ;-)

</F>




From c.evans at clear.net.nz  Mon Jan 10 05:22:42 2000
From: c.evans at clear.net.nz (Carey Evans)
Date: 10 Jan 2000 23:22:42 +1300
Subject: Python in Corel Linux
References: <947294639.1002448618@news.islandnet.com>
Message-ID: <87ya9ymdjp.fsf@psyche.evansnet>

J Dukarm <jdukarm at hydracen.com> writes:

> Has anybody out there installed Python in Corel Linux, and if so are there any
> tricks to getting it set up and working?

Corel Linux is based on Debian GNU/Linux <URL:http://www.debian.org>,
so you should just be able to point apt at a Debian mirror and install
Gregor Hoffleit's excellent Debian packages.

There may be some catches to this; I haven't done it myself, so I
don't know what else you might need to install.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

CONFIG_IPL_RDR
  Select this option if you want to IPL the image from a real card reader.


From ivanlan at callware.com  Sat Jan  1 18:38:14 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Sat, 01 Jan 2000 16:38:14 -0700
Subject: Python Press/Musings
Message-ID: <386E8FE6.EF707CB2@callware.com>

Hi All--
I don't normally read Dr. Dobb's, but a friend who is aware that I'm
trying to smuggle Python into Callware notified me that the latest issue
(January 2000, Vol 25, No. 1, Number 308), a special about "Scripting
Languages," has an article by Kirby Angell, a contributor to _The Quick
Python Book_, entitled "Python Server Pages, Part I."

PSP, an ASP-like server-side scripting engine that runs with everything
*but* M$'s IIS, looks pretty slick.  It's implemented in JPython, which
of course gives you a *ton* of goodies in addition to Python's normal
Euclid-full.  I haven't used it, but may give it a try.  One reason that
PSP doesn't work with IIS is that Angell chose to use $[ and ]$ as
script markers, instead of the ASP <% and %>.  In the article, he
laments not going with the ASP markers, since "win32all plugs a Python
interpreter into ASP", and if he'd used <% and %>, his applications
would have dropped right into ASP.

Well, it seems to me that he could change this; to avoid breaking
existing scripts, just provide the choice of <% and %> or $[ and ]$.  Or
both.

Also, the latest Linux Journal has a short essay by Eric Raymond, our
most visible friend, which mentions Python, and the latest Linux
Magazine has an extended interview with ESR--which also mentions Python.

And, as I recall, the December 1999 issue of LJ featured yet another
interview with Guido.

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

On another front, I've just spent the last week working on boiling
5000-8000 lines of C down to, um, 800 lines of Python.  The original C
took me a couple of months, at least, to write, using the Athena widget
set on Unix.  And wasn't nearly as slick as the Python/Tkinter version.

Oh, my, I do love Python.  For example:  one of the things that really
sucks about C is building dispatch tables with lots of little functions
that do nothing but map one set of function arguments to another set for
another set of functions.  That is, I'd have a callback function, say,
foo(w,d,c) (widget,data,closure) which has to call Foo(dd) ("different
data obtained from somewhere else").  So I'd have to either build a
couple of dispatch tables or write a bunch of callbacks which call the
appropriate function (that is, bar(w,d,c) simply calls Bar(dd)).  Or
something ugly and complicated.

Well, after writing the ninth or tenth function, I said, "There's gotta
be a better way."  (TGBABW)  I wrote a simple class  that takes function
as its only parameter for __init__.  __init__ stashes the function in
self.  I implemented a __call__ method which not only calls the stashed
function with the proper arguments, but stuffs the whole Tk widget
constructed by the called function into a Toplevel widget, and waits for
the user to push the OK button which my class has thoughtfully provided
as boilerplate.  14 lines of code.

I populated a list with nameless instances of my class--
flist=[Cmd(LongCount),Cmd(CalendarRound),Cmd(Stela),...] --which get
called as command callbacks from pushbuttons.  I just cruise through the
list of buttons setting the "command" property with my nameless
instances, et voila.

In um, 20 lines of code?  OK, maybe 25 if you count three for the
command-setting loop. ...

I discovered years ago that the way most people seem to program,
including me, is to think about the problem in the concrete and then
toss everything *possible* into the first implementation.  The rest of
the iterations through the code involve throwing things away ruthlessly,
paring it down until you've got no more and no less than you need.

But, like everyone else, I look at my older C or C++ (maybe I should
call it C-(> ), and I don't see ruthlessly pared down code.  What I've
got is a ton of gunk that I don't remember the function of.  Leftover
variables that I'm afraid to get rid of, functions that may or may not
be called, and functions with different, inconsistent, names that do the
same thing.

Python makes it *easy* to throw away code.  You still have to take the
time to look for those unused variables, functions and methods, but it's
far less painfull because for one thing, there's a lot less crud to sift
through.  For another, it's a trivial step to comment something out and
see what that breaks.

Then, after you throw away the excess Python code, you have the luxury
of spending time *thinking about what you're doing.*  You get to eat the
strawberry instead of thinking about the utensils you plan to eat it
with;-)  How often do you have the luxury of revisiting C++ code without
having to fix a bug, and *think* about what the code is doing, *think*
about how it could be done better.  I dunno about you, but I *never* get
to do that with C++.  With Python, it's just part of the process. 
Python Tao.

There's *lots* more slick little things I was able to do with this
project.  Maybe things like my Cmd class are old hat to everyone else,
but my point is not that I came up with such a clever idea; it's that
Python makes it so easy to see where the possibilities and opportunities
are in your own code that the enhancements just kind of throw themselves
into your arms like little licky puppies who want to be your best
friends.

<the-python-that-can-be-spoken-is-not-the-true-python>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From Robin.Friedrich at pdq.net  Mon Jan 31 23:57:43 2000
From: Robin.Friedrich at pdq.net (Robin.Friedrich at pdq.net)
Date: Tue, 01 Feb 2000 04:57:43 GMT
Subject: SNOWSPAM2000
Message-ID: <875p47$p9r$1@nnrp1.deja.com>

http://freeweb.pdq.net/robinf1/spam8pix/
(FTP to the starship isn't working so I put them up here.)
These are some of the photos taken last week at SNOWSPAM2000
--
Robin K. Friedrich     Houston, Texas


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


From cheselka at tucana.noao.edu  Thu Jan 27 15:05:34 2000
From: cheselka at tucana.noao.edu (Matt Cheselka)
Date: 27 Jan 2000 20:05:34 GMT
Subject: memory
Message-ID: <86q8ee$2g7k$1@noao.edu>

Hi,

Are there any issues with freeing memory in python?  Say for example I do a

	d = array.array('i')
	for i in range (2000):
	    d.append(0)

Do I need to free 'd' before I quit the program?

Cheers,

Matt Cheselka


From grant at nowhere.  Thu Jan 20 11:58:19 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 20 Jan 2000 16:58:19 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <XFMail.000120170005.mikael@isy.liu.se>
Message-ID: <slrn88eflb.6kp.grant@grante.comtrol.com>

In article <XFMail.000120170005.mikael at isy.liu.se>, Mikael Olofsson wrote:
>
>On 20-Jan-00 Grant Edwards wrote:
>[A detailed description about a stubborn guy at work and some
> lame bosses]
>
>Sounds like you should do what the rats are known to do: Leave the ship 
>before it sinks. I mean, if this is how your company is run it will sink
>sooner or later. Sooner rather than later.

No matter where you go, there's always somebody who's a pain.

Other than BOFH and the e-mail thing, it's a decent place. The
projects I'm on are interesting and I'm getting to do some
other stuff that's pretty cool (We're developing embedded code
for one of our ARM-based products under Linux with an
open-source RTOS from Cygnus).

-- 
Grant Edwards                   grante             Yow!  With YOU, I can be
                                  at               MYSELF... We don't NEED
                               visi.com            Dan Rather...


From tuttled at my-deja.com  Fri Jan  7 16:42:04 2000
From: tuttled at my-deja.com (tuttled at my-deja.com)
Date: Fri, 07 Jan 2000 21:42:04 GMT
Subject: Newbie Win32 Questions
Message-ID: <855mj9$ckc$1@nnrp1.deja.com>


Help please. I've got a couple of questions I
haven't been able to find any answers to.

1) Can you make a stand alone app? Meaning is
there a way to compile and distribute one or more
files without having to install Python on every
machine you want to run the app on?

2) If not, how do protect source code from prying
eyes?

TIA,
Don


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


From bwarsaw at cnri.reston.va.us  Wed Jan 19 00:58:51 2000
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Wed, 19 Jan 2000 00:58:51 -0500 (EST)
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
	<948195584.987404@news.adtranzsig.de>
	<388466D4.CC797CFC@earth.ox.ac.uk>
	<l03102800b4aac96334f7@[193.78.237.138]>
Message-ID: <14469.21147.337716.522238@anthem.cnri.reston.va.us>

>>>>> "JvR" == Just van Rossum <just at letterror.com> writes:

    JvR> To *many* people 8 spaces indentation is just too much. Look
    JvR> at the Python library: 4 spaces all over! (But had they used
    JvR> tabs anyone would've been able to decide for themselves...)

As I'm sure Just knows, but others may not, many of the older Python
modules used to be 8-spaced.  I think a few still are (Tkinter.py
perhaps, hard for me to check right now).  But the current standard
for .py files is 4-space, no tabs.

Also, if you look at Python's C code you'll see remnants of Guido's
coding preferences; all the C code is indented 8-spaced.  Sometimes
you have to get creative to fit everything within an 80 column window!

-Barry



From eddyw at lsl.co.uk  Mon Jan 17 13:37:20 2000
From: eddyw at lsl.co.uk (Edward Welbourne)
Date: Mon, 17 Jan 2000 18:37:20 GMT
Subject: [PSA MEMBERS] Re: [Types-sig] Suggestions for python 2
In-Reply-To: <388328F8.85147F85@maxtal.com.au>
References: <m129zpL-00006JC@vortigen.demon.co.uk>
            <388328F8.85147F85@maxtal.com.au>
Message-ID: <200001171837.SAA20866@lsls4p.lsl>

>>   * certain tools the functional programmers crave, notably a truly
>>     faithful implementation of currie()
>
>	not required: python lambdas subsume what can be done with
>	currying.

Unfortunately, unless I'm missing something, this is only true when you
know the call-signature of the function you are packaging.  Without safe
tunnels, I can see no way of handling even the simple case of `provide
the first argument to a function given no knowledge of how many
arguments the caller is going to give it': the currie()d result has to
use *args in its argument-list, yet we need arguments which supply it
with the function to call and the given first argument; these have to
appear before the *args (unless something has changed and no-one has
told me), consequently they will be over-ridden by the first few
arguments passed in by the caller of the constructed function:

def currieone(func, one):
    def result(*args, f=func, o=one):
	return apply(f, (o,) + args)
    return result

as safe tunnels would put it; if f and o precede *args (which they
currently must, as I understand it), the result is doomed because

  lambda f=func, o=one, *args: apply(f, (o,) + args)

called with args (a, b, c) would bind a in as f (over-riding the default
supplied, func), b as o (ditto, one), leaving args = (c,), so it would
attempt to call a(b, c) where func(one, a, b, c) was our intent.

Can you show me a piece of valid python (using lambdas or otherwise)
which achieves this effect ?

Granted, it can all be done (one of the first toys I built in python) as
a class ... but I need stuff like it before I can implement class ...
and we get so much more out of liberalising the position of *args and
**kwds relative to the name=value parameters.  Like, for instance, the
ability to have a function take arbitrarily many positional arguments
along with some keyword-arguments without the former and the latter
getting tangled up with one another - the things I call generators
commonly use (*bases, dict=None, meth=None, **what) ... unless dict and
meth keywords are used when the function is invoked, the function will
see them as None, no matter how many arguments are passed.

And, just for clarity, I don't particularly want this for the functional
tools (cute though I find them, fun though I find it to build them, and
useful though they are as sample things to check a system is powerful
enough to support) so much as for the expressive clarity of, for
instance,

def join(*strings, glue=' '):
    return string.joinfields(strings, glue)

invokable as join('hello', 'there', 'all', 'you', 'pythoneers') or, if
different padding is wanted, join('in', "Tim's", 'style', glue='-'),
without having to put [] around the sequence of strings that aren't the
keyword-supplied glue.

	Eddy.



From bjorn at roguewave.com  Thu Jan 27 15:27:37 2000
From: bjorn at roguewave.com (bjorn)
Date: Thu, 27 Jan 2000 13:27:37 -0700
Subject: memory
References: <86q8ee$2g7k$1@noao.edu>
Message-ID: <3890AA38.E1C604E1@roguewave.com>

> Are there any issues with freeing memory in python?  Say for example I do a
>
>         d = array.array('i')
>         for i in range (2000):
>             d.append(0)
>
> Do I need to free 'd' before I quit the program?

Nope, happens automagically when you can't use it anymore (aka reference
counting).

-- bjorn




From skaller at maxtal.com.au  Sat Jan  8 10:01:15 2000
From: skaller at maxtal.com.au (skaller)
Date: Sun, 09 Jan 2000 02:01:15 +1100
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <011501bf58a4$7b0db540$0100a8c0@rochester.rr.com>
Message-ID: <3877513B.73B762F4@maxtal.com.au>

Darrell wrote:
> 
> Using an exception as a "switch" statement is a cool idea.
> A dictionary approach comes to mind, with so many states.

Yeah, this will be MUCH faster for large switches,
because hashing is typically very fast, and hashtable
lookup is more or less constant time.

What would be 'generally' useful is a genuine switch 
statement -- a sort of unrolled dictionary whose values
are pieces of code. Syntax?

	switch x:
	case 1: ..
	case 2: ...
	default: ...

like C, except that each handler has an implicit
break at the end of the handler suite (no fall through,
like in C).

Comments?

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From talon at coursen.ifu.net  Wed Jan  5 22:22:21 2000
From: talon at coursen.ifu.net (Stephen Coursen)
Date: Thu, 06 Jan 2000 03:22:21 GMT
Subject: How to with **kw
References: <850vre$v3h$1@nnrp1.deja.com>
Message-ID: <slrn8782k2.2kk.talon@coursen.ifu.net>

On Thu, 06 Jan 2000 02:49:22 GMT, wiktor1325 at my-deja.com
<wiktor1325 at my-deja.com> wrote:
>Is there a way to pass **kw args from
>one function to another,like:
>
>def func1(**kw):
>  for key in kw.keys():
>     print key
>
>def func2(**kw):
>  a_way_to_call_func1_with(kw)
>
>thanks in advance
>
Something like this will work:

>>> def func1( dict={}, **kws ):
...     for key in dict.keys( ):
...             print key
...     for key in kws.keys( ):
...             print key
...
>>> def func2( **kws ):
...     func1( kws, another_key='some_value' )
...
>>> func2( key_1='hello', key_2='world' )
key_2
key_1
another_key

HTH,
Steve

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


-- 
--
Stephen Coursen                      talon at coursen.ifu.net
Q:  How many supply-siders does it take to change a light bulb?
A:  None.  The darkness will cause the light bulb to change by itself.


From a.eyre at optichrome.com  Tue Jan 18 12:52:44 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 18 Jan 2000 17:52:44 -0000
Subject: emacs debug: module reload
In-Reply-To: <387BFF31.C79AF8EC@saiph.com>
Message-ID: <000b01bf61dc$d30012c0$3acbd9c2@peridot.optichrome.com>

> but a systematic reload wud be ok too. Any idea?

import sys
for mod in sys.modules.values():
	reload(mod)

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From chriss at dnastar.com  Tue Jan 18 13:09:21 2000
From: chriss at dnastar.com (chriss at dnastar.com)
Date: Tue, 18 Jan 2000 12:09:21 -0600
Subject: Win32 BSDDB
Message-ID: <20000118120921.020261@artemis.dnastar.com>

Perhaps i'm missing something, but where dose the source for Python-Win32's
BSD-DB come from? That is, the data base itsself, not python's adaptors.



Christopher Stern
DNAStar,1228 South Park
Madison,  WI 53715-2102, USA
ph: 608.258.7420 ext. 25





From malcolmt at smart.net.au  Wed Jan  5 01:27:21 2000
From: malcolmt at smart.net.au (Malcolm Tredinnick)
Date: Wed, 5 Jan 2000 17:27:21 +1100
Subject: Folding (was Re: Python not a Very High-Level Language?)
In-Reply-To: <slrn874arb.ns.warlock@raven.myth>; from Jim Richardson on Tue, Jan 04, 2000 at 09:18:35AM -0800
References: <84lvpj$sn5$1@news1.xs4all.nl> <84njut$t74$1@pollux.ip-plus.net> <slrn874arb.ns.warlock@raven.myth>
Message-ID: <20000105172721.A1624@Ridcully.home>

On Tue, Jan 04, 2000 at 09:18:35AM -0800, Jim Richardson wrote:
> On 2 Jan 2000 13:31:09 GMT, 
>  lin li, in the persona of <linli at computer.org>,
>  brought forth the following words...:
> 
> >Message by boud at rempt.xs4all.nl (Boudewijn Rempt) on: 1/1/00 10:40:48 PM
> >>I've tried a lot of folding editors over the years, and I've never
> >>found one that was as pleasant to use as, say, Word's outline editor.
> >>Perhaps that's just because I was expecting that, instead of, for
> >>instance, Origami. What I'd like would be an editor that didn't need
> >>to insert all kind of cruft into the source files, but could work
> >>on a set of preferences and rules to present the source in a nice way.
> >
> >Just an idea, how about folding on the indentation levels? We just have
> >to forget about all these other languages then.
> >
> >Lin
> Works for me:) anyone up to chewing on the vim sources for this :)

Before we (I am interested in this, too) go leaping off into modifying vim
sources, be aware that folding is on the agenda for Vim 6. So if you are
serious about modifying sources, etc, it would be worth getting in contact
with the developers to find out what they have in mind, so at least we can be
consistent. It would be a royal pain to have one sort of folding for Python
and another type of mindset for everything else that is foldable.

Cheers,
Malcolm

--
Despite the high cost of living, have you noticed how popular it remains?




From r.p at fhtw-berlin.de  Wed Jan  5 10:49:55 2000
From: r.p at fhtw-berlin.de (reneP)
Date: Wed, 5 Jan 2000 16:49:55 +0100
Subject: +Python +Extensions +Mac?
Message-ID: <84voud$14tkj$1@fu-berlin.de>

Hi,

hope everybody have time for a little problem
im new in Python on the Mac, i have Version 1.5.2 and
want to build Extensions. So if compile a SharedLib
al is ok. I also include the e.g initSample Function in the c - source code.
But if i want to import in Python a Error occured >import error:  initSample
the specified symbol was not found<
The Lib is in the right Path, is was found.
If i compile the original Sources e.g. _tkinter there is the same problem.
The Compiler is CW Pro 2. and i have a G3-300. The projekts where only for
PPC SharedLibs.

Hope you can help, i try since 2 days .
Thanks Ren?




From doughellmann at home.com  Tue Jan 11 12:24:03 2000
From: doughellmann at home.com (Doug Hellmann)
Date: Tue, 11 Jan 2000 12:24:03 -0500
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov> <20000111114923.B42270@vislab.epa.gov>
Message-ID: <387B6733.2C49FB8A@home.com>

How is this different from function overloading, as implemented in C++? 
I'm not a C++  user, but my impression is the effect would be the same. 
Is that the point?

Doug

Randall Hopper wrote:
> 
> Randall Hopper:
>  |I had the same question yesterday and did a little surfing.  Basically,
>  |instead of switching off just the type of 'self' to determine which method
>  |to call, you switch off the type of other arguments as well.
>  |
>  |If you think best in implementation as I do:
>  |
>  |   def f( a,b ):
>  |     if   type(a) == Type1 and type(b) == Type1: f_t1_t1(a,b)
>  |     elif type(a) == Type1 and type(b) == Type2: f_t1_t2(a,b)
>  |     elif type(a) == Type1 and type(b) == Type3: f_t1_t3(a,b)
> 
> To be closer to the truth, I really should have said:
> 
>        if   issubclass( a, Type1 ) and issubclass( b, Type1 ): f_t1_t1(a,b)
>        elif issubclass( a, Type1 ) and issubclass( b, Type2 ): f_t1_t2(a,b)
>        elif issubclass( a, Type1 ) and issubclass( b, Type3 ): f_t1_t3(a,b)
> 
> This alludes to one way multimethods can get into method resolution
> troubles.  If none of the multimethods' parameter signatures match the
> argument types exactly, but multiple multimethods accepting compatible base
> class objects as parameters exist, ...
> 
> --
> Randall Hopper
> aa8vb at yahoo.com



From parker at gol.com  Sun Jan 23 09:04:17 2000
From: parker at gol.com (Ian Parker)
Date: Sun, 23 Jan 2000 14:04:17 GMT
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu>
 <867qt1$a49@news.or.intel.com>
Message-ID: <DAJ5pFA0awi4EwIe@gol.com>

In article <867qt1$a49 at news.or.intel.com>, tye4 <tye4 at yahoo.com> writes
>
>I have Python installed on Linux Redhat 6.0. Does anyone know where the
>source code for the compiler is located (any .tar file?). I intend to modify
>the language's syntax to add
>'end' delimiters for the blocks.
>And since Python is copyrighted, I don't intend to flood out versions to
>anyone else... strictly for personal use.
>
>class TestNew
>
>    def foo(s):
>        for i in range(0, 10):
>            print i,
>        end loop
>
>        if i == 10:
>            print 'internal error'
>        end if
>
>        while i < 10:
>            print i
>        end loop
>    end def
>
>end class
>
>My end delimiters are inspired by Ada, which in my personal opinion has best
>overall syntax.
>
>Thx,
>
>-tye4
>
>

Well since you're going back to ADA, I've just recalled my favourite
method of termination, in Algol 68: if.. fi, do..od convention, eg.

   if i == 10:
       print 'internal error'
   fi

A68 used "for .. do .. od" or "while .. do .. od". For Python perhaps
"rof" would be appropriate

   for i in range(0, 10):
       print i,
   rof          

Although I suppose to be more inline with Algol 68, you'd have to a
reverse colon on the for statement, which might be unclear.

I got profoundly fed up with languages that required "end xxx" to
terminate.    However, I still prefer even that to simply using
indentation (too much mixing of the message with the medium).

-- 
Ian Parker


From pierre at saiph.com  Tue Jan 11 23:12:33 2000
From: pierre at saiph.com (Pierre Imbaud)
Date: Wed, 12 Jan 2000 05:12:33 +0100
Subject: emacs debug: module reload
Message-ID: <387BFF31.C79AF8EC@saiph.com>

debugging my python scripts on Solaris, undex Xemacs,
using the python mode and the py-shell, I have a problem that
maybe someone solved? dealing with imported modules, I sometime
change them, but I then have to reload them explicitly. When
dealing with few module its ok, but it soon becomes a pain with 
many modules. Id love to have an a la make behavior, and be able to
reload automatically changed modules since last import. but a systematic
reload wud be ok too. Any idea?
-- 
Pierre Imbaud <pierre at saiph.com>
12 Rue des Bosquets 91480 Quincy Sous S?nart France
Tel:  01 69 00 94 57 Fax 79 65


From grant at nowhere.  Tue Jan  4 11:06:05 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 04 Jan 2000 16:06:05 GMT
Subject: OT: OpenSource Python Books?
References: <20000104032206.36493.qmail@hotmail.com>
Message-ID: <slrn8746jd.kr6.grant@grante.comtrol.com>

In article <20000104032206.36493.qmail at hotmail.com>, Mike Apted wrote:
>That's funny... he has "nospam" in his e-mail address, but he seems to 
>forget he has already graced us with his post and obviously does not read 
>the list.
>
>I wonder what "portion" would go back to the Python project... 1%?

0% is "a portion".

-- 
Grant Edwards                   grante             Yow!  Yow!! "Janitor
                                  at               trapped in sewer uses ESP
                               visi.com            to find decayed burger"!!


From rkruggel at coordt-edv.de  Sat Jan 22 08:48:27 2000
From: rkruggel at coordt-edv.de (Roland Kruggel)
Date: Sat, 22 Jan 2000 14:48:27 +0100
Subject: text-screen
Message-ID: <86cckj$2bgf$1@black.news.nacamar.net>

hi,

i'm a new a new python user. i am develop my programs in informix and change
now to python. i thing python is bette. i write programms text-screen
oriented. wat for modules i need to programm? who can help?
many thanks
roland...




From Stefan_Radke at compuserve.com  Thu Jan 13 21:01:05 2000
From: Stefan_Radke at compuserve.com (Stefan Radke)
Date: Fri, 14 Jan 2000 03:01:05 +0100
Subject: Threads of ASP with Python script code
Message-ID: <387E8361.29684CF9@compuserve.com>

The following ASP gives me trouble:

[HTML]
[some initialisation script code]
[more HTML]
<%
import my_python_module
%>

[more HTML, mixed with script code, creating an instance of class in
my_python_module:]
this = my_python_module.myClass()

The my_python_module has a class defnition. From this class I create an
instance, an read some file names from a disk into a list, the only member
of this class. The list is use dto create an options-list on the page.

The problem now is, that whenever I do a reload of the ASP in IE, the list
gets longer, because the same files are searched in the class-initialisation
again, but the previous instance still exists.

This even does not change, if I add a "del this" at the end. On another
machine the same list can be viewed, although a new session has been
started. To me this means that both ASP sessions use the same memory space
within the python environment :-( ...

Does anyone know how to implement this correctly ?  It would mean to have
seperate memory in each session.

Thank you for any help, Stefan.


From billtut at microsoft.com  Tue Jan 25 03:16:00 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Tue, 25 Jan 2000 00:16:00 -0800
Subject: MAPI and NT Service
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BCC73@RED-MSG-50>

The issue is described in the following KB article:

http://support.microsoft.com/support/kb/articles/Q197/8/20.ASP


To handle the MAPIInitialize call as required you can do the following:
import mapi

mapi.MAPIInitialize((mapi.MAPI_INIT_VERSION, mapi.MAPI_NT_SERVICE))

To handle the MAPILogonEx requirement the following MIGHT work from within
the service:
# assumes import mapi.

rawSession = mapi.MAPILogonEx(0, <profile name>, <password>, 
  mapi.MAPI_EXPLICIT_PROFILE | mapi.MAPI_NEW_SESSION | mapi.MAPI_NO_MAIL | 
  mapi.MAPI_NT_SERVICE)

# assume session points at a CDO/Active Messageing session object.
session.Logon(profileName=<blah>, profilePassword=<blah>, showDialog=0, 
newSession = 0, NoMail=1)

This should hopefully will allow you to do what you want, assuming your
service already is authenticated as the user who can log into the passed in
MAPI profile. The default system account can't access MAPI Profile
information.

Hope this helps,
Bill







From Jyrki.Alakuijala at iki.fi  Thu Jan 27 03:08:21 2000
From: Jyrki.Alakuijala at iki.fi (Jyrki Alakuijala)
Date: Thu, 27 Jan 2000 11:08:21 +0300
Subject: Parameters for Zope External Method
Message-ID: <388FFBE3.53CBA294@iki.fi>

I have succeeded in creating a Zope external method which
causes a python statement to be evaluated for a web request.
However, I have failed to pass parameters (other than the
special case "self") to that external method.

So, where can I find instructions on how to call an external
method with parameters? I could not find that in the
manuals.

Should I use something else than an external method?

--Jyrki




From fredrik at pythonware.com  Mon Jan 10 05:24:27 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 10 Jan 2000 11:24:27 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <852jbk$9tt$1@nntp9.atl.mindspring.net> <3875099A.2CF9333C@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net> <3876B79F.30439283@mdlink.de>
Message-ID: <001701bf5b54$e0e110d0$f29b12c2@secret.pythonware.com>

Helge Hess <helge.hess at mdlink.de> wrote:
> Hm, I wonder why it is simpler to have two separate kinds of methods
> instead of a consistent behaviour for all methods ?

umm.  in my dictionary, "consistent" means
"free from variation or contradiction".

in the current design, if you call a class object,
you *always* get an instance of that class.  in the
current design, if you inherit from a class and call
the parent's init function, it *always* initializes the
current instance.  in the current design, multiple
inheritance is *always* supported.

in your design, you've changed "always" to "it
depends".  hardly something that would make
Python's behaviour more consistent.

</F>




From aahz at netcom.com  Tue Jan 18 11:10:08 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jan 2000 16:10:08 GMT
Subject: DCOracle Large Objects
References: <Pine.LNX.4.21.0001181545370.15178-100000@fep132.fep.ru>
Message-ID: <862390$nfv$1@nntp5.atl.mindspring.net>

In article <Pine.LNX.4.21.0001181545370.15178-100000 at fep132.fep.ru>,
Oleg Broytmann  <phd at phd.russ.ru> wrote:
>
>   Anyone can help with Oracle LOBs and stored procs?
>   I want to create a LOB in a database. DB programmer created a stored
>procedure for me, but when I call it, passing data as Python string:
>
>   cursor.execute("add_image", [1028, "This is the body of my Large Object"])
>
>I got the error "invalid SQL". How do I pass the body of LOB to stored proc?

I don't know Oracle SQL, but you usually need to use "EXEC add_image" to
call a stored proc.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From PClaerhout at CREO.BE  Thu Jan 27 07:54:32 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 27 Jan 2000 13:54:32 +0100
Subject: How to differentiate Win98 from NT ?
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA7097@MSGEURO1>

As I said before, use M. A. Lemburg's platform.py
module. You can find it here:

http://starship.python.net/crew/lemburg/

Kind regards,


Pieter

Pieter Claerhout - PClaerhout at creo.be
Response Center - Applications Support

Creo Europe - Excelsiorlaan 21 - 1930 Zaventem - Belgium
Tel: +32 (2) 711 14 00 - Fax: +32 (2) 720 96 71


-----Original Message-----
From: Svante Kleist [mailto:Svante.Kleist at edt.ericsson.se]
Sent: Thursday, January 27, 2000 1:50 PM
To: python-list at python.org
Subject: Re: How to differentiate Win98 from NT ?



Not very solid, I'd say. In my case, NT lives in D:\nt40

/ Svante

In article <86du41$lv0$1 at gxsn.com>, matt at resonation.co.uk says...
> try :
>     if os.environ['windir']=="C:\\winnt":
>             # do whatever
>     else:
>             ....
> 
> 
> 
-- 
http://www.python.org/mailman/listinfo/python-list



From c.kotsokalis at noc.ntua.gr  Fri Jan  7 08:40:48 2000
From: c.kotsokalis at noc.ntua.gr (Constantinos A. Kotsokalis)
Date: 7 Jan 2000 13:40:48 GMT
Subject: mod python?
References: <Pine.WNT.4.21.0001070938001.197-100000@wks_3.e-shopwork.com> <20000107124754.A19678@mathi.uni-heidelberg.de>
Message-ID: <slrn87brao.bro.c.kotsokalis@ajax.noc.ntua.gr>

On Fri, 7 Jan 2000 12:47:54 +0100, Gregor Hoffleit wrote:
>There are alternatives to PyApache like PCGI (Persistent CGI,
>http://starship.python.net/crew/jbauer/persistcgi/) and a corresponding
>mod_pcgi (http://www.zope.org/Members/phd/mod_pcgi2).

Of course, FastCGI too: http://www.fastcgi.com
(no it's not commercial, or so I remember)

-- 
Constantinos A. Kotsokalis || c.kotsokalis at noc.ntua.gr
National Technical University of Athens
Network Management Center
Tel. No: +30 1 7721861


From fredrik at pythonware.com  Thu Jan 13 11:34:51 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 13 Jan 2000 17:34:51 +0100
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com>
Message-ID: <002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com>

dharms98 at my-deja.com wrote:
> I know there is a way to force emacs to go into python mode even if the
> file doesn't end with .py by putting something at the top of the file.

try something like:

# -*-python-*-

or

#  -*- Mode: Python; tab-width: 4 -*-

etc.  see the emacs info manual for more on this
(emacs -> major modes -> choosing modes)

</F>




From LCortes at Flash.Net  Sat Jan 22 22:43:38 2000
From: LCortes at Flash.Net (Luis Cortes)
Date: Sun, 23 Jan 2000 03:43:38 GMT
Subject: Porn python programm. The ultimate action
References: <862gl1$m3t$1@nnrp1.deja.com>
Message-ID: <20000123.3404700@negrita.brunner.org>


Do not repay evil with evil.

Do not reduce python or anything else to the level of a prostitute.  
This type of bad publicity would ruin chances of using python in the 
schools as a language to teach students.  Plus a bad image is not 
easily shaken.  It is better to promote python through word of mouth 
and friendly service ( it worked for Microsoft [ at least in the 
beginning ] ).


regards,
Luis.


>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 1/18/00, 12:58:34 PM, Manuel Gutierrez Algaba <algaba at my-deja.com> 
wrote regarding Porn python programm. The ultimate action:


> Being a bit annoyed with the fact that "anything sucessful in
> Internet tends to suck", and considering python a superb (and a bit
> forgotten in the midst of thousand of languages) I've being meditating
> if a major action would be nice to be taken. Admitting the fact that
> Porn sites (sex) are the most succesful and willingly to retain a bit
> of dignity, I think it'd be nice to develop a python program that
> generates sex histories, for this kind of programm OO would be superb,
> and we easily would attract attention, as I think this is a bit
> extreme decision to be taken, I think it's fair to ask for your 
opinion:

> What do you think about a "PornStories.py" ?

> This is not a joke, It's a serious stuff. Bad words could be expressed
> not directly but with invented (obvious words ) or long phrases.

> I think it's possible to deal with it with enough dignity and
> mock sex sites, at the same time....

> -------------
> Manolo


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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000123/8c94cba0/attachment.html>

From dworkin at ccs.neu.edu  Sat Jan 29 18:19:10 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 29 Jan 2000 18:19:10 -0500
Subject: Creating a local variable with a parameterized name
In-Reply-To: Michael Ross's message of "Sat, 29 Jan 2000 17:57:57 -0500"
References: <38937075.18B75753@rochester.rr.com>
Message-ID: <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu>

Michael Ross <mross at rochester.rr.com> writes:

> I want to create a local variable with a passed in name.
> 
> I realize I can do the following:
> 
> >>> def f(varname):
>              cmd = '%s = 123' % varname
>              exec(cmd)
>              cmd = 'print %s' % varname
>              exec(cmd)
> 
> >>> f('mike')
> 123
> >>>
> 
> But exec needs to invoke the parser, which seems to be an extremely
> costly thing to do.

As you have discovered, there are ways to do that, but none of them
are pretty.  Are you sure that you really need the name to be a
variable on its own, and that a dictionary wouldn't do?

That is, instead of things like:

cmd = '%s = 123' % varname 
exec(cmd) 
cmd = 'print %s' % varname 
exec(cmd) 

You could probably do things like this:

my_directory = {}
my_directory[varname] = 123
print my_directory[varname]

Instead of creating a variable whose name is the value of 'varname',
it makes a directory with a key equal to the value of 'varname'.
There are very few situations of this type that I can think of where
this would not be sufficient.

Directories are handy, and particularly well suited to this sort of
thing.  Why be any messier than you need to be?

-Justin

 



From tseaver at aftershock.neosoft.com  Mon Jan  3 12:43:11 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 3 Jan 2000 11:43:11 -0600
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <386E7BD4.514764EA@ndh.net> <84nv0u$bu9$1@nntp9.atl.mindspring.net> <3870DB37.761034EA@roguewave.com>
Message-ID: <75B5A3A6EB0EDE04.FCE8CED8D9B414D5.95006D8FA0ECEB31@lp.airnews.net>

In article <3870DB37.761034EA at roguewave.com>,
bjorn  <bjorn at roguewave.com> wrote:
>Aahz Maruch wrote:
>
>> In article <386E7BD4.514764EA at ndh.net>,
>> Stefan Schwarzer  <s.schwarzer at ndh.net> wrote:
>> >
>> >The other topics which the author on the O'Reilly site, Greg Wilson,
>> >mentions might also be worth considering. For example, I never needed
>> >tuples (explicitly) yet, i. e. in a sense that lists would not work.
>> >(If tuples would be taken out of the language or their use deprecated)
>> >tuple un/packing would become list un/packing and so on.)
>>
>> Anyone who claims that tuples are unnecessary has never used a tuple as
>> a dict key.  That requirement necessitates differentiating between a
>> mutable and immutable list -- might as well stick with tuples.
>> --
>
>So what you're saying is that dicts are broken since they can't take
>lists as keys <wink>.

What reasonable use can you propose for an associative container which allows
mutable keys?  You might as well revert to a list of (key,value) pairs, since
you're going to do a linear search on each lookup anyway if the keys are
allowed to change.

>
>undefined-behavior-isn't-always-bad'ly y'rs

I-want-a-word-to-mean-exactly-what-I-meant-when-I-said-it'ly

(Hmmm, TweedleDum/Dee as Pythonistas?)
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From wolf at one.net  Wed Jan 26 18:17:31 2000
From: wolf at one.net (wolf at one.net)
Date: Wed, 26 Jan 2000 18:17:31 -0500
Subject: Whitespace delimiters suck - Seen Occam?
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <86cmom$5pp$1@news-int.gatech.edu> <86h7jm$o7i@newton.cc.rl.ac.uk> <388C760A.1E66C20@gtri.gatech.edu> <86k8km$11p6@newton.cc.rl.ac.uk> <388E2D9A.C9A62D60@gtri.gatech.edu>
Message-ID: <l10v8s0semljq7tolp7gitr86g5jcr9sb9@4ax.com>

Andrew Henshaw <andrew.henshaw at gtri.gatech.edu> wrote:

>I don't have much call for Occam anymore, but I still use Origami for
>just about every programming project.  It supports a lot of other
>languages by default, and it can be set up to support most others.  It
>only took a couple of minutes to set it up for Python (including 
>keyword highlighting).

If you like Elcom's Origami you should try Loginov's ScopeEdit, more
expensive but I think it's far better. They actually have an undo
feature (unlike the version of Origami I bought from Elcom).

Respectfully,

Wolf

"The world is my home, it's just that some rooms are draftier than
others". -- Wolf


From herzog at online.de  Sun Jan 16 08:48:21 2000
From: herzog at online.de (Bernhard Herzog)
Date: 16 Jan 2000 14:48:21 +0100
Subject: xpot version 0.3
References: <oq66wv5a9h.fsf@titan.progiciels-bpi.ca>
Message-ID: <m3ln5qp1p6.fsf@greebo.nodomain.de>

Fran?ois Pinard <pinard at iro.umontreal.ca> writes:

> Hi, people.  A few days ago, I told you about a program able to extract
> a POT file out of a series of Python sources.  I made it available as:
> 
>    http://www.iro.umontreal.ca/contrib/po/xpot-0.3.tar.gz

This should be

     http://www.iro.umontreal.ca/contrib/po/xpot/xpot-0.3.tar.gz

I think.



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


From gerrit.holl at pobox.com  Thu Jan 13 14:33:54 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 13 Jan 2000 20:33:54 +0100
Subject: file width and height
In-Reply-To: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com>; from sposhua@my.pc on Thu, Jan 13, 2000 at 05:46:09PM +0000
References: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com>
Message-ID: <20000113203354.A1732@stopcontact.palga.uucp>

Sposhua wrote on 947781969:
> Is there a module to find gif/jpeg width and height?

HTMLgen uses that. Search for HTMLgen on http://www.vex.net/~x/parnassus

groeten,
Gerrit.

-- 
Please correct any bad English you encounter in my mail!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From claird at starbase.neosoft.com  Mon Jan 24 22:29:54 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 24 Jan 2000 21:29:54 -0600
Subject: tabs do WHAT?
References: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net> <slrn895gp2.rar.kc5tja@garnet.armored.net>
Message-ID: <ECDDDF77D8F2099E.DE9BA3AD946C1B64.BB59383FFE5FAEF2@lp.airnews.net>

In article <slrn895gp2.rar.kc5tja at garnet.armored.net>,
Samuel A. Falvo II <kc5tja at garnet.armored.net> wrote:
			.
			.
			.
>Now try explaining brace-delimited constructs to a newbie:
>
>Everything you want done under the IF statement needs to go inside these
>things called Braces. You need these braces to group the things you want
>done to treat them as a unit.  My recommendation is to write your code like
>this, but you don't have to.  Just make sure, whichever style you choose to
>write your code in, make it consistent so you'll know what you wrote two
>months ago:
>
>	if( a == b )
>	{
>	   do this;
>	   then do this;
>	   then do that;
>	   and finally, do that.
>	}
>
>I can absolutely guarantee a new-comer to programming, computer science
>major or otherwise, will have their head spinning with the latter
>description.  The former is a piece of cake, because _everyone alive today_
>writes outlines in almost exactly the same way, and in every way, they are
>whitespace sensitive.
			.
			.
			.
You forgot about the part where, if any more {-s and }-s
show up in the "do this" part, they have to balance.

Unless they're in quotes.

Except for ones in comments.

Or if they're themselves quoted, in some circumstances.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From mal at lemburg.com  Sun Jan 23 09:02:47 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sun, 23 Jan 2000 15:02:47 +0100
Subject: ANN: platform.py - Version 0.7.0
References: <3889CFEE.E2C0EF1E@lemburg.com> <388A1A94.C4B4B462@rubic.com>
Message-ID: <388B0A07.EF14C416@lemburg.com>

Jeff Bauer wrote:
> 
> "M.-A. Lemburg" wrote:
> > Documentation is provided through doc-strings and many
> > comments in the source code, which can be found via my
> > Python Pages:
> >
> >         http://www.lemburg.com/~python/
> 
> Actually should be: http://www.lemburg.com/python/
> URL without the tilde (~) character.

Ouch. I should be doing this more often again... ;-)
FYI, I've posted a corrected announcement to both lists.

The correct file should now also be available from the
web site by now (yesterdays upload didn't make it for some reason).

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




From ivanlan at callware.com  Fri Jan 21 13:04:33 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 21 Jan 2000 11:04:33 -0700
Subject: ConfigParser and .ini files
References: <86a1s2$7ha$1@nnrp1.deja.com> <14472.39533.223900.313932@weyr.cnri.reston.va.us>
Message-ID: <38889FB1.F692A45D@callware.com>

Hi All--

"Fred L. Drake, Jr." wrote:
> 
> super_banane at yahoo.com writes:
>  > The ConfigParser class in the default Python libraries
>  > does not support comments given on the value lines, separated
>  > by a semicolon, such as:
>  >
>  > [Section]
>  > keyword = value ; put your comments here
> 
>   You're right, it doesn't.
>   Does anyone know of a *real* specification for the .ini format?  I
> didn't know this was allowed.
> 

The spec is how Windows treats .ini files. ...

[snip]

> These should really be separated, because what most people want is a
> conventional .ini file parser.

Yup.  That's what I'm looking for this morning!

>   Anyway, I've attached a patch that allows ConfigParser to strip out
> ';' comments from .ini-style entries (using '=' instead of ':' as the
> value separator); RFC822-style entries should *not* be affected.  I
> think this is closer to what you're looking for.
>   I can check this in if this is the right approach.  I'm still not
> happy with the overloaded syntax, but this might be closer to getting
> the syntaxes properly implemented.  People should tell me if this is
> really the right thing to do, and if it breaks anything.
> 

I applied the patch, hoping it would help with my problem, but it
didn't.  Basically, what I'm seeing is two things:

1)  Section headers of the form "[Button States]" come up as
ParsingError because of the whitespace.  I read the regex for the
headers as allowing whitespace in the header, so what's going on here? 
Here's the regex:

 __SECTCRE = re.compile(
        r'\['                                 # [
        r'(?P<header>[-\w]+)'                 # `-', `_' or any alphanum
        r'\]'                                 # ]
        )

Not being a regex guru (not even an acolyte), I'm probably unqualified
to evaluate it.  What should this be to allow whitespace in section
names?  [*lots* of .ini files have whitespace in the section names. ...]

2)  The ParsingError lists the lines it thinks are bad at the end of the
run, but they come up like this on the listing:

        [line  5]: '[Button States]\012'
        [line 23]: '[CallWare Archive]\012'

Now, is this because ConfigParser thinks it's on UNIX, because it treats
everything as a UNIX file, because it is not expecting the '\r\n'
combination, or because it's broken?

<patiently-awaiting-enlightenment>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From doug at doug.speedex.net  Tue Jan 11 17:55:27 2000
From: doug at doug.speedex.net (Douglas Shawhan)
Date: Tue, 11 Jan 2000 16:55:27 -0600
Subject: Imaging-1.0 Compilation & OpenBSD
Message-ID: <Pine.BSO.4.10.10001111652020.6301-100000@doug.speedex.net>

I am having a problem getting the Imaging libs to compile properly under
OpenBSD 2.6:

--------------------begin my pain-----------------------
doug:~/Imaging-1.0# make
gcc -fpic  -IlibImaging -I/usr/local/include -I/usr/local/include
-DWITH_TKINTER -I/usr/local/include -g -O2 -I/usr/local/include/python1.5
-I/usr/local/include/python1.5 -DHAVE_CONFIG_H -c ./_imaging.c
./_imaging.c:63: warning: `WITH_TKINTER' redefined
*Initialization*:1: warning: this is the location of the previous
definition
./_imaging.c:2303: tk.h: No such file or directory
*** Error code 1

--------------------end my pain--------------------------
tk.h does indeed exist:

/usr/local/include/tk8.0/tk.h

Any clues?

Thanks.

d



From wjk at wjk.mv.com  Fri Jan  7 18:44:16 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Fri, 07 Jan 2000 18:44:16 -0500
Subject: does a variable exist
References: <B49BBF60.1373%Dan@Grassi.com> <dQsd4.586$mj7.15972@typhoon.nyroc.rr.com>
Message-ID: <38767A50.CFDEC3C2@wjk.mv.com>

Could you do something like:

try:
    print x
except NameError:
   print "No global or local x"

Just wondering?

Darrell wrote:

> "Dan Grassi" <Dan at Grassi.com> wrote in message
> news:B49BBF60.1373%Dan at Grassi.com...
> > I have forgotten and can not find the function (incantation) that can will
> > determine if a variable currently exisis.
>
> globals() and locals() are your best bet.
>
> >>> import sys
> >>> myVar=123
> >>> sys.modules[__name__].__dict__.keys()
> ['sys', '__doc__', 'myVar', '__name__', '__builtins__']
> >>>
> >>> class A:
> ...     classVar=111
> ...     def __init__(self):
> ...             self._x=1
> ...             print self.__dict__.keys()
> ...
> >>> a=A()
> ['_x']
> >>> a.__class__.__dict__.keys()
> ['__init__', '__doc__', 'classVar', '__module__']
> >>>
> >>> def f1(aVar):
> ...     print locals().keys()
> ...
> >>> f1(123)
> ['aVar']
> >>>
> >>> print globals().keys()
> ['__doc__', '__name__', 'A', '__builtins__', 'a', 'myVar', 'f1', 'sys']
> >>>
>
> --
> --Darrell



From jayantha at ccs.neu.edu  Sat Jan 29 12:16:51 2000
From: jayantha at ccs.neu.edu (Joshua C. Marshall)
Date: Sat, 29 Jan 2000 12:16:51 -0500
Subject: thread killing
Message-ID: <Pine.GSO.4.21.0001291214100.16444-100000@sirius.ccs.neu.edu>

I know it's been asked.

In my application I run untrusted code in a separate thread.  I want to
kill this thread if its execution takes too long (I don't care about
cleanup).

Is there any way to do this?  Is it possible for a spawned thread to
handle signals?



From linli at computer.org  Sun Jan  2 08:31:09 2000
From: linli at computer.org (lin li)
Date: 2 Jan 2000 13:31:09 GMT
Subject: Python not a Very High-Level Language?
References: <84lvpj$sn5$1@news1.xs4all.nl>
Message-ID: <84njut$t74$1@pollux.ip-plus.net>

Message by boud at rempt.xs4all.nl (Boudewijn Rempt) on: 1/1/00 10:40:48 PM
>I've tried a lot of folding editors over the years, and I've never
>found one that was as pleasant to use as, say, Word's outline editor.
>Perhaps that's just because I was expecting that, instead of, for
>instance, Origami. What I'd like would be an editor that didn't need
>to insert all kind of cruft into the source files, but could work
>on a set of preferences and rules to present the source in a nice way.

Just an idea, how about folding on the indentation levels? We just have
to forget about all these other languages then.

Lin


From jstok at bluedog.apana.org.au  Sun Jan 23 23:37:03 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Mon, 24 Jan 2000 15:37:03 +1100
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com> <86fl0m$gmj$1@inputplus.demon.co.uk>
Message-ID: <VuQi4.9271$3b6.42990@ozemail.com.au>

Ralph Corderoy wrote in message <86fl0m$gmj$1 at inputplus.demon.co.uk>...

>OK, I learnt C from the excellent K&R (Old Testament), all the Unix
>little languages, and many more besides.  I learnt Perl 4 years ago
>purely from the man pages, and Perl 5 at the same time from
>eavesdropping on the perl5-porters mailing list.  I like well written,
>terse, books.  Given all of that it sounds to me as if I should remove
>_Programming Python_ from my shopping basket and add _Python Essential
>Reference_.  Your views are sought.


Learning Perl from the manpages sounds exceedingly godlike, so  _Python
Essential Reference_ should suit you fine.  I just picked it up.  For me,
Python's modules and multiple inheritance trigger repressed memories from
Perl, Python's functional aspects twinge memories of Haskell, and Python's
classes have an exceedingly smalltalkish feel to them.  So long as you've
already encountered what the reference is describing, the reference should
serve you fine.  Even the online docs, for that matter, but the reference is
nicer.




From hamish_lawson at yahoo.co.uk  Mon Jan 17 07:05:42 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Mon, 17 Jan 2000 12:05:42 GMT
Subject: Version history of python
Message-ID: <85v0ih$409$1@nnrp1.deja.com>

I'm looking for a version history of Python that gives details of when
particular features were introduced into the language and when
particular libraries were included in the standard library. I've tried
looking on the python.org website and searching the web, but I've only
found an essay by Guido tracing Python's influences, but not detailing
when these features made it into the language.

Hamish Lawson


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


From michael.stroeder at inka.de  Wed Jan 19 07:58:56 2000
From: michael.stroeder at inka.de (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Wed, 19 Jan 2000 13:58:56 +0100
Subject: Python Rocks! - get rid of colons
References: <860ftk$bas@news.or.intel.com> <8610or$nsu$1@news.udel.edu> <861jm5$nrj$1@mach.vub.ac.be> <Pine.WNT.4.21.0001191125160.243-100000@wks_3.e-shopwork.com>
Message-ID: <3885B510.4A43BEA7@inka.de>

Sposhua wrote:
> 
> (oh, and wouln't it make life easier to change == to = and let python 
> work out when it is assingment and when it is comparison?)

No!

Ciao, Michael.


From sanderson at ttm.com  Mon Jan  3 12:24:23 2000
From: sanderson at ttm.com (Scott Anderson)
Date: Mon, 03 Jan 2000 12:24:23 -0500
Subject: Doing things in one line?
References: <78D9D2737D484D028525685B005D29CA.005D2ACF8525685B@ttm.com>
Message-ID: <3870DB47.51B4C5AA@ttm.com>

Actually, just search the mailing list archives back just before the
holidays, because I asked that very question. There were about 30
responses IIRC so I'm not going to summarize them here. :-)

Here's a one-liner I did the other day for the heck of it:

reduce(lambda x, y: x + y, range(1, int(raw_input('Enter a number: ')) +
1))

It accepts a number from the user and adds 1 to it. :-)

Regards,

-scott

gerrit.holl at pobox.com wrote:
> 
> Hello,
> 
> I know there's a FAQ entry called "Can I write obfuscated one-liners
> in Python?", but for fun, I'm trying to do it now.
> 
> Can I do this in one line:
> if a: foo()
> elif b: bar()
> else: foobar()
> 
> Is it possible with an intelligent construct that looks like:
> a and foo() or a and bar() or foobar()



From neilh at hare.net.au  Sat Jan 15 22:46:35 2000
From: neilh at hare.net.au (Neil Hodgson)
Date: Sun, 16 Jan 2000 14:46:35 +1100
Subject: How to use Pythonwin's ability to access windows dll's???
References: <1264133143-1099206@hypernet.com>
Message-ID: <007701bf5fd4$4b1d4c60$01646464@computer>

> Seriously what is the purpose of the LoadLibrary function if not to
> access Dll's>

   One thing you can do is load DLLs that define window classes and then
create windows of those classes. Then you send messages to those windows.
This is how Scintilla is integrated into PythonWin.

   I'd like to see calldll in the Python distribution rather than in
PythonWin. Its only a very small wafer of code that provides access to all
OS functionality. Most of my chain saw scars heal eventually anyway.

   Neil





From artymiak at safenet.pl  Fri Jan  7 16:53:48 2000
From: artymiak at safenet.pl (Jacek Artymiak)
Date: Fri, 07 Jan 2000 21:53:48 GMT
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net> <3875e3f1.230595438@news.oh.verio.com>
Message-ID: <387666A5.A1A6D28F@safenet.pl>

try this address:

http://www.corrt.com/info/pyisp-list.html

Jacek Artymiak

-------------------------------------------------------------------
 See how many people have registered at the Python Users Counter!
http://www.wszechnica.safenet.pl/cgi-bin/checkpythonuserscounter.py
-------------------------------------------------------------------
    Autor/Dziennikarz/Konsultant - Author/Journalist/Consultant
       artymiak at safenet.pl, http://www.wszechnica.safenet.pl
co-author:  StarOffice for Linux Bible (IDG Books Worldwide, Inc.)
http://www.amazon.com/exec.obidos/ASIN/0764533630/polskawszechnica
-------------------------------------------------------------------


From juxjux at hotmail.com  Wed Jan 12 10:00:49 2000
From: juxjux at hotmail.com (jux)
Date: Wed, 12 Jan 2000 16:00:49 +0100
Subject: Sorry...
References: <85i46b$jrl$1@zingo.tninet.se>
Message-ID: <85i4ti$nmu$1@zingo.tninet.se>

I Really didn't mean to send the same message twice.
It's no spam, I promise. (I HATE spam and wouldn't ever do it)

I Had problems with my e-mail program, so I send the message again... Didn't
think it would succeed in the first try.

That's what happens! (for me sometimes)





From mal at lemburg.com  Sat Jan 22 13:03:24 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 22 Jan 2000 19:03:24 +0100
Subject: ANN: platform.py - Version 0.7.0
References: <3889CFEE.E2C0EF1E@lemburg.com>
Message-ID: <3889F0EC.6E8BDA4B@lemburg.com>

ANNOUNCING:

                  platform.py - Version 0.7.0

        A Python module providing platform information APIs.


WHAT IT IS:

platform.py was intially developed for the mxCGIPython project
to automate the creation of file names including platform
specific data for users to be able to recognize and download
the appropriate Python interpreter binary for their (or their
ISPs) setup.

Since then the module has been extended and redesigned to
provide a generic interface for querying platform information
both in human readable form and for other application purposes
through platform dependent special APIs.

platform.py supports (at least) these setups: most Windows
setups, MacOS, various Linux distributions, various Unices,
OpenVMS, Java VM and DOS.


WHAT'S NEW ?

Version 0.7.0 is the first new version after the intial release
and update cycles last year.


WHERE CAN I GET IT ?

Documentation is provided through doc-strings and many
comments in the source code, which can be found via my
Python Pages:

        http://www.lemburg.com/~python/


WHAT DOES IT COST ?

It comes with a Python-style license, but is otherwise free for
commercial and non-commercial use.


WHERE CAN I GET SUPPORT ?

Commercial support for all my packages, tools and other
Python software will soon be available through my newly
founded software company. Keep an eye open on this list
for an announcement.


REFERENCE:

<P><A HREF="http://www.lemburg.com/~python/">
platform.py 0.7.0</A> - Generic platform information APIs.
(22-Jan-2000)


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





From dsnider at mindspring.com  Fri Jan 28 14:41:55 2000
From: dsnider at mindspring.com (Dan Snider)
Date: Fri, 28 Jan 2000 14:41:55 -0500
Subject: ObjectDomain status?
References: <86pqmb$d22$1@nnrp1.deja.com>
Message-ID: <86sr5r$3sd$1@news.inc.net>

Hi Harri,

> I wonder if the
> company has gone belly-up?

Absolutely, 100% not!! We are doing quite well but extremely busy preparing
a public beta of a new release. This should be posted real soon... You know
the deal about dates :-)

Sorry for the delay in some of the pre-sale support emails. We are trying to
hire
more people to accommodate the demand but the new release has us a little
behind.

Regards,

Dan

Dan Snider
President, Object Domain Systems Inc.
http://www.objectdomain.com
dsnider at objectdomain.com


<hpasanen at my-deja.com> wrote in message news:86pqmb$d22$1 at nnrp1.deja.com...
> Hi,
>
> This brushes on Python, so I though someone here might know...
>
> I've recently evaluated ObjectDomain UML modeller for our internal use.
> (http://www.objectdomain.com/)   The unique feature of their product is
> that it contains an embedded JPython as a scripting language, while
> the rest of it resembles Rational Rose a whole lot from the UI
> perspective.  (perhaps too much?)
>
> All in all, ObjectDomain 2.5 seems to work reasonably well, but I can't
> get any response from ObjectDomain to my queries.  I wonder if the
> company has gone belly-up?
>
> Anyone in the know?  Would be a shame to see a prominent Python product
> going down the drain, but perhaps they'll Open-Source it? ;-)
>
> Regards,
>
> Harri
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From jdukarm at hydracen.com  Fri Jan  7 20:17:14 2000
From: jdukarm at hydracen.com (J Dukarm)
Date: Fri, 7 Jan 2000 17:17:14 -0800
Subject: Python in Corel Linux
Message-ID: <947294639.1002448618@news.islandnet.com>

I just installed Corel Linux ("standard" version on CD) on a PC where I
can also boot Win98, and as a Windows refugee I was very pleased to see how
easy it was to access my Windows file system, read floppies, etc.  Only problem
is -- it seems that Corel left Python out of the distribution!

Has anybody out there installed Python in Corel Linux, and if so are there any
tricks to getting it set up and working?

Thanks,

Jim Dukarm
DELTA-X RESEARCH



From phd at phd.russ.ru  Mon Jan 10 10:00:30 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 10 Jan 2000 15:00:30 +0000 (GMT)
Subject: class destructor
In-Reply-To: <3879DB51.B9928BCE@obop.com.pl>
Message-ID: <Pine.LNX.4.21.0001101459290.24821-100000@fep132.fep.ru>

On Mon, 10 Jan 2000, [iso-8859-2] Przemys?aw G. [iso-8859-2] Gawro?ski wrote:
> Is there 'automatic' destructor in python, like AClass::~AClass() in C++
> ?

   Method __del__(self) will be called when python is about to destroy the
object (ie, reference count dropped to zero).

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From grant at nowhere.  Thu Jan 20 10:44:46 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 20 Jan 2000 15:44:46 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com> <slrn886oku.2ac.grant@grante.comtrol.com> <38871EA1.8AF33147@sage.att.com>
Message-ID: <slrn88ebbe.67b.grant@grante.comtrol.com>

In article <38871EA1.8AF33147 at sage.att.com>, Garrett G. Hodgson wrote:

>> The admin shut off both POP and IMAP to try prevent people from
>> using mail clients other than Outlook.  He's convinced
>> management that people who don't use Outlook were increasing
>> his workload.
>> 
>> IMAP used to be enabled to allow people to check e-mail from
>> home using Outlook Express, but once he figured out how to
>> enable the web interface, he told people to use that from home
>> and then he shut off IMAP (I was using fetchmail and mutt
>> at the time).
>
>while i expect you'll learn a lot solving your technical
>problem,

My motto: "You never know what will be useful at the next place
           you work."

>you should also pay some attention to your political problem.
>an out of control sysadmin doesn't do anyone any good. you
>might try and engage your other colleagues, gather some data,
>and enlist some management support.

Well, it's a long story.  Feel free to hit [tab] at any time:

I've been here about six months.  I'm told there was a big
fight about this a few months before I started.  Apparently,
before BOFH started, we had a Linux machine that acted as the
mail server (POP3 and SMTP).  People used whatever mail client
they felt like (Eudora, Elm, Outlook, Outlook Express, etc.).

Everybody was happy.

Then BOFH showed up and trashed the Linux machine and put in
Windows NT.

Then he complained that it was too much work supporting
POP/IMAP, and he wanted everybody to use Outlook.  People
resisted (particularly the Unix users, most of whom left
shortly thereafter).  The fight went all the way to the
president of the company, and the decision was made that
everybody use Outlook.  (Why is management always so big on
monotony and so down on individuality?)

Since people wanted to be able to read their e-mail from home,
he was forced to leave IMAP/SMTP services enabled so that they
could use Outlook Express or Netscape. 

I showed up and started using mutt via IMAP/SMTP. Everthing was
fine as far as I was concerned [Except that our DNS server was
broken so that forward and reverse lookups didn't match for
DHCP clients.  I pointed this out to him the first week I was
here, but I don't think he understood what I was talking
about.]

He noticed I was using something other than Outlook (he's never
heard of mutt, sendmail, fetchmail, or anything else non-MS),
so he disabled plaintext logins in the IMAP server and also
required SMTP authentication.  He claimed it was for security
reasons, but it happened about two hours after he came and
asked me how I was getting my mail -- something he was unable
to figure out on his own.  Apparently the broken DNS server
resulted in the MS SMTP server not knowing where to send bounce
notifications when I sent e-mail to an invalid address.  Rather
than fix the problem with the DNS/SMTP, his solution is to tell
users not to do that.  [IOW: don't fix a server's
configuration, shut it down!]

So I set up an out-of-office rule to forward all of my mail to
somewere else (where I could use ssh/mutt). He noticed I was
doing that, so he shut off forwarding mail to outside
mailboxes.  (Also, we presume, justifiable on security
grounds.) BOFH may have problems getting a server to work
right, but he can always figure out how to shut off a server.

I added NTLM authentication code to mutt and fetchmail (and
appropriate authentication code to ssmtp). I then resumed using
mutt via IMAP/SMTP.  

He complained to my boss (who is also his boss -- which
complicates things politically).  Our boss more or less ignored
the complaint. He did ask me what I used for e-mail, and I told
him.  He sort of implied that BOFH had told him that people
using IMAP required too much support. I told my boss that I had
never asked BOFH for any support, and I asked why having the
IMAP server enabled was such a burden. He admitted that he
didn't understand why it should be.

BOFH then set up the Outlook Web thingy and shut off IMAP
completely.

As a temporary measure, I installed VNC, set a Windows machine
up over in the corner underneath the table, and now have a nice
Windows desktop that shows up in an X window on my _real_
computer.

Meanwhile, I've got Python code running on the Windows box that
can read my inbox.  I've also got prototype code to shove the
messages out to the SMTP server on my real computer.  After I
figure out a general way to transform Exchange format addresses
into SMTP style addresses, I'll bolt the whole thing together
and once again be using mutt for my MUA.

[I use an outside mail account for 90% of my mail traffice, so
this is all probably less important that I make it sound.]

>f'rinstance, why is *his* workload the dominant factor?  

I pointed out to my boss that BOFH could reduce his workload
even further by not supporting any network services at all.

>have him defend and document what actual, measured effects
>there would be to his workload if he enabled POP and IMAP.
>gather info from outlook admins to challenge his assumptions.

My boss (who is also BOFH's boss) thinks the guy does a good
job -- which he does, as long as everything is done his way and
you don't need anything done that he doesn't care about (like
having a DNS server that works right).

>or just get some big guys and go kick his ass.

I'm still building up "political capital" as it were.  I'm not
going to start a fight unless I know that if push comes to
shove, I'm not the one who gets the shove.

My colleagues all find the whole thing amusing in a Quixotic,
Children's Crusade sort of way.

-- 
Grant Edwards                   grante             Yow!  Where's th' DAFFY
                                  at               DUCK EXHIBIT??
                               visi.com            


From bernhard at alpha1.csd.uwm.edu  Sun Jan  2 15:47:23 2000
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 2 Jan 2000 20:47:23 GMT
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com>
Message-ID: <slrn86vear.6ed.bernhard@alpha1.csd.uwm.edu>

On Sat, 01 Jan 2000 21:40:47 GMT, randy_shaffer at my-deja.com <randy_shaffer at my-deja.com> wrote:

>I have experimented with tkinter and the PMW library.  Things seem to
>operate a bit slow.  The other options/considerations I?ve pondered are:
>
>Move the gui to the Win platform ? I?m not very proficient in MFC-land
>and I?m worried.  Is there Python portability to Windows
>problems/issues as well?
>
>Can Viper help out here?  I get the feeling that it is not quite ready
>for prime time.  Am I wrong?
>
>What about JPython and the fabulous Java Swing library?  Am I asking
>for trouble here?
>
>What are your thoughts, please?

My thoughts are: 
	- viper will not help.

	- I also don't believe in Java. This language is not stable
	enough right now to be used for anything serious

	- You could use the win32 extentions and operate MFC from there,
	but you would need to learn that nasty and proprietory GUI library.

	- I would go with a faster crossplatform GUI library.
	Only WxPython and the python bindings for FOX and FLTK come
	to mind. 

	- I only tested wxPython briefly. It definatly is still 
	evolving a bit, but IMO it is most promissing as there are
	a bit of real wxWindows applications popping up. And it works
	pretty stable for me under Linux and Windows.

Bernhard
-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.  (www.uwm.edu/~bernhard)
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From drew46 at netscape.net  Fri Jan 14 02:14:50 2000
From: drew46 at netscape.net (Andrew McCumiskey)
Date: Thu, 13 Jan 2000 23:14:50 -0800
Subject: CallDLL and Long Far Pointer Return
Message-ID: <mvDTB9lX$GA.167@hawkeye.i3s.net>

I've been trying to get calldll to return a long far pointer back from a dll
function call.  I've tried every PyArg_ParseTuple type there is and still
crash upon calling the function.  All of my other functions work flawlessly
in this program and I've narrowed it down to the return type.

I was able to compile a module with C that returns the proper type but want
to avoid having to do that for the other 20 or so dll functions that return
the same type.

The function prototype claims it is returning a LONG FAR PASCAL.

Has anyone seen this and/or been able to get around it?

I'm running Python 1.5.2 with Win32 extensions build 127.

Thanks ahead of time for any help.

Drew





From darrell at dorb.com  Thu Jan  6 12:06:22 2000
From: darrell at dorb.com (Darrell)
Date: Thu, 6 Jan 2000 12:06:22 -0500
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com>
Message-ID: <005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>

From: "Skip Montanaro"
>     Emile> Doesn't this do that now? or is there some objection to this
usage?
>
>     [ ... snip try/except used to implement state machine ... ]
>
> I find nothing wrong with it.  Some people object to using try/except for
> anything but handling errors however.

Don't know if this if fair. But exceptions are slow when raised.
import time

def exception(cnt):
    for x in range(cnt):
        try:
            raise 0
        except:
            pass

def loop(cnt):
    for x in range(cnt):
        if x > cnt/2:
            pass
        else:
            pass

size=100000
t1=time.time()
exception(size)
print 'Time:',time.time()-t1

t1=time.time()
loop(size)
print 'Time:',time.time()-t1

######## Output
# Time: 3.59399998188
# Time: 0.25






From hartmut at oberon.noris.de  Sun Jan  2 08:58:30 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Sun, 02 Jan 2000 14:58:30 +0100
Subject: Pseudo-SQL: searching for Parser/Rewriter
Message-ID: <16600378@oberon.noris.de>

Hello everyone,

I'm looking for a general purpose tokenizer/parser/rewriter for Python. If a
parser is available, it best should return a simple parse tree.

I need to rewrite Pseudo-SQL statements for hiding complex table layouts.
This sould convert eg.

  FromName = 'hugo' or FromName = 'Hans' and ToName = 'me'

into

  select ID from messages where
    sender in (select ID from adresses where name='hugo' or name='Hans')
  AND
    ID in (select msgID from receivers where
              type = 'to'
           and
              addrID in ( select ID from addresses where name='me' )
          )

Huh, now I see how complicated this will become, I consider dropping this
feature ;-)


+++hartmut


From tim_one at email.msn.com  Sat Jan 29 16:26:52 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 29 Jan 2000 16:26:52 -0500
Subject: Case Sensitivity and Learnability
In-Reply-To: <86uoo7$1a4$1@vvs.superst.iae.nl>
Message-ID: <000501bf6a9f$8f501e00$78a0143f@tim>

[possibly Will Rose]
> ... If unskilled programmers find case too difficult or distracting,
> the cure is to force a single case, either upper or lower.  This could
> be done by an (optional?) compiler switch.

[Carel Fellinger]
> this would make it impossible to import library modules that do depend
> on case distinction. Or would it be a per module switch, settable in
> the source?  Now that will improve things:)

Relying on case-sensitive file names is already broken in *portable* Python,
as some filesystems don't support the notion (e.g., while NT's NTFS
*preserves* filename case, system functions ignore it).  And Python had to
grow the PYTHONCASEOK envar switch for filesystems that don't even preserve
case.

> I, for one, vehemently dislike case insensitive languages, as
> they force me to use ugly naming conventions like:
>
>   class tv_class: pass
>   tv = tv_class()
>
> yack, instead of the much clearer:
>
>   class TV: pass
>   tv = TV()
>
> Admitted you have to know this convention of using case to
> distinct classes from variables:), but then in natural language
> we use this same trick to distinguish persons from objects like
> in Bill and bill.

As in Bill of Materials, or

> groetjes, carel

<wink>?  The problem seems more that in programming we use a great many
compound "made up" words, and it can be a real bitch to remember e.g.
whether some function is named moveUp or MoveUp or moveup.  Some moron API
designer may even supply all three, but doing slightly different things.
You can say a consistent set of rules "should be" followed, but that's a lot
like saying newbies "shouldn't get" confused to begin with <wink>.

> So how hard can it be for a newbie:) OTOH, I'm easily convinced
> by proper usability studies.

The great thing about newbies is that they pay no attention whatsoever to
arguments about how they should behave <0.9 wink>.  I recently posted a link
to Matt Conway's doctoral dissertation (see DejaNews), where his project's
difficulties in using Python in a newbie-oriented system are explained.  The
Alice system ships with a modified interpreter that ignores case (and also
treats int/int as a floating-point operation).

ever-use-a-case-sensitive-web-search?-ly y'rs  - tim





From robin at jessikat.demon.co.uk  Sat Jan  8 21:06:45 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 9 Jan 2000 02:06:45 +0000
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
References: <20000105082703.A9441@vislab.epa.gov>
 <1265040530-3892638@hypernet.com> <ONAe+FAsO4c4Ew5h@jessikat.demon.co.uk>
 <pqKd4.2780$Ef6.704898@news.shore.net>
 <IN2z0BAYY5d4EwgO@jessikat.demon.co.uk> <87embs8eqf.fsf@ifm.uni-kiel.de>
Message-ID: <0bnwLEA10+d4EwWI@jessikat.demon.co.uk>

In article <87embs8eqf.fsf at ifm.uni-kiel.de>, Janko Hauser
<jhauser at ifm.uni-kiel.de> writes
>Sorry to jump in so latly. I haven't used it myself, but I understand
>that the ``fpectlmodule.c'' from standard distribution has some of the
>needed infrastructure to catch exceptions from the C-code and handle
>them.
>
>But the code talks about hardware specific things. Have you looked at
>this module already or am I offtopic?
>
>HTH
>__Janko
>
No I guess I will have to look there as well.
-- 
Robin Becker


From niels at endea.demon.nl  Tue Jan 11 15:25:25 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Tue, 11 Jan 2000 21:25:25 +0100
Subject: ioctl
References: <85ft10$1kdg$1@noao.edu>
Message-ID: <387B91B5.CB7F4517@endea.demon.nl>


Matt Cheselka schreef:
>         >>> xxx = 0
>         >>> val = fcntl.ioctl (fd.fileno(), RNDGETENTCNT, xxx)
>         Traceback (innermost last):
>           File "<stdin>", line 1, in ?
>         IOError: (14, 'Bad address')
> 
> So it appears as though I'm not passing the correct value of 'xxx' since it
> looks like fcntl.ioctl is interpreting 'xxx=0' to be the address of where to
> place the result, which is in the ROM and not legal.
> 
> So I've got two questions:
> 
>         1. is this essientially the correct way to do ioctl stuff or is there
>            a better way?

It would be a little cleaner to put things like this into a C module, so
you could do something like
 val = rnddev.getentcnt()
but for occasional use this is probably acceptable.

>         2. how do I pass this method the correct address?

First, you need to set up a mutable buffer to accept the returned value.
An array object with one integer element seems most appropriate:
  val = array.array('i', [0])
Then you can do
  fcntl.ioctl (fd.fileno(), RNDGETENTCNT, val.buffer_info()[0])
and the entropy count will be in val[0].

-- 
Niels Diepeveen
Endea automatisering



From python at scoobysnacks.com  Sat Jan  1 13:06:06 2000
From: python at scoobysnacks.com (Marshall)
Date: Sat, 01 Jan 2000 18:06:06 GMT
Subject: pickle problem:cant reload object saved with binary option
References: <386e1de1.998928692@news-server> <YHpb4.914$Sr1.104202@typhoon.nyroc.rr.com>
Message-ID: <386e31c2.1004019140@news-server>

Thanks, now I have a new problem.  It still works without the binary
option.  I think I have the organization of my data screwed up
somehow, but I don't understand why it works without the binary, and
breaks if I turn it on.  It doesn't matter if I use pickle or cPickle,
I still get errors.  The object I'm pickling is self.info=crccvars()
Thanks in advance for any insight.

	def save(self,file='crcc.sav'):
		blah=open(file,'w')
		cPickle.dump(self.info,blah,1)
		blah.close()
	def load(self,file='crcc.sav'):
		print 'print self     :',self
		print 'print __name__ :',__name__
		print 'print self.info:',self.info
		print ''
		blah=open(file,'r+b')
		self.info=cPickle.load(blah)
		blah.close()
		self.buildlist()

each time I open the app, save, then load and here's what I get ...

print self     : <__main__.crcc instance at 7b3830>
print __name__ : __main__
print self.info: <__main__.crccvars instance at 7b6df0>

Exception in Tkinter callback
Traceback (innermost last):
  File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 764, in
__call__
    return apply(self.func, args)
  File "new.py", line 250, in load
    self.info=cPickle.load(blah)
ImportError: No module named __main__



On Sat, 1 Jan 2000 11:15:15 -0500, "Darrell" <news at dorb.com> wrote:

>Try this:
>        fp=open(fn,'r+b')
>
>
>--
>--Darrell
>"Marshall" <python at scoobysnacks.com> wrote in message
>news:386e1de1.998928692 at news-server...
>> I created a class instance (self.info) to hold all my variables and
>> I'm trying to save it for use in later sessions.  The binary file is
>> much smaller but if I set the binary flag while saving, it blows an
>> error when I try to load it.  I'm trying this on Win98 and I'm calling
>> it from a Menu widget using command=self.save and I change the default
>> bin=0 to bin=1 and it won't load anymore.  The load error is at the
>> bottom.  Any direction is greatly appreciated.  Thanks.
>>
>
>
>



From kern at caltech.edu  Sat Jan  8 17:12:33 2000
From: kern at caltech.edu (Robert Kern)
Date: 8 Jan 2000 22:12:33 GMT
Subject: is variable a list?
References: <38729120.7616054D@simplypets.com>
    <86hfgp6o7m.fsf@ruby.somewhere.washington.us>
    <857jnf$ne4$1@nntp5.atl.mindspring.net>
Message-ID: <858coh$lv1@gap.cco.caltech.edu>

In article <857jnf$ne4$1 at nntp5.atl.mindspring.net>,
	aahz at netcom.com (Aahz Maruch) writes:
> In article <86hfgp6o7m.fsf at ruby.somewhere.washington.us>,
> Evgeny Roubinchtein  <eroubinc at u.washington.edu> wrote:
>>
>>Hmm, I guess TIMTOWDY _sometimes_ applies to Python :-).
> 
> Well, sure, that's going to be true of pretty much any powerful
> programming language, but TMTOWTDI (note spelling)

Personally, I think TIMTOWDY is far more appropriate for Python.

TIM-Tells-Of-Ways-Damnably-Yucky-ly yours,

Robert Kern
kern at caltech.edu


From orlov at diasoft.ru  Wed Jan 12 04:05:54 2000
From: orlov at diasoft.ru (Oleg Orlov)
Date: Wed, 12 Jan 2000 12:05:54 +0300
Subject: Python BUG?
Message-ID: <85hg1q$2nli$1@gavrilo.mtu.ru>

This script:

d = [1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L,
    10L, 9L, 8L, 7L, 6L, 5L, 4L, 3L, 2L, 1L]
print reduce(lambda x, y: x*x + y*y, d)

hangs up my python interpreter (on NT4).

Is it a bug or my mistake?
How about UNIX?

--
====================
Oleg Orlov
orlov at diasoft.ru




From tim.hochberg at ieee.org  Tue Jan  4 20:01:39 2000
From: tim.hochberg at ieee.org (Tim Hochberg)
Date: Wed, 05 Jan 2000 01:01:39 GMT
Subject: Simple cartesian product
References: <iacpuvhqwdp.fsf@vier.idi.ntnu.no> <iacn1qlqvw7.fsf@vier.idi.ntnu.no> <iack8lpqvoi.fsf@vier.idi.ntnu.no> <iachfgtqvj9.fsf@vier.idi.ntnu.no>
Message-ID: <TJwc4.948$R51.11406@news1.rdc1.az.home.com>

"Magnus L. Hetland" mailto:mlh at vier.idi.ntnu.no wants a Cartesian product:

If you can use Numeric and you're limiting yourself to numbers, there are
several ways to do it, one of which is shown below. It's not as short as
yours, but I almost guarentee its faster. Figuring out why this code is the
way it is is left as an excercise for the reader.

def product(a, b):
    a = asarray(a, 'O')
    b = asarray(b, 'O')
    result = zeros((len(a), len(b), 2),  'O')
    result[:,:,0] = result[:,:,0]  + a[:,NewAxis]
    result[:,:,1] =  result[:,:,1]  + b[NewAxis,:]
    result.shape = len(a) * len(b), 2
    return result.tolist()

If you need to use objects other than numbers, I believe the following
should work. (It may get confused if you feed the elements of the original
lists are sequences). This code is weird because of the rules as to when
Numeric makes copies of arrays. Figuring it out is another excercise
<0.5wink>.

def product2(a, b):
    result = zeros((len(a), len(b), 2),  'O')
    reshape(result , (-1,2))[:,0] = a*len(b)
    result = array(transpose(result, (1,0,2)))
    reshape(result , (-1, 2))[:,1] =  b*len(a)
    result = reshape(transpose(result, (1,0,2)), (-1, 2))
    return result.tolist()


-tim




From scarblac-spamtrap at pino.selwerd.cx  Tue Jan 18 08:06:53 2000
From: scarblac-spamtrap at pino.selwerd.cx (Remco Gerlich)
Date: 18 Jan 2000 13:06:53 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
Message-ID: <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl>

tye4 wrote in comp.lang.python:
> I'm a python newbie.. but I like Python better than Perl (no ugly @$#@.. in
> syntax).
> 
> Python does have one quirk: no end if or end loops
> 
> if x == 10:
>     print 'x is 10'
> print 'Hello world'
> 
> # easy to get confused whether this line is in 'if' block or not

If you have problems with that, try indenting it more, and/or use
blank lines. The following looks rather clear to me:

if x == 10:
      print 'x is 10'
	 
print 'Hello world'

And of course, get used to it. It's pretty easy to do.

if-whitespace-isn't-enough-use-more-whitespace-ly-yrs, etc
-- 
Remco Gerlich,  scarblac at pino.selwerd.cx
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From gmcm at hypernet.com  Thu Jan 20 17:12:04 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 20 Jan 2000 17:12:04 -0500
Subject: How to store a function pointer in class?
In-Reply-To: <867ssq$a3c$1@mordred.cc.jyu.fi>
Message-ID: <1263716168-11209979@hypernet.com>

Sami Hangaslammi writes:

> Justin Sheehy <dworkin at ccs.neu.edu> wrote in message
> news:vnd66wo1nz2.fsf at betelgeuse.ccs.neu.edu...
> > You could store the function inside a mutable class data member.
> 
> Yeah, I figured this out myself, but it still seems like an
> unneccessary and ugly hack. 

Perhaps if you understood the mechanics, it would not appear 
that way.

 obj.method(args)

becomes (in psuedo code):
  apply(getattr(obj, method), args)

For instance objects, getattr first checks in the instance, then 
in instance.__class__, then a walk of the bases. It finds a 
function object, (unless it's actually on the instance, but that's 
rare). This is, on the fly, transformed into an unbound method, 
which (when it trickles back to the instance) becomes a 
bound method. This can be called (and the "self" parameter is 
safely squirreled away in the binding).

If you want things on the class that _stay_ function objects, 
you'd need some way to distinguish those which _should_ be 
transformed from those that _shouldn't_. No such flag exists, 
and no such hook exists.

People ask for this regularly. But there are those (like 
yourself) who want to be able to get a plain old function, and 
there are those who want "class methods", ie, methods bound 
to the class object, not the instance object.

- Gordon



From erw at dde.dk  Tue Jan 25 07:34:10 2000
From: erw at dde.dk (Erwin S. Andreasen)
Date: 25 Jan 2000 12:34:10 GMT
Subject: Embedding -- getting error/traceback information not to stderr?
References: <000101bf6695$977d6140$3acbd9c2@peridot.optichrome.com>
Message-ID: <slrn88r54f.ebi.erw@erw-pc.dde.dk>

On Mon, 24 Jan 2000 18:05:26 -0000, Adrian Eyre <a.eyre at optichrome.com> wrote:

>> Did I miss some other similar function that will return the nicely
>> formatted exception information in a variable?
>>  1) copy PyErr_Print(), modify it to append to a file (yuck!)
>>  2) Modify sys.stderr to append to an internal buffer instead of a file

>3) Set stderr to be stdout

Well, I wanted the information to be in a variable in memory, which I can
then send to the user (which isn't looking at stdout :) ). I suppose I could
set sys.stderr to something like a file which I then open and read after
each use of PyErr_Print -- or a pipe the other end of which I watch -- but
that doesn't seem very... elegant.

-- 
==============================================================================
Erwin Andreasen   Herlev, Denmark <erw at dde.dk>          UNIX System Programmer
<URL:http://www.andreasen.org>              <*>              Goodbye, Richard!
==============================================================================


From mikael at isy.liu.se  Fri Jan 28 13:03:01 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 28 Jan 2000 19:03:01 +0100 (MET)
Subject: tabs do WHAT?
In-Reply-To: <949076889.194238@news.adtranzsig.de>
Message-ID: <XFMail.000128190301.mikael@isy.liu.se>

On 28-Jan-00 Dirk-Ulrich Heise wrote:
 >  What bothers me even more, they don't mind writing functions without the
 >  function comment headers that shall be used in the project.
 >  You end up with functions containing not a single char of comment.

Guilty as charged.

 >  It would be another improvement to Python if it would force you to
 >  write a doc string for each function of method.

I would take that punishment. I would need that. No, wait! I think I 
should rewrite python to log me out if I forget a docstring.

But-then-again-maybe-not'ly y'rs

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    28-Jan-00
Time:    18:58:26

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From aa8vb at yahoo.com  Fri Jan  7 09:28:32 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 7 Jan 2000 09:28:32 -0500
Subject: XML to dictionary?
In-Reply-To: <m3puvf5p4h.fsf@havenrock.com>
References: <8518kq$59n$1@nnrp1.deja.com> <m3puvf5p4h.fsf@havenrock.com>
Message-ID: <20000107092832.A25967@vislab.epa.gov>

Matt Gushee:
 |from pyxie import File2PYX    # PYX is the notation pyxie uses for
 |import string		      # parsed XML
 |
 |f = File2PYX('hithere.xml')
 |data = f.read()
 |f.close()
 |data = string.split(data)
 |
 |dict = {}
 |for line in data:
 |    if line[0] = '(':         # '(' at the beginning of the line in 
 |			      # PYX indicates an element

Pyxie is a good tool.  But just to pick nits, using string.split isn't safe
with Pyxie output.  :-)

Consider:

    <?xml version = "1.0"?>
    <Greeting>
    Wishing You A Merry Christmas!
    From all of us (here at the White House)
    Author: --Unknown
    </Greeting>

string.split gives you a new start tag, attribute, and text element you
didn't know you had ;-)

-- 
Randall Hopper
aa8vb at yahoo.com



From gerrit.holl at pobox.com  Mon Jan 17 10:50:29 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 17 Jan 2000 16:50:29 +0100
Subject: Version history of python
In-Reply-To: <85v0ih$409$1@nnrp1.deja.com>; from hamish_lawson@yahoo.co.uk on Mon, Jan 17, 2000 at 12:05:42PM +0000
References: <85v0ih$409$1@nnrp1.deja.com>
Message-ID: <20000117165029.A3761@stopcontact.palga.uucp>

Hamish Lawson wrote on 948107142:
> I'm looking for a version history of Python that gives details of when
> particular features were introduced into the language and when
> particular libraries were included in the standard library. I've tried
> looking on the python.org website and searching the web, but I've only
> found an essay by Guido tracing Python's influences, but not detailing
> when these features made it into the language.

In the source distribution:
Misc/HISTORY

File attached.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From gerrit.holl at pobox.com  Thu Jan 27 09:10:22 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 27 Jan 2000 15:10:22 +0100
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
In-Reply-To: <m12DjXD-000CxiC@artcom0.artcom-gmbh.de>; from pf@artcom-gmbh.de on Thu, Jan 27, 2000 at 08:40:07AM +0100
References: <m12DjXD-000CxiC@artcom0.artcom-gmbh.de>
Message-ID: <20000127151022.A1920@stopcontact.palga.uucp>

Peter Funk wrote on 948958807:
> Look at the following coding alternatives often seen in generic python code:
> 
> class A:
>     def load(self): 
>         # ... do something e.g. load a file
>         try:
>             self.update()
>         except AttributeError:
>             pass
> 
> class B:
>     def load(self):
>         # ... do something e.g. load a file
>         if hasattr(self, 'update'):
>             self.update()
> 
> Which coding idiom is better?  The second form is one line shorter than
> the first, but it has the disadvantage, that the method (or attribute) must
> be named twice.  Both idioms are used in the standard libary.

The second one: is self.update() contains an error raising an
AttributeError, you don't want to catch it.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From mjackson at wc.eso.mc.xerox.com  Thu Jan 20 16:01:12 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 20 Jan 2000 21:01:12 GMT
Subject: Python Rocks! - get rid of colons
References: <867qt1$a49@news.or.intel.com>
Message-ID: <867t2o$q2j$2@news.wrc.xerox.com>

"tye4" <tye4 at yahoo.com> writes:
> 
> I have Python installed on Linux Redhat 6.0. Does anyone know where the
> source code for the compiler is located (any .tar file?). I intend to modify
> the language's syntax to add
> 'end' delimiters for the blocks.
> And since Python is copyrighted, I don't intend to flood out versions to
> anyone else... strictly for personal use.
> 
> class TestNew
> 
>     def foo(s):
>         for i in range(0, 10):
>             print i,
>         end loop
> 
>         if i == 10:
>             print 'internal error'
>         end if
> 
>         while i < 10:
>             print i
>         end loop
>     end def
> 
> end class

If you use a slightly variant spelling you'll find that the stock
Python compiler will recompile itself on the fly:

[don't forget the colon]
             |
             V
class TestNew:

    def foo(s):
        for i in range(0, 10):
            print i,
        #nd loop

        if i == 10:
            print 'internal error'
        #nd if

        while i < 10:
            print i
        #nd loop
    #nd def

#nd class

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
    Faced with the choice between changing one's mind and proving
    there is no need to do so, almost everybody gets busy on the proof.
				- John Kenneth Galbraith




From fredrik at pythonware.com  Fri Jan 21 07:21:33 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 13:21:33 +0100
Subject: regex question
References: <roy-835F11.21172620012000@netnews.nyu.edu> <01eb01bf63ec$532be300$f29b12c2@secret.pythonware.com> <20000121214110.A1880@Ridcully.home>
Message-ID: <004401bf640a$11561d10$f29b12c2@secret.pythonware.com>

Malcolm Tredinnick <malcolmt at smart.net.au> wrote:
> umm .. speaking as an "official person who has never seen the 1.6 regex
> implementation", could you explain what you are talking about?

http://www.deja.com/getdoc.xp?AN=571652264

> It sounds interesting, but I cannot find anything on the Python website
> about what's on the hotlist for 1.6 (am I going blind?).

the original plan contains lots of stuff, but it may boil
down to unicode, distutils, and assorted bug fixes in 1.6,
and all the rest in 1.7.  or whatever.

I suppose we'll know more after the developer's day
on next week's conference...

</F>




From didier.rano at wanadoo.fr  Thu Jan 20 15:28:26 2000
From: didier.rano at wanadoo.fr (didier.rano)
Date: Thu, 20 Jan 2000 21:28:26 +0100
Subject: using PMW's scrolledframes
Message-ID: <867r7f$kk$1@wanadoo.fr>

     Hello,

     I've used your PMW to insert two scrolledFrames (sf1 and sf2) within a
     main one (sf).

     This works fine but what I would like to do now is to make the sf
     horizontal scrollbar move sf2 while sf1 remains visible.
     I've tryed to modify _xview which I renamed in xview2 but this doesn't
     work.
     When I invoke the following command

        self.sf._horizScrollbar.configure( command=self.sf2.xview2)

     I get an AttributeError on xview2.



     I would be very greatful if you could help me on that one.
     Thanks in advance,


     Didier Rano




From News at Titanic.co.uk  Fri Jan 14 14:35:30 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Fri, 14 Jan 2000 19:35:30 GMT
Subject: Database to Web
References: <jmsu7sk17q2hg0dbrf3qi8lobednf72m4a@4ax.com>
Message-ID: <ejuu7s4r94q6a3scu6qm52grn5tq8h8k73@4ax.com>

What I needed was:

Name = tuple[record]
 for Field in Name
	print "<td>" + Field + "</td>"


Why is Python so easy?

Simon
Simon Faulkner


From anders.eriksson at morateknikutveckling.se  Sun Jan 16 10:07:54 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Sun, 16 Jan 2000 16:07:54 +0100
Subject: CGI, Apache, and mxODBC
References: <6p1g4.35$iL4.1041@news.rdc1.tx.home.com> <gX1g4.48$iL4.1177@news.rdc1.tx.home.com>
Message-ID: <b9qBODYB7K8gObMCEbChFPLIGi9N@4ax.com>

On Sat, 15 Jan 2000 17:15:56 GMT, "Paul E. Bible" <pbible at home.com>
wrote:

>File "c:/apache/cgi-bin/sql.py", line 13, in Main
>db = Odbc.Windows.Connect('Allegro','helistest','oilwell')
>mxODBC.OperationalError: ('IM002', 0, '[Microsoft][ODBC Driver Manager] Data
>source name not found and no default driver specified', 4265)

Not being a guru I would say that the error is in the ODBC setup. You
don't have any ODBC DSN with the name of 'Allegro'.

Check the spelling and the upper/lower case...

// Anders

---
Enlish isn't my first or second language, 
so anything that you find offending is not there on 
purpose it's just due to the translation ;-)


From esj at harvee.billerica.ma.us  Sat Jan  1 22:37:41 2000
From: esj at harvee.billerica.ma.us (Eric S. Johansson)
Date: 2 Jan 2000 03:37:41 GMT
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg>
Message-ID: <84mh65$2uv$1@harvee.billerica.ma.us>

Ajith Prasad <aprasad at magix.com.sg> wrote:
> http://www.oreilly.com/news/vhll_1299.html is an article by Greg Wilson
> casting doubts on the effectiveness/value of Python and other very high
> level scripting languages. Wilson comments that:
> "Over the past few years, I have done several medium-sized projects using
> both Perl and Python. At first, I was very excited by what these Very-High
> Level Languages (VHLLs) let me do, and how quickly. The more I played with
> them, however, the less satisfied I was. In particular, I no longer believe
> that they deserve the 'V' in their name. This article explores why, and
> suggests some ways in which they could evolve." Worth responding to as it
> includes detailed criticisms of Python in particular.

interesting article and a good set of commentary from other folks in
this thread.

for me, most of the suggestions were of the "yup, good ideas that will
be ignored" type.  He left out one thing that I think is very
important from Eiffel which is the concept of the "obsolete" keyword.
It's a very useful device which allows you to tag what public names
are going away so that users of your toolkit can get reasonable
notification before the names actually vanish.  There is a school of
thought that says that no functions or public names should ever go
away but I believe that's a reaction to the lack of a mechanism for
notification.

Of all the things he suggested, I personally want to see programming
by contract support.  I've done it in C/C++ using assert.h and it's
wonderful for picking out bugs in your thinking as well as your
implementation.

The XML based code suggestion was also fairly interesting.  The
reminded me of my experiences with literate programming, another very
powerful but underrated tool.  Using literate programming techniques,
I was able to generate code some 50 percent faster with higher
reliability because the technique mapped to how I think of software
blocks.

A life changing event has significantly altered my relationship to
software development.  About five years ago I was injured from too
many hours on the keyboard and I am now dependent on speech
recognition for most computer use.  As a result, I've developed a
strong interest in speech driven programming techniques.  Shifting
from the keyboard focus to a vocal focus highlights some of what Greg
Wilson was talking about in terms of high level vs. very high-level
languages.  It has also altered my thinking to the point that people
suspect me of consuming way too many mushrooms when I talk about
programming by voice.  ;-)

If you can, try the experiment of stepping back from using your
keyboard when writing code.  Think about and vocalize what you would
tell someone to write your code for you.  Another time, think about
how you would write code differently if you had an intelligent aid
looking over your shoulder and filling in detail for your shorthand.

Some of things the come out of this is the necessity for an editing
environment to be aware of context, being able to separate the
difference between "type" and "purpose", smart abbreviations
(proforms) based on your context and name scope rules, smart help for
function definitions, schemas which let you alter one data element
within the schema and then reflect the change throughout the entire
schema.

The world gets very strange when you move away from a keyboard focus.
You find yourself wanting to operate at a very high level with big
chunks and let the context fill in the details.

--- eric

-- 
Eric S. Johansson    esj at inguide.com	esj at harvee.billerica.ma.us
This message was composed almost entirely using NaturallySpeaking


From DOUGS at oceanic.com  Wed Jan 12 13:38:21 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Wed, 12 Jan 2000 08:38:21 -1000
Subject: Ping in Python
Message-ID: <5650A1190E4FD111BC7E0000F8034D26A0F250@huina.oceanic.com>

This gives me a chance to plug one of the new Python books, delivered just
yesterday from Amazon.  The quick skim looks very promising for 'Python
Annotated Archives' by Martin C. Brown, and the following leapt out;  from
Chapter 2, Networking, a section titled 'Emulating Ping'.  Good info that
answers some of the "how does a PING work" question and this quote:

"This script is part of the standard Python distribution and can be found in
the following location: Demo/sockets/udpecho.py".

I think on Windows you need to download the source to get it, but it was on
my RedHat system under /usr/doc/python-demos-1.5.2/Demo/sockets/udpecho.py.

Hope this helps,
-Doug-

> -----Original Message-----
> From: Anders M Eriksson 
> [mailto:anders.eriksson at morateknikutveckling.se]
> Sent: Wednesday, January 12, 2000 6:16 AM
> To: python-list at python.org
> Subject: Ping in Python
> 
> 
> Hello!
> 
> Everyday about 1.30 pm my Intenet just dies. Everything takes forever
> and IE nearly almost will TimeOut on all pages. So I need to test and
> document this so I can tell my ISP that there is something wrong.
> 
> So I thought: Why don't I create an Python program that will PING a
> list of hosts and log the response! 
> 
> But how does a PING work?
> 
> tips, articles, source code will be very wellcome!!
> 
> // Anders
> 
> ---
> Enlish isn't my first or second language, 
> so anything that you find offending is not there on 
> purpose it's just due to the translation ;-)
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From roy at popmail.med.nyu.edu  Thu Jan 20 23:02:43 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Thu, 20 Jan 2000 23:02:43 -0500
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com>
Message-ID: <roy-11C630.23024320012000@netnews.nyu.edu>

"tye4" <tye4 at yahoo.com> wrote:
> I don't oppose indentation either. It's good for program readability, and
> readability only. It shouldn't be part of the syntax. Whitespace end
> delimiters are invisible and hence confusing and prone to error.

When I was first learning programming (a quarter of a century ago!), 
there was the idea of the "telephone test", i.e. your code should be 
simple enough that if you read it to somebody on the phone they would 
understand it.  Python not only fails the phone test, it fails the 
printed page test; I can print two programs on a piece of paper, which 
differ only in tabs and spaces, and they will appear identical yet which 
do different things.  Barf.

I'm in love with python, but this idea of white space as significant 
syntax is the biggest wart on the language by far.


From chuck.meyers at lmco.com  Mon Jan 17 18:07:49 2000
From: chuck.meyers at lmco.com (Chuck Meyers)
Date: Mon, 17 Jan 2000 15:07:49 -0800
Subject: Is there a tool like VIB in ICON?
Message-ID: <3883A0C5.4494F4B3@lmco.com>

Does python have a tool similar to VIB (Visual Interface Builder) in
ICON. VIB looks like a drawing program for drawing GUIs, but it produces
ICON code for the GUI.



From kc5tja at garnet.armored.net  Mon Jan 10 01:56:46 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 06:56:46 GMT
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu> <slrn87fol9.15t.kc5tja@garnet.armored.net> <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk> <slrn87imad.thc.wtanksle@hawking.armored.net>
Message-ID: <slrn87j11s.vc0.kc5tja@garnet.armored.net>

>>I tried com for many years without coming close to understanding it. If
>>M$ continues to have huge leverage it will continue to be arcane and for
>>wizards only!

I don't know why, but I keep missing the original messages.  I'm starting to
get annoyed with our newsserver... *sigh*

COM is exceptionally //easy// to understand.  If you can't understand it,
it's solely because you are deliberately making it more complicated than it
really is.

I'll be blunt:

	COM objects do not follow the traditional, "politically-correct"
	class-hierarchy object model that most other programming languages
	follow.

Once you get the idea that an "object" is something which belongs to a
class, and "is-a" something else, out of your mind, you can move on to the
next big revelation, which is a vastly more powerful concept anyway:

	An object is a thing.  It has an identity.  It can do stuff.

Note that nowhere does it indicate that it belongs to a class.  Nowhere does
it indicate that it exhibits an is-a relationship with another class (how
can it, if it doesn't belong to a class of objects)?

Let me give you a real-world example.  Your computer monitor has a viewing
surface on it; it could be an LCD screen, or it could be a CRT tube.  It
doesn't matter.  Likewise, your TV set also has a display surface on it.  It
follows that a TV "is a" monitor.  After all, it lets you see ("monitor")
things, no?

But doesn't a sheet of paper /also/ exhibit a viewing surface?  It too is an
object that exposes the same basic functionality as a computer monitor or TV
set -- you can SEE things on a sheet of paper.  Yet, a sheet of paper is not
a monitor.

Here, we have an example of an object which exhibits pretty much the same
functionality (interfaces) as a TV, yet is used for a radically different
purpose.

In terms of COM, we could define the following interfaces using some
hypothetical interface description language:

	interface IViewingSurface:
		def See( self ): returns None

Now, the paper object can define the See method to expose the same image
over and over again.  It's a completely static image.  Yet, our TV object
can define the See method to render a different image everytime it's called,
depending on what channel it's tuned to, and what frame of video is being
transmitted.

Now, in order to actually change the channels, we need another interface --
the knobs:

	interface ITVKnobs:
		def SetChannel( self, channel ): returns None
		def GetChannel( self ): returns channel
		def SetVolume( self, volume ): returns None
		def GetVolume( self ): returns volume

The TV object would implement the ITVKnobs interface, whereas our sheet of
paper wouldn't -- you cannot change the channel on a sheet of paper
(remember: we're dealing with a sheet of paper object, not a book object).
And even if we could, how could we change the volume of a sheet of paper?

COM eschews the concept that each object "belongs" to one or more classes
because, in reality, objects rarely do belong to a finite set of
classes.  It simply doesn't make sense to attempt to be academically correct
when, in the real world, it has little applicability.  (Aside: It is pretty
much a proven track record that real-world issues and academic solutions
rarely, if ever, meet in the middle.)

Instead, COM concentrates on the concept that an object just "does it" (to
steal an ad from Nike).  It is up to the object itself to determine which
interfaces it exposes.  It is merely the collection of interfaces it
supports that determines its type.  For example, we know that if we receive
an object that supports the IViewableSurface interface, we know that it's
some kind of viewable surface.  If it ALSO supports the ITVKnobs interface,
we know unambiguously that it is, indeed a TV.  Never mind whatever else it
supports -- the fact that it supports those two interfaces makes it a TV.
Period.  No questions asked.

Please don't take this the wrong way -- it's not meant as an offense.  But
if you get your mind out of the gutter for a bit, and realize that this
whole "object oriented" thang has been advocated pretty much by the academia
alone, you'll quickly realize why OO is the way it is: in order to save
time, students draw from prior art when writing their theses.  As a result,
entire WORLDS of alternative, yet perfectly valid, forms of object
orientation have gone unexplored.  Again, please don't take the above as an
offense.  I just don't know how else to express it.

>Keep in mind the fact that although working with COM may be simple (and
>Sam is the authority on this, since he has implemented a reasonably
>complete implementation, with a start at documentation), working with
>Windows involves more than COM -- you've gotta work out the whole
>API/framework classes.

Precisely.  In fact, the COM API is by far the simplest and most well
designed Windows API I'd ever seen.  Everything else is horribly written,
with inconsistent error handling techniques, poor nomenclature resulting in
similar-sounding API functions, but radically different jobs, and worse,
several functions, all with similar signatures, all of which do largely the
same thing -- a hallmark of very poor factorization.

And COM objects themselves are quite easy to work with.  What can be easier
than a table of function pointers?  Ever worked with the Berkeley DB
interface?  There's a perfect example of a COM-like object that *ISN'T* from
Microsoft, yet everybody and their grandmother hails Sleepycat for their
superior database API.

All COM _really_ does is give a standard layout to the first three function
table entries of each object's interface.  That's it!

>>Python at least allowed me in without a lot of exotic
>>verbiage.

Then I'd suspect you're a neophyte to object oriented technology.  Again, no
offense towards you.  But it seems that you're not well versed in object
oriented concepts, or your fresh out of, or still in, college.  The concept
behind COM is one of the easiest to understand concepts, addressing all the
complicated issues behind multiple inheritance and more, in a simple,
easy-to-read specification document.  A document which, by the way, is open
to the public.  No secret handshakes or high wizardry here.  In fact, I'll
even give you the URL to the COM specification, so you can read it yourself:

	http://msdn.microsoft.com/library/specs/S1D139.HTM

Or you can get a printable copy (Microsoft Word format) from:
	
	http://www.microsoft.com/com/resources/specs.asp

In fact, these are the documents I'm using to implement my GCOM package for
Linux and Dolphin.

Note that these documents describe the COM 0.9 specification.  This is the
last specification to be in the hands of Microsoft.  1.0 was donated to the
Distributed Computing Environment (DCE), thus making COM a commercially open
standard.

But you don't need DCE to get the latest spec, either.  Just browse
Microsoft's MSDN on-line library, and you'll get ALL the latest
specifications for the latest version of COM, including the interfaces used
for asynchronous, non-blocking method calls, transaction services,
MAPI/TAPI/DirectX, and more.  So starting with the 0.9 spec, and
incorporating the more modern APIs and COM objects documented in the other
on-line manuals will give you, pretty much, COM+ without ever having to
spend a dime on Windows 2000.

>>I'm not a whizz at COM, but Bill Tutt, Hark Hammond and the
>>other COMerati regularly throw up their hands at the complexity and
>>opacity of it all.

And you don't get this with CORBA?  Or with SOM?  Or with binaries written
that you don't have source code to?  I have experience with CORBA, and let
me tell you, it ain't pretty.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From drek at MonsterByMistake.Com  Mon Jan 24 12:13:38 2000
From: drek at MonsterByMistake.Com (Agent Drek)
Date: Mon, 24 Jan 2000 12:13:38 -0500 (EST)
Subject: OFFTOPIC: *-ly y'rs, 
In-Reply-To: <pcobt6c635a.fsf@math.ntnu.no>
Message-ID: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com>

I know this is a stupid question but at least I'm not complaining
about delimiters :) here goes ...

I have only ever seen these sign-offs on this list:

*-ly y'rs

Where did they come from and are they patented :)
How can one join the signoff-ly club?

these-are-super-fun-ly y'rs,

    derek

--
Monster By Mistake Inc > 'digital plumber'
http://www.interlog.com/~drek




From cjc26 at nospam.cornell.edu  Sat Jan 29 20:45:13 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sun, 30 Jan 2000 01:45:13 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <slrn88n40a.2qh6.cjc26@bridget.mindriot.net> <948708715.22839@news.adtranzsig.de> <slrn891l2f.5tk.cjc26@bridget.mindriot.net> <86se6p$qu9$1@nntp5.atl.mindspring.net>
Message-ID: <slrn899qb3.1ai.cjc26@bridget.mindriot.net>

Pada 28 Jan 2000 15:56:09 GMT, Aahz Maruch bilang:
| >
| >	kicked = lambda(X) (lambda(Y) (if Y kicked X: 1
| >	                               else: 0))
| 
| My head hurts.

And you're a programmer.  Imagine a roomful of linguists and
philosophers trying to grasp the concept of lambda.  That's what
my semantics class last year was like. :)


-- 
cliff crawford  -><-  http://www.people.cornell.edu/pages/cjc26/
                     "All I need is a small town
                      with good Internet connectivity" - Nadine


From hei at adtranzsig.de  Tue Jan 25 04:02:10 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 25 Jan 2000 10:02:10 +0100
Subject: tabs do WHAT?
References: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net> <86h5ep$iq$1@mach.vub.ac.be> <948721268.245257@news.adtranzsig.de> <pcoya9f4bza.fsf@math.ntnu.no>
Message-ID: <948794680.612775@news.adtranzsig.de>

Harald Hanche-Olsen schrieb in Nachricht ...
>for me.  The editor also has features for indenting or outdenting
>blocks of code to the appropriate level for those circumstances when
>you change the structure of the code or copy blocks of code from one
>place to another place with a different indentation leve.  Use these
>features, and all concerns about the unfriendliness of this or that
>indentation scheme, parentheses or curly brackets etc melt away like
>icicles in the sun.

<evil grin> Can it... fold?
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From emwpeso at emw.ericsson.se  Thu Jan 27 03:53:57 2000
From: emwpeso at emw.ericsson.se (Per Soderstam)
Date: Thu, 27 Jan 2000 09:53:57 +0100
Subject: Installing quirk...
Message-ID: <389007A5.6B22105C@emw.ericsson.se>

Hi!

I'm trying to build the Python1.5.2 release on a SPARC5 workstation
running Solaris5 and i want another functionality for my pyhon promt
(tsch type, with the up/down/left/right arrow positioning and command
history as well as backspace instead of delete). Setting the SHELL
(../tcsh) variable does not seem to work, but I'm no guru :). Do anyone
understand my question? If so, do anyone have an answer?

/Per
-- 
Per S?derstam, Ericsson Microwave Systems AB, SE-431 84 Molndal, SWEDEN, 
tele: (+46) (0)31 7470195, e-mail: per.soderstam at emw.ericsson.se
********************************************************************
The views expressed above is my own and does not nessecarily
reflect the policy of my employer. http://www.ericsson.se/microwave/


From guido at CNRI.Reston.VA.US  Thu Jan 27 11:23:37 2000
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Thu, 27 Jan 2000 11:23:37 -0500
Subject: Call for Participation: Open Source Convention in Monterey (July)
Message-ID: <200001271623.LAA03648@eric.cnri.reston.va.us>

I received the following request from O'Reilly.  The deadline is very
tight (Feb 1); this is an exciting opportunity to teach other open
sourcers about Python!

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

------- Forwarded Message

From:    Vee McMillen <vee at oreilly.com>

Hello,

We are  accepting proposals for tutorial sessions and
conference talks for the Open Source Convention in Monterey,
July 17-20, 2000.  If you are interested in presenting, please
go to the following web sites and follow the instructions for
submitting a preliminary proposal:

Apache, Linux, Python, Tcl and Open Source Business:
http://conferences.oreilly.com/oscon2000/call.html

Hope to see you in Monterey,

Vee McMillen
Speaker Coordinator


------- End of Forwarded Message




From skip at mojam.com  Fri Jan  7 08:44:51 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 7 Jan 2000 07:44:51 -0600 (CST)
Subject: Text indexing and searching (was "Re: Python stuff")
In-Reply-To: <3875813C.A2F29ECB@trojanslair.zzn.com>
References: <3875813C.A2F29ECB@trojanslair.zzn.com>
Message-ID: <14453.60883.281202.910741@beluga.mojam.com>

    keagan> Thanks for replying to my post on the newsgroup, but as i said,
    keagan> i am a newbie. i didn't understand all of your post. I was
    keagan> wondering if i could creat a dictionary that would search
    keagan> through a file to find matches:

Well, indirectly.  I'm not much of a text search person, but here's one
approach, assuming you're only interested in whole words.  Suppose your file
was

The quick brown fox ...
My country 'tis of thee ....
Mavis told the commissioner the brown fox was too slow.

You could create a dictionary whose keys are the words in the file and whose 
values are lists of offsets into the file.  Let's also assume
case-insensitive searching, so you might have a dictionary that looks like

{
    'the': [0, 64, 81],
    'quick': [4],
    'brown': [10, 85],
    etc.

}

Then, if someone searched for "brown", you would access the dictionary using
"brown" as the key.  This would return the list [10, 85].  You'd then
highlight the words at offsets 10 and 85.

Obviously, for large files, that dictionary could get very big, and most of
the time you wouldn't use much of it.  The natural way to store it is in a
dbm file of some sort, which acts pretty much like a dictionary but is
stored on disk.

Here's a simpleminded script that takes a single text file and creates an
index such as the one I described.

    #!/usr/bin/env python

    """given a file in sys.argv[1], create a keyword-in-context index in the
    file in sys.argv[2], e.g.:

	build-offset.py my-story.txt my-story-index.db

    """

    import re, shelve, string, sys

    # note the conspicuous lack of any error checking!
    f = open(sys.argv[1])
    db = shelve.open(sys.argv[2])
    wordpat = re.compile('(\W+)')

    print "indexing", sys.argv[1], "into", sys.argv[2]

    offset = 0
    wordchars = string.letters + string.digits
    while 1:
	line = f.readline()
	if not line: break
	line = re.split(wordpat, string.lower(line))
	# pick through the resulting list of words and non-words registering
	# offsets...
	for i in range(len(line)):
	    if line[i] and line[i][0] in wordchars:
		lst = db.get(line[i], [])
		lst.append(offset)
		db[line[i]] = lst
	    offset = offset + len(line[i])

    f.close()

    # demonstrate that we have a valid db...
    keys = db.keys()
    keys.sort()
    for k in keys:
	print "%s: %s" % (k, db[k])
    db.close()

For the three-line file above, when run it displays:

    indexing fox.txt into fox.db
    brown: [10, 85]
    commissioner: [68]
    country: [27]
    fox: [16, 91]
    mavis: [53]
    my: [24]
    of: [40]
    quick: [4]
    slow: [103]
    the: [0, 64, 81]
    thee: [43]
    tis: [36]
    told: [59]
    too: [99]
    was: [95]

(These offsets all assume Unix line endings (LF only).  On a PC you will see
slightly different values because of the CRLF line endings.)

One thing that's also worth noting...  When you're looking for help with
something, it's best to followup to the list to which you originally posted
your question.  There are plenty of people out there who can help.  By
limiting your followups to the first person who happens to respond, you lose
the collective benefit of the community as a whole.  In particular, I would
be quite surprised if someone on the list didn't have a better idea about
text searching or even pointers to modules that solve the problem in a much
better way than I've outlined.  Accordingly, I've cc'd
python-list at python.org to shed some more light on the subject.  (The tutor
list (python-tutor at python.org) is also an excellent source of help.)

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 juno79 at my-deja.com  Mon Jan 17 23:37:09 2000
From: juno79 at my-deja.com (juno79 at my-deja.com)
Date: Tue, 18 Jan 2000 04:37:09 GMT
Subject: Csound-The World's Most Powerful Synthesizer & DSP Software
Message-ID: <860qlj$evg$1@nnrp1.deja.com>

Dear hackers...

Turn your computer into the ultimate software synthesizer and FX processor. 
Finally, the world's most powerful software synthesis and signal processing
language, Csound,developed at the MIT Media Lab, and used in major
universities around the world, can be yours.  Edited by Dr.Richard Boulanger,
a professor of Music Synthesis at The Berklee College of Music in Boston, The
Csound Catalog with Audio is an HTML-based CD-ROM with over 2000 incredible
sounding Csound instruments and compressed audio files for each to show you
the full range of possibilities - and these "patches" are all ready for you
to modify and personalize.Everything you need to learn, use and ultimately
master Csound is included on this single ISO/Hybrid CD-ROM for Mac
OS,Windows,Linux,and BeOS.  All the manuals, dynamically linked tutorials,
fully functional versions of this incredible software, and much much more. 
And you won't believe the price - $19.95 plus shipping and handling. Check it
out and order your copy today! You will receive your order in about a week.

By the way, there are two versions of this product : QDesign (you need Apple
QuickTime) for Mac OS and Windows,MP3 for Linux and BeOS (MP3 version is not
available yet,but it is coming soon)

Still need more info for Csound? Check out Csound Front Page @ MIT Press :
http://www.csound.org/

To purchase your copy of The Csound Catalog with Audio:
http://www.csounds.com/catalog/

P.S:What the hell is csounds.com? Leave your e-mail @ http://www.csounds.com/



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


From djc at itga.com.au  Thu Jan  6 23:50:33 2000
From: djc at itga.com.au (Dave Cole)
Date: 07 Jan 2000 15:50:33 +1100
Subject: Help with sockets!
References: <853jd1$etf$1@mulga.cs.mu.OZ.AU>
Message-ID: <871z7upjs6.fsf@heresy.itga.com.au>

> I've started tinkering with sockets in Python and have the following
> modified version of the example program in the library docs.
> 
> This is intended to forward everything from a port on one machine to
> a telnet port on another machine.
> 
> The program works fine except that when I terminate the connection the
> socket remains in use:
> 
> prompt% ./redir
> Listening for connections on X port 60016
> Connected by ('X.X.X.X', 62268)
> Forwarding to 23 on X
> Closing connections. Bye.
> prompt%
> prompt% ./redir
> Listening for connections on X port 60016
> Traceback (innermost last):
>   File "./redir", line 16, in ?
>     s.bind(FORWARD, PORT)
> socket.error: (125, 'Address already in use')
> 
> Any ideas why this isn't working? The sockets are being closed before
> the program terminates.

Try adding the following line:

> --- CUT HERE ---
> #!/usr/local/bin/python
> # Echo server (p143 Python lib man)
> from socket import *
> import select
> # from select import *
> 
> def sock_avail(x):
>         """Test whether data is available on the socket."""
>         return select.select([x], [], [], 0) == ([x], [], [])
> 
> FORWARD = 'host1'
> DEST = 'host2'
> PORT = 60016
> print 'Listening for connections on', FORWARD, 'port', PORT
> s = socket(AF_INET, SOCK_STREAM)

s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)

> s.bind(FORWARD, PORT)
> s.listen(1)
> conn, addr = s.accept()
> print 'Connected by', addr
> try:
>         s1 = socket(AF_INET, SOCK_STREAM)
>         s1.connect(DEST,23) # telnet port
>         print 'Forwarding to 23 on', DEST
>         while 1:
>                 if sock_avail(conn):
>                         data = conn.recv(1024)
>                         s1.send(data)
>                 if sock_avail(s1):
>                         data = s1.recv(1024)
>                         conn.send(data)
> except error:
>         s.close()
>         conn.close()
>         s1.close()
>         print 'Closing connections. Bye.'

- Dave


From fredrik at pythonware.com  Fri Jan 21 07:03:57 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 13:03:57 +0100
Subject: Screen depth
References: <2B1262E83448D211AE4B00A0C9D61B03BA7075@MSGEURO1>
Message-ID: <002c01bf6407$99b8b170$f29b12c2@secret.pythonware.com>

Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> how can I determine how many colors a screen can show?

it depends on the platform and what user interface
toolkit you are using.

</F>




From aloy at ctv.es  Wed Jan 26 14:17:47 2000
From: aloy at ctv.es (Antoni Aloy)
Date: Wed, 26 Jan 2000 20:17:47 +0100
Subject: Is there a text mode user interface for python?
Message-ID: <86q1bt$fkm$1@lola.ctv.es>

Hi!

I'm looking for a text mode user interface for Python, that works with
M$Windows.

Any help would be greatly appreciated!

Antoni Aloy




From s.schwarzer at ndh.net  Sat Jan  1 17:12:36 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Sat, 01 Jan 2000 23:12:36 +0100
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl>
Message-ID: <386E7BD4.514764EA@ndh.net>

Hello all,

Boudewijn Rempt schrieb:
> Yes, I read that too, a while ago, and it got me thinking about what
> a language like he proposes would look like. He made two striking
> suggestions - being able to draw part of the logic, in the sidebar of the
> code, as it were, and having a sort of underlying representation of the
> code (he said xml, but that's what everyone is saying these days),

Because "everyone" sais it, it doesn't necessarily mean that it is
bad ;-) .

> so that
> a particular development tool could show the code in various helpful ways.
> 
> I think, however, that he has confused the language with the creation
> and maintenance tool (I hesitate to call it an editor anymore), Already
> in Steve MacConnels Code Complete, there was a proposal for an ideal
> programming environment. I do think even the best editors aren't
> getting close to the sophistication enjoyed by users of basic stuff
> like webbrowsers (the KDE2 general browser Konqueror is incredibly
> sophisticated) or office suites - so there he has a point.
> [...]
> I'd like to write a programming environment that not only included
> syntax highlighting, but also linking to documentation, a rich
> presentation of the program and its logic, but I suspect programmers
> will always go back to the ascii source.

This separation of code and maintanance tool is in part already
common: think of folding editors. An advantage of these is that you
can still view the code with a "normal" text editor and this
advantage should remain in future ideas. XML may be a good starting
point.

Of course, this approach rises the complexity of the "editor", and
would (as said) lead to a distinction of source and its view. On the
other hand, you can _still_ use a "usual" ASCII editor, in the same
way you can edit HTML with any text editor. The kind of editor you
use (plain text or a special program browser/editor) would be optional.

The other topics which the author on the O'Reilly site, Greg Wilson,
mentions might also be worth considering. For example, I never needed
tuples (explicitly) yet, i. e. in a sense that lists would not work.
(If tuples would be taken out of the language or their use deprecated)
tuple un/packing would become list un/packing and so on.)

The idea to provide "true" multidimensional arrays sounds also very
good to me. Treating multidimensional arrays as lists of lists (or
vice versa) complicates the use of them, and may it be that a
special wrapper module is needed to provide the corresponding
funtionality.

Stefan


From eddyw at lsl.co.uk  Fri Jan 21 11:44:35 2000
From: eddyw at lsl.co.uk (Edward Welbourne)
Date: Fri, 21 Jan 2000 16:44:35 +0000
Subject: Forcing exit(); unqualified exception catchers
Message-ID: <E12BhAp-0005EL-00@lsls4p>

Note that calling _exit() means any dangling `finally:' clauses will get
skipped.  If these were just waiting to close a few files, no problem;
the system will do so for you.  However, if you had (say) some files
locked by a dumb-but-portable locking scheme that creates a lock-file
you are beholden to tidy away in a finally: clause, you have a problem.
Likewise if you've got any kind of persistent resource claimed.

Most likely, though, your problem is that your code is using unqualified
exception catchers - that is,

try:
    # ...
except:
    # ...

so that your SystemExit exception is getting caught and discarded - as
opposed to

try:
    # ...
except AnticipatedError:
    # ...

Use of unguarded except: is commonly a result of laziness, e.g.

try: 
    n = string.atoi(text)
except:
    n = 0

where someone `knows' that if an exception got raised it can only have
been the ValueError and they know what to do in that case ... except
that if the user tries to interrupt the program (KeyboardInterrupt)
during the execution of string.atoi, the program will ignore the user
and act as though the text had been unreadable.  This trains users to
type ^C^C^C^C^C^C^C^C in frustration, while pieces of your process take
it in turn to do fragments of their jobs, potentially making the mess
worse and worse at each layer ...

That's the noddy case; the more severe cases involve someone not knowing
*which* exception might be coming back from something they're doing, or
maybe there are several possibilities and they can't be bothered to work
out an exhaustive list of them, but again they presume that any
exception raised is `something wrong with the args I passed down' and
they think they know what to do in such a case.  So they just catch all
exceptions and totally ignore the fact that something down under there
might have legitimate grounds for trying to exit (say), or (again) the
user might KeyboardInterrupt.

Note that many programmers believe that the relevant piece of code is so
quick they can rely on the interrupt not happening then.  This is, of
course, a superstition: the interrupt is going to happen somewhen and
the universal law of natural cussedness says the probability of it
happening in any given piece of code is directly proportional to the
amount of grief that will arise if it does - and totally independent of
the length of time the relevant code takes to run.

In general, if you use a unqualified except: clause, ask yourself:
  * if the user tries to interrupt the process, and by ill luck does so
    during this operation, do I really want to ignore the user's request ?
  * if some subordinate operation I haven't anticipated works out that
    it *has* to exit *now*, do I really believe I know better ?

Note that, where several exceptions might get raised, you *can* specify
an except clause that catches them - by using a tuple of exceptions:

try:
    return db.connect(user, password)
except (TypeError, ValueError):
    return None

The other way one *might* approach this, particularly in the case where
many exceptions are possible, would be:

try:
    return db.connect(user, password)
except (KeyboardInterrupt, SystemExit), what:
    raise what
except:
    return None

which will ensure you propagate the ones you realise you don't mean to
lose while catching the rest.  However, you'll have to think long and
hard about your list of which exceptions to forward (e.g. how about
MemoryError ?), lest you go catching something that communicates a state
of affairs that's really so bad you don't want to be responsible for the
aftermath.

So: the real lesson here is, use SystemExit or exit(), not _exit(), and
fix the wanton except: clauses that are preventing your SystemExit from
having its intended effect.  That way, all your finally: clauses will
get to do their stuff.

	Eddy.



From tjg at avalongroup.net  Tue Jan 18 19:42:38 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Tue, 18 Jan 2000 16:42:38 -0800
Subject: More DBI question
References: <1263885125-1047272@hypernet.com>
Message-ID: <3885087E.91A8A205@exceptionalminds.com>

Gordon McMillan wrote:
> > Howevever, the second time through this sequence, the first execute
> > returns an error along the lines of "Expecting 3 arguments received 1"
> >
> > Now I understand where the three arguments part comes from, the cursor
> > just did something that required three arguments, but how do I "reset"
> > what the cursor expects, it seems clunky to call a db.commit() after
> > each execute.
 
> As you might expect by now, much depends on what DB,
> what driver and what DBI implementation you're using. I
> suggest, however, closing a cursor once you're done with an
> execute and using a new cursor.

Thanks Gordon, I appreciate the input. I'm am now going through my code
and closing all cursors after I get the result table.

As I stop to think about it though, If I were to need to make a
different query mid-fetch(), I would then need to open a separate cursor
and use it. I'm significantly new enough to this type of DB programming
that I just don't know what is good style and what is not.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From mmueller at dgfz.de  Mon Jan 31 06:30:03 2000
From: mmueller at dgfz.de (Mike Mueller)
Date: Mon, 31 Jan 2000 12:30:03 +0100
Subject: sending keystroke to a dos application
Message-ID: <3.0.6.32.20000131123003.009693c0@orion.dgfz.de>

I have a dos application that runs in a dos window under NT or win98 that
takes interactive input. Basically numbers to choose from menu items and
return keys. I have not found a method to script this application by
sending keystrokes to the dos window. Is there a any possibility to do this
with python?
Does the dos window offer COM support I could use to pass some keystrokes?

Thanks 
Mike

---------------------------------------------------------
	Dipl.-Ing. Mike Mueller, M.Sc.
	Dresden Groundwater Research Center
	Meraner Str. 10
	D-01217 Dresden
	Germany
Tel.:  	+49 351/4050675
Fax.:  	+49 351/4050679
e-mail: mmueller at dgfz.de
http://www.dgfz.de
---------------------------------------------------------- 



From newshoes at mail.com  Wed Jan 19 17:00:50 2000
From: newshoes at mail.com (NewShoes)
Date: 19 Jan 2000 22:00:50 GMT
Subject: Python Rocks! (delimiters)
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com> <8653jg$esi@news.or.intel.com> <002401bf62b8$c30da8a0$f29b12c2@secret.pythonware.com> <8656ne$h1j@news.or.intel.com> <slrn88c7qh.ja9.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>
Message-ID: <8EC0866A9newshoesmailcom@205.147.63.94>

i think the original recommendation was that if an 'END'
delimiter was used, python would be nit-picky about
indentation.
while this seems a decent proposal to make two camps
semi-happy at first, it would be approximate to a
'very bad thing' when using nested ifs/whiles/etc
to have an either-or blocking method



From Dan at Grassi.com  Fri Jan  7 16:05:52 2000
From: Dan at Grassi.com (Dan Grassi)
Date: Fri, 07 Jan 2000 16:05:52 -0500
Subject: Help: does a variable exist
Message-ID: <B49BBF60.1373%Dan@Grassi.com>

I have forgotten and can not find the function (incantation) that can will
determine if a variable currently exisis.

Thanks,

Dan



From jjlucsy at concentric.net  Tue Jan 18 08:42:44 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Tue, 18 Jan 2000 08:42:44 -0500
Subject: PythonWin and Dialogs in DLLs
References: <38839d87$0$22473@senator-bedfellow.mit.edu> <388418ec.8757803@news.ricochet.net>
Message-ID: <k5_g4.3484$iy5.225935@typ12.deja.bcandid.com>

> 2.  Add a file called say frog.c containing something like
> #include <windows.h>
> BOOL WINAPI DllMain( HINSTANCE hInstance, DWORD dwReason, LPVOID frog)
> {
>     return 1;
> }
> i think the DLL loader needs an entry point called DllMain and the
> parameters will allow you to check for stuff, but i don't know why
> that would be useful; anyway it doesn't cost much, and using a
> resource DLL is a fairly standard thing.

Nope, simply add /noentry in the link tab settings. No DllMain/WinMain/or
main required.





From sanderson at ttm.com  Thu Jan 13 15:30:33 2000
From: sanderson at ttm.com (Scott Anderson)
Date: Thu, 13 Jan 2000 15:30:33 -0500
Subject: Doing things in one line?
References: <60A42251CAC873748525686500604B69.00604BD385256865@ttm.com>
Message-ID: <387E35E9.30C8D0BD@ttm.com>

mwh21 at cam.ac.uk wrote:
[...]
> > reduce(lambda x, y: x + y, range(1, int(raw_input('Enter a number: ')) +
[...]
> > It accepts a number from the user and adds 1 to it. :-)
> >
> 
> No it doesn't. It computes the nth "triangular number", where n is the
> number the user enters.

My bad. Slight brain dysfunction during thought completion.  That's what
I meant to say.

Regards,

-scott



From pf at artcom0.north.de  Mon Jan 17 02:39:21 2000
From: pf at artcom0.north.de (Peter Funk)
Date: 17 Jan 2000 08:39:21 +0100
Subject: fintl.py -- a python module to aid i18n (was Re: International Python)
References: <oqk8ldg2jp.fsf@titan.progiciels-bpi.ca>
Message-ID: <85ugv9$gtb$1@artcom0.north.de>

In <oqk8ldg2jp.fsf at titan.progiciels-bpi.ca> 
	=?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> writes:
>Things are improving in the area of Python internationalisation! :-) After
>having made PO mode able to mark strings for translation in a collection
>of Python sources, I finally got `xpot' to extract those strings into a
>neater POT file.  I'm grateful to Barry Warsaw for his `pygettext' tool,
>which has been useful to cross-check the results of 'xpot' and debug it.

Seeing this subject I like to use it as a shameless plug to release
the following module, which I've written to aid the i18n of my Python apps.

I'm still seeking for hints how to mimic the locale functionality on the
WinXX and Mac platforms.  However on Linux it works as is.  Have fun.

Regards, Peter
-- 
Peter Funk, Oldenburger Str.86, D-27777 Ganderkesee, Germany, Fax:+49 4222950260
office: +49 421 20419-0 (ArtCom GmbH, Grazer Str.8, D-28359 Bremen)
---- 8< ---- 8< ---- cut here ---- 8< ---- schnipp ---- 8< ---- schnapp ----
#!/usr/bin/env python
"""i18n (multiple language) support.  Reads .mo files from GNU gettext msgfmt

If you want to prepare your Python programs for i18n you should 
add the following lines to the top of a BASIC_MAIN module of your py-program:
    try:
	import fintl
	gettext = fintl.gettext
	fintl.bindtextdomain(YOUR_PROGRAM, YOUR_LOCALEDIR)
	fintl.textdomain(YOUR_PROGRAM)
    except ImportError:
	def gettext(msg):
	    return msg
    _ = gettext
and/or also add the following to the top of any module containing messages:
    import BASIC_MAIN
    _ = BASIC_MAIN.gettext
	    
Now you should use _("....") everywhere instead of "...." for message texts.

Once you have written your internationalized program, you can use
the suite of utility programs contained in the GNU gettext package to aid
the translation into other languages.  

You ARE NOT REQUIRED to release the sourcecode of your program, since 
linking of your program against GPL code is avoided by this module.  
(Although it is possible to use the GNU gettext library by using the 
intl.so module written by Martin von L?wis if this is available, it is 
not required to use it in the  first place)
"""
# Copyright 1999 by pf at artcom-gmbh.de (Peter Funk)
#  
#                         All Rights Reserved
#
# Permission to use, copy, modify, and distribute this software and its
# documentation for any purpose and without fee is hereby granted,
# provided that the above copyright notice appear in all copies.

# ArtCom GmbH AND Peter Funk DISCLAIMS ALL WARRANTIES WITH REGARD TO
# THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
# AND FITNESS, IN NO EVENT SHALL ArtCom GmBH or Peter Funk BE LIABLE
# FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
# WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN
# AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
# OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.

_default_localedir = '/usr/share/locale'
_default_domain = 'python'

# check out, if Martin v. L?wis 'intl' module interface to the GNU gettext
# library is available and use it only, if it is available: 
try:
    from intl import *
except ImportError:
    # now do what the gettext library provides in pure Python:
    error = 'fintl.error'
    # some globals preserving state:
    _languages = []
    _default_mo = None # This is default message outfile used by 'gettext'
    _loaded_mos = {}   # This is a dictionary of loaded message output files

    # some small little helper routines:
    def _check_env():
        """examine language enviroment variables and return list of languages"""
	languages = []
	import os, string
	for envvar in ('LANGUAGE', 'LC_ALL', 'LC_MESSAGES', 'LANG'):
	    if os.environ.has_key(envvar):
		languages = string.split(os.environ[envvar], ':')
		break
        # use locale 'C' as default fallback:
	if 'C' not in _languages:
	    languages.append('C')
	return languages

    # Utility function used to decode binary .mo file header and seek tables:
    def _decode_Word(bin):
        # This assumes little endian (intel, vax) byte order.
	return  ord(bin[0])        + (ord(bin[1]) <<  8) + \
	       (ord(bin[2]) << 16) + (ord(bin[3]) << 24)

    # Now the methods designed to be used from outside:

    def gettext(message):
        """return localized version of a 'message' string"""
	if _default_mo is None: 
	    textdomain()
	return _default_mo.gettext(message)

    _ = gettext

    def dgettext(domain, message):
        """like gettext but looks up 'message' in a special 'domain'"""
	# This may useful for larger software systems
	if not _loaded_mos.has_key(domain):
	    raise error, "No '" + domain + "' message domain"
	return _loaded_mos[domain].gettext(message)

    class _MoDict:
        """read a .mo file into a python dictionary"""
        MO_MAGIC = 0x950412de # Magic number of .mo files
	def __init__(self, domain=_default_domain, localedir=_default_localedir):
	    global _languages
	    self.catalog = {}
	    self.domain = domain
	    self.localedir = localedir
	    # delayed access to environment variables:
	    if not _languages:
	        _languages = _check_env()
	    for self.lang in _languages:
		if self.lang == 'C':
		    return
		mo_filename = "%s//%s/LC_MESSAGES/%s.mo" % (
						  localedir, self.lang, domain)
		try:
		     buffer = open(mo_filename, "rb").read()
		     break
		except IOError:
		     pass
	    else:
		return # assume C locale
            # Decode the header of the .mo file (5 little endian 32 bit words):
	    if _decode_Word(buffer[:4]) != self.MO_MAGIC :
		raise error, '%s seems not be a valid .mo file' % mo_filename
	    self.mo_version = _decode_Word(buffer[4:8])
	    num_messages    = _decode_Word(buffer[8:12])
	    master_index    = _decode_Word(buffer[12:16])
	    transl_index    = _decode_Word(buffer[16:20])
	    buf_len = len(buffer)
	    # now put all messages from the .mo file buffer in the catalog dict:
	    for i in xrange(0, num_messages):
		start_master= _decode_Word(buffer[master_index+4:master_index+8])
		end_master  = start_master + \
		              _decode_Word(buffer[master_index:master_index+4])
		start_transl= _decode_Word(buffer[transl_index+4:transl_index+8])
		end_transl  = start_transl + \
		              _decode_Word(buffer[transl_index:transl_index+4])
		if end_master <= buf_len and end_transl <= buf_len:
		    self.catalog[buffer[start_master:end_master]]=\
				 buffer[start_transl:end_transl]
		else: 
		    raise error, ".mo file '%s' is corrupt" % mo_filename
		# advance to the next entry in seek tables:
		master_index= master_index + 8
		transl_index= transl_index + 8

	def gettext(self, message):
	    """return the translation of a given *message*"""
	    try:
		return self.catalog[message]
	    except KeyError:
		return message
	# _MoDict instances may be also accessed using mo[msg] or mo(msg):
	__getitem = gettext
	__call__ = gettext

    def textdomain(domain=_default_domain):
        """Sets the 'domain' to be used by this program. Defaults to 'python'"""
	global _default_mo
	if not _loaded_mos.has_key(domain):
	     _loaded_mos[domain] = _MoDict(domain)
	_default_mo = _loaded_mos[domain]

    def bindtextdomain(domain, localedir=_default_localedir):
	global _default_mo
	if not _loaded_mos.has_key(domain):
	    _loaded_mos[domain] = _MoDict(domain, localedir)
	if _default_mo is not None: 
	    _default_mo = _loaded_mos[domain]

def _testdriver(argv):
    message   = ""
    domain    = _default_domain
    localedir = _default_localedir
    if len(argv) > 1:
        message = argv[1]
	if len(argv) > 2:
	    domain = argv[2]
	    if len(argv) > 3:
		localedir = argv[3]
    # now perform some testing of this module:
    bindtextdomain(domain, localedir)
    textdomain(domain)
    info = gettext('')  # this is where special info is often stored
    if info:
	print ".mo file for domain %s in %s contains:" % (domain, localedir)
	print info
    else:
	print ".mo file contains no info"
    if message:
        print "Translation of '"+ message+ "' is '"+ _(message)+ "'"
    else:
        for msg in ("Cancel", "No", "OK", "Quit", "Yes"):
	    print "Translation of '"+ msg + "' is '"+ _(msg)+ "'"

if __name__ == '__main__':
    import sys
    if len(sys.argv) > 1 and (sys.argv[1] == "-h" or sys.argv[1] == "-?"):
        print "Usage :", sys.argv[0], "[ MESSAGE [ DOMAIN [ LOCALEDIR ]]]"
    _testdriver(sys.argv)



From gerrit.holl at pobox.com  Fri Jan 21 07:54:51 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 13:54:51 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <38866E35.F04D4ADE@ndh.net>; from s.schwarzer@ndh.net on Thu, Jan 20, 2000 at 03:08:53AM +0100
References: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse> <38866E35.F04D4ADE@ndh.net>
Message-ID: <20000121135451.B2265@stopcontact.palga.uucp>

Stefan Schwarzer wrote on 948334133:
> Hi Evguenii
> 
> Evguenii Smogailov schrieb:
> > Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.
> 
> I consider it a bad idea, to add synonyms for things that are
> already in the language. So _adding_ "greater" to the language (do you
> really find "greater" or just "gt" more readable as ">"?) is not
> the same as using "or" _instead_ of "||".

It would break code.

eq='aaa'

Wouldn't be possible anymore.

> Perhaps the ".." operator is something to consider (if it isn't
> incompatible with somethings else in the language), but most of
> the times "range" does it's job very well.

Unfortunately, not yet. It's quite hard to get a string which
would be a..k in Perl. Maybe a nice feature for range()?
>>> string.join(map(lambda i: chr(i), range(ord('a'), ord('k'))), '')
'abcdefghij'

Doesn't look very well, does it?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From tim_one at email.msn.com  Sun Jan 23 00:06:47 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 23 Jan 2000 00:06:47 -0500
Subject: Python dumps core with this.. how come?
In-Reply-To: <86di3r$tj0$1@nntp4.atl.mindspring.net>
Message-ID: <000701bf655f$a632aa60$732d153f@tim>

> ... he invited people to just *try* writing a Python program
> without using any __xxx__ hooks. ...

[Aahz]
> I assume that Guido's advice specifically excluded __init__;
> otherwise, it makes no sense to me.

No, he didn't specifically exclude it -- I suppose he figured he was talking
to grownups <wink>.

__init__-isn't-really-"a-hook"-ly y'rs  - tim





From drek at MonsterByMistake.Com  Mon Jan 17 16:21:05 2000
From: drek at MonsterByMistake.Com (Agent Drek)
Date: Mon, 17 Jan 2000 16:21:05 -0500 (EST)
Subject: Sucking e-mail from Exchange Server vis HTTP?
In-Reply-To: <slrn88702k.2fm.grant@grante.comtrol.com>
Message-ID: <Pine.BSF.4.10.10001171617110.22701-100000@bunny.monsterbymistake.com>

|I'm going to figure out how to get my mail without running
|Outlook if it takes hundres of years and costs thousands of
|lives...
|
|-- 
|Grant Edwards                   grante             Yow!  I want you to
|                                  at               MEMORIZE the collected
|                               visi.com            poems of EDNA ST VINCENT
|                                                   MILLAY... BACKWARDS!!

Keep fighting the good fight!

    =derek

Monster By Mistake Inc > 'digital plumber'
http://www.interlog.com/~drek




From Vladimir.Marangozov at inrialpes.fr  Wed Jan 26 06:16:18 2000
From: Vladimir.Marangozov at inrialpes.fr (Vladimir Marangozov)
Date: Wed, 26 Jan 2000 12:16:18 +0100
Subject: Global mutex/lock
References: <388CB723.7EE04D7C@ina.fr>
Message-ID: <388ED782.E1B8F00A@inrialpes.fr>

Richard Gruet wrote:
> 
> Hi Python gurus,
> 
> Is there a portable global mutex implementation somewhere ?

What do you mean by global?

> AFAIK, the locks in the threading/thread modules work only between
> threads of the same module.

They work for all threads within a given (Python) process, yes.

> OTOH, there are file locks for Windows and Unix (respectively in modules
> msvcrt and fcntl), that can serve as locks.

Yes. Locking a shared resource can be used for inter-process
synchronization.

> 
> Am I right or am I missing something ?
> 
> cheers,
> Richard

It depends on what you're looking for, but it sounds like you need
inter-process synchronization of some sort... Different solutions
exist, some are "portable", some aren't.

For more feedback, tell us more about your problem. :-)

-- 
       Vladimir MARANGOZOV          | Vladimir.Marangozov at inrialpes.fr
http://sirac.inrialpes.fr/~marangoz | tel:(+33-4)76615277 fax:76615252


From bjorn at roguewave.com  Wed Jan 19 13:04:55 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jan 2000 11:04:55 -0700
Subject: SQL escape sequence problems
References: <8644bj$lrd$1@snipp.uninett.no>
Message-ID: <3885FCC7.C6FEB0BE@roguewave.com>

I would probably do:

    sql = "SELECT FROM TABLE WHERE MESSAGEID='%s'" % MySQL.escape(father)

instead, since you really only want to escape the values you are inserting to
the database, and not the sql string itself...

-- bjorn


Asle Pedersen wrote:

> I'm trying to run the following code:
>
> father="something"
> sql="SELECT FROM TABLE WHERE MESSAGEID='%s'" % (father)
> MySQL.escape(sql)
>
> but there is a problem since ' is really an escape character. How can I get
> around this problem? I'm using MySQLmodule-1.2. running under win32
> enviroment. I have tried to get a compiled version of the MySQLdb module
> whithout any success.
>
> regards,
> Asle
>
> --
> http://www.python.org/mailman/listinfo/python-list




From bjorn at roguewave.com  Wed Jan 19 16:14:48 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jan 2000 14:14:48 -0700
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse>
Message-ID: <38862948.5AEE964C@roguewave.com>

Evguenii Smogailov wrote:

> On 19 Jan 2000, Stephen Coursen wrote:
> > When did this become a trait of Perl?  Gods, what's so hard about knowing
> > that == is the equality operator?
>
> Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.

I see... so you don't really want eq, gt, and ne; you want equal, greater, and
not equal.

add-1-to-cobol-giving-cobol'ly y'rs
-- bjorn





From mikael at isy.liu.se  Thu Jan 20 04:47:43 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 20 Jan 2000 10:47:43 +0100 (MET)
Subject: Python Rocks!
In-Reply-To: <865lsm$s3p@news.or.intel.com>
Message-ID: <XFMail.000120104743.mikael@isy.liu.se>

On 20-Jan-00 tye4 wrote:
 >  I suggest you go to other newsgroups and let them know how brilliant Python
 >  is because it has an invisible block terminator.

But they aren't invisible, I see the delimiters all the time.

This discussion is getting nowhere. If you want delimiters, try LISP.
I've been polite up 'til now. I suggest you (tye4) wake up, grow up, 
get a life, et cetera. 

I suggest the rest of us ignore this ... ahem ... individual from now on.
At least I will.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    10:35:03

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From hei at adtranzsig.de  Thu Jan 27 05:26:26 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 27 Jan 2000 11:26:26 +0100
Subject: SourceForge question
References: <388E7D5B.7840983F@erols.com> <388F140B.C0AA1C99@erols.com> <388f88ca.1925884@news.erols.com> <948966042.615175@news.adtranzsig.de>
Message-ID: <948972527.962905@news.adtranzsig.de>

Hi! I'm a clueless Windows user and i want to start a
tiny project in SourceForge.
I have a CVS binary and the files for my first release.
I need: Advice on where to get a SSH binary for Windows
and how to set up the initial CVS tree. It doesn't need
to be perfect, this will not become a big team effort,
i suppose. Somebody who already has set up a
SourceForge project out there? Tanks for any
advice.

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





From tseaver at aftershock.neosoft.com  Wed Jan  5 11:13:24 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 5 Jan 2000 10:13:24 -0600
Subject: Progress at internationalising the TP robot
References: <vr3dsd2rgw.fsf@trex.IRO.UMontreal.CA>
Message-ID: <E0104151A5BBF1F1.38F742594E1F82B8.BA7E244F23C6C05C@lp.airnews.net>

In article <vr3dsd2rgw.fsf at trex.IRO.UMontreal.CA>,
=?ISO-8859-1?Q?Fran=E7ois_Pinard?=  <pinard at IRO.UMontreal.CA> wrote:
>Hello, my friends.  Please accept my best wishes!
>
>I made some progress at internationalising the Python robot, which is
>used to process submissions in the Translation Project.  Let me share a
>quick status report with you, by repeating part of a message I sent to
>the national team leaders, earlier today.  Comments, if any, are welcome.
>

Oh, darn -- I thought you were going to say you'd found a way to put an
Armani suit on the Timbot, in preparation for his unveiling at IPC8.

Clothes-maketh-the-bot'ly,

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From michael at camelot-it.com  Sun Jan 16 08:19:09 2000
From: michael at camelot-it.com (Michael Tiomkin)
Date: Sun, 16 Jan 2000 15:19:09 +0200
Subject: __name__
References: <387F4F42.B9E1DB8E@kw.igs.net>
Message-ID: <3881C54D.C1BD656C@camelot-it.com>

  All the class info is a part of the class, and not of an object.  For an object a, you can use
a.__class__.name instead.

  Michael

JB wrote:

> Could anyone tell me how to get a classes name:
>
> class MyClass:
>    def className( self ):
>        return self.__name__
>
> This doesn't fly. The docs say a class has the __name__ attribute but...
>
> thanks
> jb
>
> --
> There are bugs and then there are bugs.  And then there are bugs.
> -- Karl Lehenbauer



From aa8vb at yahoo.com  Fri Jan  7 09:33:30 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 7 Jan 2000 09:33:30 -0500
Subject: XML to dictionary?
In-Reply-To: <20000107092832.A25967@vislab.epa.gov>
References: <8518kq$59n$1@nnrp1.deja.com> <m3puvf5p4h.fsf@havenrock.com> <20000107092832.A25967@vislab.epa.gov>
Message-ID: <20000107093330.B25967@vislab.epa.gov>

Randall Hopper:
 |Matt Gushee:
 | |from pyxie import File2PYX    # PYX is the notation pyxie uses for
 | |import string		      # parsed XML
 | |
 | |f = File2PYX('hithere.xml')
 | |data = f.read()
 | |f.close()
 | |data = string.split(data)
 | |
 | |dict = {}
 | |for line in data:
 | |    if line[0] = '(':         # '(' at the beginning of the line in 
 | |			      # PYX indicates an element
 |
 |Pyxie is a good tool.  But just to pick nits, using string.split isn't safe
 |with Pyxie output.  :-)

In case it wasn't obvious to some, the fix is to change split to:

    data = string.split(data, '\n')

-- 
Randall Hopper
aa8vb at yahoo.com



From aa8vb at yahoo.com  Thu Jan  6 07:31:58 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 6 Jan 2000 07:31:58 -0500
Subject: Python and GUI options
In-Reply-To: <XFMail.000105210217.Vadim.zeitlin@dptmaths.ens-cachan.fr>
References: <20000105141519.A13656@vislab.epa.gov> <XFMail.000105210217.Vadim.zeitlin@dptmaths.ens-cachan.fr>
Message-ID: <20000106073158.A12345@vislab.epa.gov>

Vadim Zeitlin:
 |Randall Hopper:
 |
 |> From a user standpoint, it would be useful if the build system were changed
 |> to a GNU configure sense-and-testwhat-you-have scheme rather than a
 |> configure-this-file, put-overrides-in-this-file, and see-what-breaks
 |> scheme.
 |
 | Sorry, I don't understand what are you referring to. wxWindows uses
 |configure exclusively
...
 |> hope they do), more docs on creating the build.local file are a must.
 |
 | Sorry again, but I don't even know what is it?

Oh, I see the confusion.  I'm guessing you are a wxWindows person, and
these refer to problems with wxPython, the wxWindows interface layer for
Python.

 |> Also, wxGTK and wxPython sense different
 |> compilers (SGI vs. GCC).  I had to coerce them to use the same ones.  
 |
 | Hmm... How did this happen? Did the value of your CXX env var changed
 |in between somehow?

No, I hardly ever specify CC/CXX.  wxPython doesn't use configure though.

Thanks,

-- 
Randall Hopper
aa8vb at yahoo.com



From tim_one at email.msn.com  Tue Jan 11 22:31:23 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 11 Jan 2000 22:31:23 -0500
Subject: [ANN] Multimethod.py -- multimethods for Python
In-Reply-To: <87zouclb8m.fsf@baal.winnegan.de>
Message-ID: <000301bf5cad$801c5020$da2d153f@tim>

[Siggy Brentrup]
> ...
> From what I see above it [multimethods] might as well be
> some dynamic variant of C++'s template functions.

That's one way to look at it, although a strained one <wink>.  Multimethods
are the general case; someone looking at C++ thru MM eyes would view all of
the latter's varieties of static, virtual, overloaded, and templatized
(sure, hypothetically dynamically templatized too) dispatching gimmicks as
special cases of MM.

It's a natural & elegant generalization of both generic functions and OO
dispatch.  Whether it's a generalization that's needed is open to debate.

i-still-don't-have-cable-tv<wink>-ly y'rs  - tim





From arno at mtb.ee  Wed Jan 19 10:59:57 2000
From: arno at mtb.ee (arno at mtb.ee)
Date: Wed, 19 Jan 2000 17:59:57 +0200
Subject: Python and MS SQL (Microsoft SQL server)
References: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in> <3884C60B.BE820782@psychosis.com>
Message-ID: <slrn88bnrt.vr.arno@arnokodu.mtb.internal>

On Tue, 18 Jan 2000 14:59:07 -0500, Dave Cinege <dcinege at psychosis.com> wrote:
>"George Thom at s" wrote:
>> 
>> Hi Dave,
>> The odbc module that comes as part of the win32 extensions bundled with
>> Python 1.5.2 for the various flavours (!) of Windows should do just fine.
>
>So you're saying python has to be runnign on NT for this to work?
>
>> Additional functionality (with some tweaking ... ymmv ;-) can be obtained
>> with marc Lemburg's mxODBC module.
>
>What about this? we're looking to run on solaris or linux.
>

You could use the FreeTDS library:

http://www.freetds.org/

It supports both Sybase and Microsoft servers and exports 
db-, ct- and ODBC- compliant API's. I have used it with great success
in combination with Python's sybasemodule.



HTH,

Arno


From sameerc1 at yahoo.com  Thu Jan 13 23:20:31 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Thu, 13 Jan 2000 20:20:31 -0800
Subject: os.popen under Win32 Using Apache
Message-ID: <013101bf5e46$b2d83a20$d8c9fea9@desktop>

I came across some reference indicating that popen does not work in all
conditions under win32.  I believe Mark Hammond includes popen2 in the
win32all package that is supposed to replace popen under Windows.  I would
check Python's windows page for more information.

-----Original Message-----
From: Nicolas L?onard <leonar_n at epita.fr>
Newsgroups: comp.lang.python
To: python-list at python.org <python-list at python.org>
Date: Thursday, January 13, 2000 4:33 AM
Subject: os.popen under Win32 Using Apache


>Hit there
>
>
>I have a problem when I try to use os.popen() in a cgi with Apache,
>under Win32.
>
>It works in command-line mode, but don't work through Apache.
>
>Any idea ?
>
>Thanks.
>
>--
>Nicolas L?onard
>T?l : (33) 06-61-40-30-31
>E-Mail : leonar_n at epita.fr
>
>
>

__________________________________________
NetZero - Defenders of the Free World
Get your FREE Internet Access and Email at
http://www.netzero.net/download/index.html



From PartyG2816 at aol.com  Mon Jan  3 11:12:15 2000
From: PartyG2816 at aol.com (PartyG2816 at aol.com)
Date: Mon, 3 Jan 2000 11:12:15 EST
Subject: Considering using python for something, and need a bit of info :)
Message-ID: <0.3903b60a.25a2245f@aol.com>

Allo list,
    I'm involved in writing a MUD server from scratch with a small group of 
other people.  For you who do no know, a MUD is a multi-user text game.  I've 
been looking around for a scripting language... I was considering making my 
own that was custom tailored to my exact needs, but in the process of 
thinking about that i got repeated mentions of look into Python. :) So here's 
the main reason for this post: I wanted to know how suitable Python would be 
to embed into an environment like I mentioned earlier to be used by mainly 
non-programmers to perform scripting goals.
    A script would be used for (in most cases) to "liven" up the place.  
Let's say this person is a builder who is making an "area". An area would be 
part of the textual world that contains the things the player can interact 
with.. IE a set of the rooms the players character would be in, the thing 
sthey could interact with (a chair, a sword, whatever), and the NPC's (non 
player characters).  They may need to make a script to perform a special case 
thing.. Like say when player a enters room b, under condition c, i need to do 
d.
    Could the Python interpreter be embedded and used in a way that would 
allow me to use it like this within my application?  I've downloaded the 
interpreter and gone through parts of the tutorial... I love it. *chuckles* 
I'm a C++ programmer at heart, but I can definately see some advantages to 
using Python for osme things.. I think it would be fairly easy to teach a 
non-programmer how to use it within the context of my application would be 
fairly easy with a person dedicated to helping the nephytes and a few docs.
    Another concern of mine is: how secure can it be? IE can i restrict 
certain aspects of it based upon a condition? Say I have a a normal builder, 
ie someone who simply makes areas and may perform a few scripts for them, I 
wouldn't want to let him do certain things... Whereas me as head programmer, 
I would have access to many more options within my scripts :) Or take a 
normal "player".  I'm going to be allowing small scripts and macros and such 
for my players to automate a few 'boring' things, I would want to give them 
access to a limited set of use, and not let them say re-arrange everything in 
the room. *chuckles*
    Ah well, maybe you peoples understand what I'm asking and can help, I'd 
appreciate any response :) If you have questions about something I may need 
to clarify, then by all means ask and I'll try my hardest. ;)

JD

p.s. - If it turns out that yall think Python could be suitable for osmething 
like this, could you include a few linkst o places that discuss and give some 
working examples of embedding python in another application and running 
scripts through it using whatever API is inherent in Python for that use? :)



From Martin.Preishuber at stuco.uni-klu.ac.at  Sat Jan  8 06:13:57 2000
From: Martin.Preishuber at stuco.uni-klu.ac.at (Martin Preishuber)
Date: Sat, 08 Jan 2000 12:13:57 +0100
Subject: ftplib exceptions
Message-ID: <38771BF5.D3CECC6C@stuco.uni-klu.ac.at>

Hi,

I've a question concerning exception handling in the ftplib module.
My program contains some code like:

try:
    ftp.cwd(directory)
    ftp.dir(self.LsCallBack)
except all_errors, detail:
    HandleError(detail)

The strange thing is, that this way messages like

200 PORT command successful
200 Type set to A

are handled by the exception, even this messages are no
errors from my point of view ...

any ideas ? thanks,

Martin

-- 
Martin Preishuber - Student, ECLiPt Core Member, SysAdmin
http://eclipt.uni-klu.ac.at,
mailto:Martin.Preishuber at stuco.uni-klu.ac.at

Wernher von Braun settled for a V-2 when he coulda had a V-8.


From jblazi at netsurf.de  Fri Jan 21 14:20:24 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Fri, 21 Jan 2000 20:20:24 +0100
Subject: function "overloading"
References: <3888787d_2@goliath.newsfeeds.com> <20000121173150.A1498@stopcontact.palga.uucp> <3888a12e_3@goliath.newsfeeds.com>
Message-ID: <3888afa9_4@goliath.newsfeeds.com>

Now I see that isinstance + the builtin function types + the standard module
types solve my problems.

J.B.


Janos Blazi <jblazi at netsurf.de> schrieb in im Newsbeitrag:
3888a12e_3 at goliath.newsfeeds.com...
> This was not exctly what I meant.
> These classes are not subclasses of each other. What I mean is: I have
> essentially the same function from the mathematical point of you
> (intersection) and I have to call it with different arguments. Of course I
> could code the type information into the classes, so during runtime
> intersection could find out whith which types of arguments it has been
> called, but maybe there are more elegant possibilities.
>
> Janos Blazi
>
> Gerrit Holl <gerrit.holl at pobox.com> schrieb in im Newsbeitrag:
> 20000121173150.A1498 at stopcontact.palga.uucp...
> > Janos Blazi wrote on 948468281:
> > > I have 3 classes called Tline, Tsegment and Tpoint, repectively. I'd
> like to
> > > write a function "subset" that cheks if one of the instances of the
> above
> > > classes  is a subset of another instance (as long as this is
sensible).
> > >
> > > I am afraid I cannot overload the name "subset" and additionally when
> subset
> > > has control it cannot decide the type of its arguments: Does this mean
I
> > > have to write several *different* subset functions having different
> names?
> >
> > >>> print issubclass.__doc__
> > issubclass(C, B) -> Boolean
> >
> > Return whether class C is a subclass (i.e., a derived class) of class B.
> >
> > regards,
> > Gerrit.
> >
> > --
> > Please correct any bad English you encounter in my email message!
> > -----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
> > Version: 3.12
> > GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V
> PS+ PE?
> > Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
> > -----END GEEK CODE BLOCK-----
> >
>
>
>
>
>   -----------== Posted via Newsfeeds.Com, Uncensored Usenet News
==----------
>    http://www.newsfeeds.com       The Largest Usenet Servers in the World!
> ------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers
==-----




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From kc5tja at garnet.armored.net  Sun Jan  9 20:36:46 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 01:36:46 GMT
Subject: Python Host
References: <bJ3e4.8$78.764@news.uswest.net>
Message-ID: <slrn87ie9t.sb7.kc5tja@garnet.armored.net>

In article <bJ3e4.8$78.764 at news.uswest.net>, Andrew Ferk wrote:
>Are there any web hosts that support Python?

Oops.  I forgot to post our web address.  http://www.armored.net/

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From mal at lemburg.com  Wed Jan 12 13:18:12 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 12 Jan 2000 19:18:12 +0100
Subject: ODBC and mxODBC
References: <387B9540.BE617781@exceptionalminds.com>
Message-ID: <387CC564.D0EC75D8@lemburg.com>

Timothy Grant wrote:
> 
> Hi,
> 
> Has anyone got any tips on configuring mxODBC, or ODBC in general on a
> Linux box?
> 
> I have a Linux box running PostgreSQL with the ODBC system installed.
> 
> I can access it just fine from a Windows machine running mxODBC, but I
> have yet to be able to figure out how to get it to work from my own
> development machine which is a Linux notebook running RH6.1
> 
> I can access the database just fine using the PyGreSQL module, but I
> really want to use mxODBC so that I can do cross platform development.
> 
> This is probably more of a ODBC question than a Python question, but any
> help would be appreciated.

ODBC on Linux is not hard, but neither very easy to get up
and running.

The best solution is usually to add an ODBC manager layer
between the application and the driver. There are
two managers for Linux: iODBC and unixODBC (there should
be links to both of them from the mxODBC page).

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





From echuck at mindspring.com  Thu Jan  6 07:06:54 2000
From: echuck at mindspring.com (Chuck Esterbrook)
Date: Thu, 06 Jan 2000 07:06:54 -0500
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net> <005501bf582f$b2860ce0$c02b2bc1@martelli>
Message-ID: <3874855E.DA81DC0B@mindspring.com>

Alex Martelli wrote:

> Chuck Esterbrook writes:
>
> > Does anyone know who would host a CGI Python site?
>
> I've noticed the same thing, and posed the same question, a
> month ago, or so.
>
> The only answer I got on this group was to join the PSA and
> get a Starship account -- I did the former, but I'm waiting
> for the latter to happen when the Starship admin gets some
> time.
>
> I also sent requests to all providers of free web hosting that
> currently house Perl CGI only.

[snip]


So here's what I've collected from various people:


Interland.com supports Python on a Unix server (but not on Windows/NT).


http://www.webaxxs.net/
They have Python 1.4 installed but I installed 1.5 into my account.
Since they give me 500meg of storage that's not a problem.


Armored Internet (http://www.armored.net) supports Zope and Python CGIs.
Zope is supported under our Unix web hosting packages, and Python CGIs are
supported under both Unix and Windows NT web hosting packages.


(belated) favourable response was www.prohosting.com;




From robin at jessikat.demon.co.uk  Sat Jan  1 01:40:42 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 1 Jan 2000 06:40:42 +0000
Subject: fun with stock quotes
References: <Fnn296.A3w@world.std.com> <ncREqFA3nZb4Ew1G@jessikat.demon.co.uk>
Message-ID: <UO8EUHAqFab4EwDS@jessikat.demon.co.uk>

In article <ncREqFA3nZb4Ew1G at jessikat.demon.co.uk>, Robin Becker
<robin at jessikat.demon.co.uk> writes
>In article <Fnn296.A3w at world.std.com>, Will Ware <wware at world.std.com> writes
>>I recently discovered that Yahoo's stock-quote-serving web
>>page is now offering historical stock data in machine-friendly
>>form. The following scripts will pull in a bunch of stock data
>>and format it as a C module. Then you can back-test any silly
>>investment algorithms you come up with. Nerdy greedy fun for
>>the whole family.
>>
whoops some bugs in the last thing. here goes version 2; keeps all dates
for correlation analyses etc etc

#######################################################
import urllib, string, normalDate, math, regex, sys, os

class _YahooData:
    def __init__(self,ticker,date,Open,High,Low,Close,Volume):
        self.ticker=ticker
        self.start=date[0]
        self.finish=date[-1]
        self.date=date
        self.Open=Open
        self.High=High
        self.Low=Low
        self.Close=Close
        self.Volume=Volume

def fetchRangeAsCSV(ticker, first, last):
    '''go to Yahoo and fetch a range of historical stock
    data as a CSV file, where 'first' and 'last' are normalDates'''
    url = 'http://chart.yahoo.com/table.csv?s=%s' \
          '&a=%d&b=%d&c=%d&d=%d&e=%d&f=%d&g=d&q=q&y=0&z=%s' \
          '&x=.csv' % (ticker,
                       first.month(), first.day(), first.year(),
                       last.month(), last.day(), last.year(),
                       ticker)
    return urllib.URLopener().open(url).readlines()

def _yahoo_to_normalDate(date):
    date = string.split(date, '-')
    date[0] = eval(date[0])
    date[2] = eval(date[2])
    if date[2] > 90: year = date[2] + 1900
    else: year = date[2] + 2000
    month = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep',
            'Oct','Nov','Dec'].index(date[1]) + 1
    return normalDate.ND(year * 10000 + month * 100 + date[0])

def fetchRange(ticker, first, last):
    date = []
    Open = []
    Low = []
    High = []
    Close = []
    Volume = []
    for x in fetchRangeAsCSV(ticker,first,last)[1:]:
        F = string.split(x, ',')
        date.append(_yahoo_to_normalDate(F[0]))
        Open.append(eval(F[1]))
        High.append(eval(F[2]))
        Low.append(eval(F[3]))
        Close.append(eval(F[4]))
        Volume.append(eval(F[5]))

    # Yahoo gives reverse time order
    date.reverse()
    Open.reverse()
    High.reverse()
    Low.reverse()
    Close.reverse()
    Volume.reverse()
    if len(date) == 0:
        raise 'ouch'
    return _YahooData(ticker,date,Open,High,Low,Close,Volume)

if __name__=='__main__':
    # just do a quick test for now
    masterList = [
        # tech stocks
        'LNUX', 'MSFT', 'AMZN', 'INTC', 'RHAT', 'DELL', 'GTW',
        'CSCO', 'AAPL', 'AOL', 'COMS', 'CPQ', 'LCOS', 'LU',
        'ORCL', 'SUNW', 'TXN', 'YHOO',
        # single-letters and blue chips
        'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
        'J', 'K', 'L', 'N', 'O', 'P', 'R',
        'U', 'W', 'X', 'Y', 'Z',
        # Dow Jones stocks
        'AA', 'AXP', 'T', 'BA', 'CAT', 'CHV', 'KO',
        'DIS', 'DD', 'EK', 'GE', 'GM', 'GT', 'HWP',
        'IBM', 'IP', 'JNJ', 'MCD', 'MRK', 'MMM', 'JPM',
        'MO', 'PG', 'S', 'TRV', 'UK', 'WMT',
        # biotech
        'AFFX', 'BIOM', 'CRA', 'ENMD', 'HGSI',
        'MATX', 'MLNM', 'RGEN', 'SAFS', 'DNA', 'GENE',
        'INCY', 'GLGC', 'PFE',
        'AVXT', 'CTIC', 'CRXA', 'GZMO', 'IDPH',
        'ILXO', 'ISIP', 'RZYM', 'SUPG', 'TGEN', 'VICL',
        # retailers, general interest
        'BBY', 'COST', 'UPS', 'HD',
        ]
    masterList = masterList[:12]
    start = normalDate.ND(19950101)
    finish = normalDate.ND()   # today
    d = {}

    for t in masterList:
        d[t]=fetchRange(t, start, finish)
        print    "data for %s obtained start=%d(%f) finish=%d(%f)" % \
                (t,d[t].start,d[t].Close[0],d[t].finish,d[t].Close[-1])
-- 
Robin Becker


From skip at mojam.com  Thu Jan 27 13:00:06 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 27 Jan 2000 12:00:06 -0600 (CST)
Subject: A couple of (minor) things I'd like to see in the Python _environment_
In-Reply-To: <yu14sbza12z.fsf@black-racer.atg.com>
References: <yu14sbza12z.fsf@black-racer.atg.com>
Message-ID: <14480.34726.8155.378590@beluga.mojam.com>

    chris> * A command line driven documentation browsing system, similar to 
    chris>  'perldoc' in the Perl world.  (Python's entities have doc
    chris>  strings - maybe those could be leveraged?)

A weak substitute perhaps, but in my .pythonrc file I define the following
help function:

    def help(x):
	"""display an object's __doc__ attribute if it has one"""
	if hasattr(x, "__doc__") and x.__doc__ is not None:
	    print x.__doc__
	else:
	    print "no docstring available"

    chris> * An interactive debugger invokable with a command line option to
    chris>   the Python interpreter. - I know about PDB, and it does in fact
    chris>   work, but oh how I miss the ability to simply do:

I'm mostly a print kind of guy, but I seem to recall that IDLE has some
debugging builtin.  Others have worked on debugging both Python and the
underlying C code.  Someone released some gdb magic awhile ago.  You might
search deja.com for it.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098
"Languages that change by catering to the tastes of non-users tend
not to do so well." - Doug Landauer



From aa8vb at yahoo.com  Mon Jan 24 12:02:49 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 24 Jan 2000 12:02:49 -0500
Subject: Python - SIGPIPE handled bug
In-Reply-To: <x7ya9h5c9o.fsf@lindm.dm>
References: <20000121162925.A189033@vislab.epa.gov> <x7ya9h5c9o.fsf@lindm.dm>
Message-ID: <20000124120249.B358575@vislab.epa.gov>

Dieter Maurer:
 |Hallo Randall

Hi.  Thanks for the reply.

 |>   ------------------------------------------------------------------------
 |>     > python -c "for i in xrange(1000): print 'Some output'" | head -2
 |>     Some output
 |>     Some output
 |>     Traceback (innermost last):
 |>       File "<string>", line 1, in ?
 |>     IOError: [Errno 32] Broken pipe
 |>   ------------------------------------------------------------------------
 |> 
 |> python getting a SIGPIPE is appropriate behavior.  "Whether" it gets a
 |> SIGPIPE is up to the shell, in UNIX:
 |
 |It seems, that in your case Python does not get a SIGPIPE but
 |the "write" returns "-1" and "errno==EPIPE".
 |
 |According to POSIX, this is the behaviour, if SIGPIPE is not handled
 |by the default action:
 |       EPIPE  fd is connected to a pipe or socket  whose  reading
 |              end  is closed.  When this happens the writing pro??
 |              cess will receive a SIGPIPE signal; if it  catches,
 |              blocks or ignores this the error EPIPE is returned.
 |
 |The SIGPIPE handling under Python 1.5.2/Linux 2/Bash seems to be
 |to ignore the signal (possibly inherited from the bash!).

I'm getting this behavior with tcsh 6.04.00 on SGI IRIX, which delivers
SIGPIPE.

Looking at the code, the problem is that Python resets this signal so that
it is ignored (see pythonrun.c), deferring the handling until the failed
read() gets EPIPE as you said.  This is why an otherwise quiet-exit
scenario (delivery of SIGPIPE) is being transformed into a noisy error
condition in Python.

What's at issue is whether every Python app meant to be piped into another
program has to munge exceptions or signals to avoid emitting bogus errors.
Take for example:

    python -c "for i in xrange(100000): print 'Lots of output'" | more

If we quit our pager before scrolling to the bottom of the python-generated
stream, is that an error?  Well no, of course not.  So as with default
delivery of SIGPIPE (which calls _exit(2), Python IMO shouldn't generate
errors for this situation.

-- 
Randall Hopper
aa8vb at yahoo.com



From bruce_dodson.nospam at bigfoot.com  Fri Jan 14 19:20:05 2000
From: bruce_dodson.nospam at bigfoot.com (Bruce Dodson)
Date: Fri, 14 Jan 2000 20:20:05 -0400
Subject: Python and ArcView GIS
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>   <85hs33$4t9$2@newsserver.rrzn.uni-hannover.de>   <XXcf4.16787$2x3.303971@newscontent-01.sprint.ca> <85k96e$hn1$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <j5Pf4.20537$2x3.367615@newscontent-01.sprint.ca>

I think the confusion is on ArcView 3 vs. ArcView 4.  There is no such thing
as ArcView 4 yet, but the hype is that it will be based on a subset of
ArcInfo 8, and therefore will be totally COM based like ArcInfo.  ArcView 3
does not use COM in any way, shape or form.  It has a rich class heirarchy,
but those classes are not exposed through COM.  Nor can ArcView even act as
a COM/ActiveX client.  My theory is this is because everyone at ESRI who
knows anything about COM for has been working on ArcInfo 8 for the past 2-3
years.

Really, my extension is about controlling/executing Python from ArcView.
However, it's fair to say that the extension does allow for tighter
integration in the other direction as well.  It provides a function,
arcview.avexec, which allows you to run arbitrary Avenue scripts from
Python.  Since Avenue is the only published way to access the ArcView class
heirarchy, this is the tightest integration that I am able to achieve if I
want to publish the source code.

For Python scripts that are run from anywhere else than my extension, you
have to use use DDE to control ArcView (ugh!).

Hope this helps clear things up.

Bruce


Bernhard Reiter wrote in message
<85k96e$hn1$1 at newsserver.rrzn.uni-hannover.de>...
>There is still confusion in my head. :)
>I am not exactly sure if ArcView exposes COM objects right now.
>If is does, python could control ArcView and then you can also control
>other COM objects with python.
>
>If you embed python in ArcView, then python might get more control about
>ArcView. Of course when inside python you can also control other COM
objects.
>
>So the question is:
> Does your extension enable python to have more control about ArcView
> as it would have controling it over ArcView's COM objects from outsite?
>
> Do you need the python extension to expose ArcView capabilities
> to other programs?




From robin at jessikat.demon.co.uk  Tue Jan 18 14:27:04 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 18 Jan 2000 19:27:04 +0000
Subject: WIN32 : how to list the current processes by their executable na me?
References: <326536345498D311B3BC00105A39802A07457E@newsintern.dspace.de>
 <CxKdRDAwDEh4Ewzj@jessikat.demon.co.uk> <862936$g33$1@nnrp1.deja.com>
Message-ID: <r$Ob4HAI6Lh4Ewm8@jessikat.demon.co.uk>

In article <862936$g33$1 at nnrp1.deja.com>, MHammond at skippinet.com.au
writes
>In article <CxKdRDAwDEh4Ewzj at jessikat.demon.co.uk>,
>  Robin Becker <robin at jessikat.demon.co.uk> wrote:
>
>> this fails for me with win95.
>
>Unfortunately, win32pdh only works on NT/2000.  I dont know how to do
>the same thing on Win95/98...
>
>Mark.
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
Well there's the famous demos of PS and Kill which are claimed to work;
certainly they use different methods depending on the system. Kill
doesn't always quite do it on Win95.
-- 
Robin Becker


From a.eyre at optichrome.com  Tue Jan 25 06:00:03 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 25 Jan 2000 11:00:03 -0000
Subject: Importing a dll with a different name
In-Reply-To: <86ig7d$2pf$1@nnrp1.deja.com>
Message-ID: <000101bf6723$54d5d780$3acbd9c2@peridot.optichrome.com>

>   I have managed to figure out that it is because pythoncom is defined
> in the windows registry to point to a specific dll
> (c:\windows\system\pythoncom15.dll). I am also having the same problem
> with pywintypes (and pywintypes15.dll).

I always thought this was a bit of an oddity with Pythonwin. Better
to rename the DLLs to the name that python expects, and stick them
somewhere on the PYTHONPATH. I'm assuming the installer proglet can
handle this.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From grant at nowhere.  Thu Jan 13 11:45:52 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 13 Jan 2000 16:45:52 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com>
Message-ID: <slrn87s0a0.83o.grant@grante.comtrol.com>

In article <387DF8C4.4B3214CE at netscape.com>, Michel Orengo wrote:
>I've just started to play around with MAPI. I'm far from being a expert but here
>is a sample of what you can do to send a message:
>
>import win32com.client
>session = win32com.client.Dispatch("MAPI.Session")
>session.Logon("Michel Orengo")
>outbox=session.Outbox
>collmsg=outbox.Messages
>msgo=collmsg.Add("my subject","my
>text",win32com.mapi.mapitags.PR_MESSAGE_CLASS,1)
>msgo.Recipients.Add("Michel","morengo at morganlabs.com")
>msgo.Sender.Address="morganea at bellatlantic.net"
>msgo.Send(1,0)
>
>If I've time, I'll write a more useful example. For now, I'm still playing...
>Hope this will help

If anybody needs a little project to work on, I could really
use a program that sucks messages from a MAPI server and
forwards them to an SMTP server. ;)

-- 
Grant Edwards                   grante             Yow!  My forehead feels
                                  at               like a PACKAGE of moist
                               visi.com            CRANBERRIES in a remote
                                                   FRENCH OUTPOST!!


From ullrich at math.okstate.edu  Mon Jan 31 11:28:25 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Mon, 31 Jan 2000 10:28:25 -0600
Subject: Dynamic class construction?
References: <000301bf6bd4$b9210150$3acbd9c2@peridot.optichrome.com>
Message-ID: <3895B829.39067306@math.okstate.edu>


Adrian Eyre wrote:

> [...]
>
> There's already a module which does this...
>
> import new
>
> def anOp(self):
>         return 'This is an instance of the class %s' % \
>                 self.__class__.__name__
>
> someOps={'__str__':anOp}
> G=new.classobj('G', (Group,), someOps) # Group defined elsewhere

    I sorta thought there might be a module that does this, but I
couldn't find it.

    Um, I still can't find it. It appears to be a builtin module -
new.__doc__
says

'     Functions to create new objects used by the interpreter.

You need to know a great deal about the interpreter to use this!',

which is helpful, I guess, but I can't find any more about it in
the docs (or I'm not looking in the right place? It's not in the
"module index", I don't find any interesting files contaning
the text "classobj", etc.)

    This means I got the docs-for-weenies and it's documented
in the manly version, or what?


> --------------------------------------------
> Adrian Eyre <a.eyre at optichrome.com>
> http://www.optichrome.com



From fdrake at acm.org  Tue Jan 18 13:11:24 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Tue, 18 Jan 2000 13:11:24 -0500 (EST)
Subject: conflict between my package and os standard module
In-Reply-To: <Dkcg4.19933$zU5.216301@wbnws01.ne.mediaone.net>
References: <Dkcg4.19933$zU5.216301@wbnws01.ne.mediaone.net>
Message-ID: <14468.44236.110905.563714@weyr.cnri.reston.va.us>

Ben Wittner writes:
 > In my_proj/os.py I put the following line:
 > 
 > import os
 > 
 > Since my_proj is NOT on my PYTHONPATH, I thought that the line above
 > would cause the standard module os to be imported.

  The reasons it works the way it does have already been explained, so 
I won't repeat the explanation.
  I haven't seen anyone suggest doing the import of the "real" os
module like this, in my_proj/os.py:

os = __import__("os")

This isn't the prettiest, but it works and is about as clean as it
gets short of a "virtual" package that provides access to the real
things.  The only weirdnesses with __import__() are with dotted module 
names; see the chapter 2 of the Library Reference for more
information.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From robin at jessikat.demon.co.uk  Wed Jan 19 08:55:13 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 19 Jan 2000 13:55:13 +0000
Subject: Tkinter hangs again under Windows
References: <20000119064523.A15259@foxboro.com>
 <3885BACA.C08F575B@endea.demon.nl>
Message-ID: <Z4snZDABJch4Ewxx@jessikat.demon.co.uk>

In article <3885BACA.C08F575B at endea.demon.nl>, Niels Diepeveen
<niels at endea.demon.nl> writes
...
>window, not to the controls. Also, the following tcl code shows the same
>problem in Wish 8.0:
>  button .b -text "Click here" -command {
>    tk_messageBox -message "Button pressed"
>  }
>  pack .b
>  tk_messageBox -title "Welcome" -message "Hello, bug!"
>
>> 
...
I tried this with 8.2.3 and it worked fine. I work python with tk 8.2.x
and so this problem probably doesn't happen there.
-- 
Robin Becker


From guido at CNRI.Reston.VA.US  Tue Jan 18 23:57:59 2000
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Tue, 18 Jan 2000 23:57:59 -0500
Subject: More problems....
In-Reply-To: Your message of "Tue, 18 Jan 2000 23:28:32 EST."
             <000201bf6235$a53d25e0$9a2d153f@tim> 
References: <000201bf6235$a53d25e0$9a2d153f@tim> 
Message-ID: <200001190457.XAA18536@eric.cnri.reston.va.us>

> [Guido, is this a known bug?  Not in the bugs list, but it's
>  become a frequent complaint.]
> 
> [TechSpood]
> > ...
> > I will write a simple program (in the IDLE enviroment) then
> > when I run it, it will run it til it gets to the first input
> > statement, at which time it will stop reading the program.
> > I don't get a syntax error or any other kind of error.

This was indeed a bug in IDLE 0.4 as released.  In that version, if
you run a script from inside IDLE (by pressing f%), all output goes to
a separate *Output* window.  When you use input(), the prompt text is
therefore displayed in that window, but entering text there does
nothing.  If you enter text in the *Python Shell* window (where an
innocent Python prompt is displayed) and hit return, that text
actually goes to the program and the program can continue.

It's probably time to make an official release of IDLE 0.5.  Maybe
after the conference...

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



From jamesp at simplypets.com  Tue Jan  4 19:32:32 2000
From: jamesp at simplypets.com (James Punteney)
Date: Tue, 04 Jan 2000 16:32:32 -0800
Subject: is variable a list?
Message-ID: <38729120.7616054D@simplypets.com>

Hello,

Is there some way to determine whether a variable is list or a string?
Basically I have a variable that I would like to be able to be a string
or a list, but I need some way to tell which it is.

Thanks for the help,
--James Punteney

jamesp at simplypets.com



From rgruet at ina.fr  Fri Jan  7 06:19:43 2000
From: rgruet at ina.fr (Richard Gruet)
Date: Fri, 07 Jan 2000 12:19:43 +0100
Subject: urllib.urlopen returns 'connection refused' on Solaris
Message-ID: <3875CBCF.72E26CE7@ina.fr>

Hi everybody,

On Solaris 2.6 (Sun OS 5.6), the fct urllib.urlopen() returns an error
'Connection refused' for all URLs:

>>> urlopen('http://www.python.org')
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/urllib.py", line 59, in urlopen
    return _urlopener.open(url)
  File "/usr/local/lib/python1.5/urllib.py", line 157, in open
    return getattr(self, name)(url)
  File "/usr/local/lib/python1.5/urllib.py", line 253, in open_http
    h = httplib.HTTP(host)
  File "/usr/local/lib/python1.5/httplib.py", line 51, in __init__
    if host: self.connect(host, port)
  File "/usr/local/lib/python1.5/httplib.py", line 79, in connect
    self.sock.connect(host, port)
IOError: [Errno socket error] (146, 'Connection refused')

Could it be my build of Python 1.5.2 on Solaris ?? This fct works well
on other platforms.
The equivalent fct in Java works fine, too (so this is not a matter of
host config).

Any ideas ?

Richard



From apederse at siving.hia.no  Wed Jan 19 05:41:20 2000
From: apederse at siving.hia.no (Asle Pedersen)
Date: Wed, 19 Jan 2000 11:41:20 +0100
Subject: SQL escape sequence problems
Message-ID: <8644bj$lrd$1@snipp.uninett.no>

I'm trying to run the following code:

father="something"
sql="SELECT FROM TABLE WHERE MESSAGEID='%s'" % (father)
MySQL.escape(sql)

but there is a problem since ' is really an escape character. How can I get
around this problem? I'm using MySQLmodule-1.2. running under win32
enviroment. I have tried to get a compiled version of the MySQLdb module
whithout any success.

regards,
Asle





From tim_one at email.msn.com  Tue Jan 11 23:23:29 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 11 Jan 2000 23:23:29 -0500
Subject: a small problem
In-Reply-To: <20000112015923.A8450@xs4all.nl>
Message-ID: <000701bf5cb4$c72320a0$da2d153f@tim>

[knotwell at my-deja.com]
>> Is the following a bug?
>>
>> >>> q = [(1,'dd'),(2,'ddd'),(3,'ddddd')]
>>
>> >>> reduce(lambda x,y: x[0] + y[0],q)
>>
>> Traceback (innermost last):
>>   File "<stdin>", line 1, in ?
>>   File "<stdin>", line 1, in <lambda>
>> AttributeError: __getitem__

[Thomas Wouters, explains it, then ...]
> ...
> >>> reduce(lambda x,y: (x[0] + y[0],),q)
> (6,)
>
> or perhaps
>
> >>> reduce(lambda x,y: (x[0] + y[0],),q)[0]
> 6
>
> depending on what you want. But I think this is roughly the
> point where people like Guido(*) start wondering if reduce
> is a good idea at all. I'd say the above reduce is almost a
> good argument for perl ;) Usually, what you're trying to do
> so forcefully with just reduce and lambda is much more
> elegantly (and readably!) done with map and reduce (and
> lambda, in this case):
>
> >>> import operator
> >>> tmplist = map(lambda x: x[0], q)
> >>> tmplist
> [1, 2, 3]
> >>> reduce(operator.add, tmplist)
> 6

Whereas people like Guido <wink> would not have considered writing anything
other than

>>> sum = 0
>>> for n, s in q:
        sum = sum + n
>>>


> (This may seem like nitpicking for the above example, but it
> makes for readable code later on, when you write longer and
> longer and more complicated map/filter/reduce sequences.

People like Guido would agree, but would say it applies with three times the
force to the dirt-simple and dirt-obvious code just above.

Seriously, I don't think I've ever seen a good use for reduce in Python --
although I've certainly used it <wink>.

> ...
> (*) I'm assuming here. I dont actually know so, because
> i'm not foolish enough to consider myself knowledgable in
> all things Guido, but these are the hints i get from other
> postings on the subject, by other people.

My favorite line from Guido on this is available *somewhere* in DejaNews,
where I responded to someone going over the edge with these things that they
were intended to be minor conveniences, and Guido jumped in to amend that to
"minor nuisances".

In the Nov '98 issue of Linux Journal, there's an interview with Guido.  In
response to "What feature of Python are you least pleased with?",

    Guido:  Sometimes I've been too quick in accepting
    contributions, and later realized that it was a mistake.
    One example would be some of the functional programming
    features.  lambda is a keyword that lets you create a
    small anonymous function; built-in functions such as map,
    filter and reduce run a function over a sequence type,
    such as a list.

    In practice, it didn't turn out that well.  Python has
    only two scopes:  local and global.  This makes writing
    lambda functions painful, because you often want to
    access variables in the scope where the lambda was
    defined, but you can't because of the two scopes.  There's
    a way around this, but it's something of a kludge.  Often
    it seems much easier in Python just to use a for loop
    instead of messing around with lambda functions.  map
    and friends work well only when a built-in function that
    does what you want already exists.

I suppose finally posting that will rekindle a few old flamewars!  Brrrr.

There's another reason to look up this interview:  there's a great staged
photo of Andrew Kuchling and GvR "discussing the future of Python".  You can
tell it's staged because Andrew is paying attention, and Guido is holding a
coffee cup (with what appears to be a pencil floating in it) instead of his
usual chainsaw <wink>.

can't-believe-he'd-drink-coffee-with-a-straw-ly y'rs  - tim





From khimj at gto.net.om  Fri Jan 21 21:02:13 2000
From: khimj at gto.net.om (J Dubal)
Date: Sat, 22 Jan 2000 06:02:13 +0400
Subject: Python compiler?
Message-ID: <86duuq$8a1@OM9.omantel.net.om>

Hello good people!

I am new to python and I understand that although it is a free-to-use
language, it can be used to build commercial, propritary applications.
Is this correct?
Does it have a compiler for Linux and/or windows that can produce native
executables?
If not, do we have to distribute python application sources?
Are there modules or libs available to bulid data entry/query applications
for character terminals?
Thanks in advance.
Best wishes.
Dubal.




From gmcm at hypernet.com  Tue Jan  4 21:58:09 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 4 Jan 2000 21:58:09 -0500
Subject: what if i dont want a 32 bit int?
In-Reply-To: <38752609@MailAndNews.com>
Message-ID: <1265082304-1433932@hypernet.com>

jerry smith asks:

> what if all i want is really just an 8 bit or 16 bit number? so
> if i do this
> 
> i=0x80
> i=~i
> print '%x' %i
> ffffff7f
> 
> that is too many leading f's so i have to have an anding mask of 
> i & ff? is there any alternative? 

Nope.

> to select how much space i want
> to dedicate for a python object to be stored as? 

You think a Python int only takes 32 bits? Sorry to disappoint, 
but they're objects(*). Now if you want to store them externally 
as something else, you can use the struct module to get them 
into or from C values.

(*) The first 100 ints are singleton objects:

>>> x = 1
>>> import sys
>>> sys.getrefcount(x)
40


- Gordon



From joec at brain.acme.org  Fri Jan 21 08:40:04 2000
From: joec at brain.acme.org (Joe Cotellese)
Date: Fri, 21 Jan 2000 13:40:04 GMT
Subject: Talking to embedded python
References: <m3snzsdq7l.fsf@brain.acme.org>
Message-ID: <m3puuv1qpi.fsf@brain.acme.org>

Joe Cotellese <joec at brain.acme.org> writes:


> My question is how can I communicate with the python interpreter built
> into my application? The idea being that once the program is running I
> can send commands in the form of python scripts to the running
> program.
> 
I've thought about COM but if we want to port our application to other
platforms then COM would't help me (plus it's way to much overhead for
the things we want to do).




From tismer at tismer.com  Wed Jan 26 03:29:26 2000
From: tismer at tismer.com (Christian Tismer)
Date: Wed, 26 Jan 2000 03:29:26 -0500
Subject: Stackless Python 1.0 + Continuations 0.6
References: <mt2.0-27365-948450827@news.informatik.uni-bonn.de> <86f8uu$3ds$1@kopp.stud.ntnu.no> <388BDEFA.8D57A66F@tismer.com> <388C91E1.500B272A@hassler.de>
Message-ID: <388EB066.3DACCDEE@tismer.com>


Gerald Klix wrote:
> 
> Christian Tismer wrote:
> 
> > Magnus Lie Hetland wrote:
...
> > > would it be possible to stop and pickle a thread? This might be very
> > > interesting for Mobile Agent Technology
...
> Chris, please add this point to your arguments for stackless python!

I did so. Beneath the PalmIII-opportunity, it seems to be
a major argument. Thanks a lot!

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From w6uv at hotmail.com  Wed Jan 26 15:41:38 2000
From: w6uv at hotmail.com (Jerry Gardner)
Date: Wed, 26 Jan 2000 20:41:38 GMT
Subject: Multidimensional Dictionaries: Thanks!
References: <slrn88ppsg.49t.w6uv@voyager.home.com>
Message-ID: <slrn88ue3f.ejp.w6uv@voyager.home.com>

On 25 Jan 2000 00:00:16 GMT, Jerry Gardner wrote:
>
>Are multidimensional dictionaries possible in Python? I'm sure they're
>possible, but I can't figure out the syntax.

Thanks to all of you who took the time to answer my question,
especially Richard Brodie, who's answer was exactly what I was looking
for.

-- 
Jerry Gardner     | "Bill Clinton has all the steely resolve
w6uv at hotmail.com  | of a kamikaze pilot on his 37th mission."


From usenet-mm-yyyy at julesallen.com  Fri Jan  7 21:05:49 2000
From: usenet-mm-yyyy at julesallen.com (Jules Allen)
Date: Sat, 08 Jan 2000 02:05:49 GMT
Subject: Python in Corel Linux
References: <947294639.1002448618@news.islandnet.com>
Message-ID: <38769B7A.CC3A6AC3@julesallen.com>

Jim, Corel Linux is basically Debian (http://www.Debian.org/) give or take about
seven packages. You can get the Python package from


ftp.debian.org/debian/dists/Debian2.1r4/main/binary-i386/interpreters/python-base_1.5.1-7.deb

or


ftp.debian.org/debian/dists/unstable/main/binary-i386/interpreters/python-base_1.5.2-6.deb

if you want a more up to date Python.

If you use the Corel update tool, chances are it's on Corel's site somewhere.

Good luck!

Cheers,
Jules

--
Put the current month and year in my address to reply via email.

J Dukarm wrote:

> I just installed Corel Linux ("standard" version on CD) on a PC where I
> can also boot Win98, and as a Windows refugee I was very pleased to see how
> easy it was to access my Windows file system, read floppies, etc.  Only problem
> is -- it seems that Corel left Python out of the distribution!
>
> Has anybody out there installed Python in Corel Linux, and if so are there any
> tricks to getting it set up and working?
>
> Thanks,
>
> Jim Dukarm
> DELTA-X RESEARCH



From loewis at informatik.hu-berlin.de  Sat Jan 15 15:29:01 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 15 Jan 2000 21:29:01 +0100
Subject: International Python
References: <oqk8ldg2jp.fsf@titan.progiciels-bpi.ca>
Message-ID: <p6qwvpbxenm.fsf@informatik.hu-berlin.de>

Fran?ois Pinard <pinard at iro.umontreal.ca> writes:

> One option to `xpot' allows for automatic extraction of all doc strings,
> which never need explicit marking for this reason.  I think I got it right:
> doc strings should be correctly found even after very complex initialisation
> of keyword parameters.  I do not expect this to be common use, but this
> might be useful in highly interactive programs giving access to Python
> internal code, a bit like Emacs already does for its own LISP.

Are there plans to have the interpreter put the translations into the
__doc__ strings? If so, how would that work?

Regards,
Martin



From thierry at acm.org  Thu Jan 27 17:31:41 2000
From: thierry at acm.org (Thierry Thelliez)
Date: Thu, 27 Jan 2000 15:31:41 -0700
Subject: Charts
Message-ID: <3890C74D.786411BC@acm.org>

Is there any packages for creating charts (histograms, scattered plots,
line charts,...) in Python ?



Thanks
Thierry



From sanderson at ttm.com  Tue Jan  4 10:19:12 2000
From: sanderson at ttm.com (Scott Anderson)
Date: Tue, 04 Jan 2000 10:19:12 -0500
Subject: CPAN for Python?
Message-ID: <38720F70.A5DC6F33@ttm.com>

Hi all,

Is there an equivalent of Perl's CPAN for Python?

Regards,

-scott



From grant at nowhere.  Mon Jan  3 11:20:30 2000
From: grant at nowhere. (Grant Edwards)
Date: Mon, 03 Jan 2000 16:20:30 GMT
Subject: RPM-interface/module
References: <386B3CDB.FF843280@bibsyst.no> <386BCE53.A8D93A67@es.co.nz> <3daemsrru3.fsf@amarok.cnri.reston.va.us> <iacg0wkrp3c.fsf@vier.idi.ntnu.no>
Message-ID: <slrn871j2e.jg1.grant@grante.comtrol.com>

In article <iacg0wkrp3c.fsf at vier.idi.ntnu.no>, Magnus L. Hetland wrote:
>akuchlin at mems-exchange.org (Andrew M. Kuchling) writes:
>> > Thomas Weholt wrote:
>> >
>> > > Is there a RedHat package interface/module available for Python??
>> 
>> Red Hat actually wrote a Python rpm module for use in their
>> config. scripts; it's still there in both RH 6.0 and 6.1.  (In fact,
>> it looks like some functions were added in 6.1...)
>
>Hm. I was under the impression that rpm was written in Python... I
>take it it is not? 

IIRC, rpm was originally written in Python, and was later
re-written in C to improve performance.

-- 
Grant Edwards                   grante             Yow!  If Robert Di Niro
                                  at               assassinates Walter Slezak,
                               visi.com            will Jodie Foster marry
                                                   Bonzo??


From ivanlan at callware.com  Mon Jan 24 15:35:49 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Mon, 24 Jan 2000 13:35:49 -0700
Subject: OFFTOPIC: *-ly y'rs,
References: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com> <3dvh4jz0m9.fsf@amarok.cnri.reston.va.us>
Message-ID: <388CB7A5.BD34B1A4@callware.com>

Hi All--

"Andrew M. Kuchling" wrote:
> 
> Agent Drek <drek at MonsterByMistake.Com> writes:
> > Where did they come from and are they patented :)
> > How can one join the signoff-ly club?
> 
> The first occurrence, at least in the python-list archives, was quite
> a while ago:
> 
>   in-any-case-the-best-christmas-present-i-got-today!-ly y'rs - tim
>         -- Tim Peters, 29 Dec 91
> 
> I imagine Tim probably used this on other mailing lists before
> python-list.
> 

One thing to be careful of, as Tim himself may point out, is the
placement of the apostrophe.  Many people mistakenly position it as if
for a possessive:  yr's.  This is, however, incorrect, as the
construction functions as a contraction for "yours", which is not a
possessive.  Thus, the correct format is: y'rs, Agent Drek.

<and-where-is-agent-dross?>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From mjackson at wc.eso.mc.xerox.com  Sun Jan  9 21:10:10 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 10 Jan 2000 02:10:10 GMT
Subject: control structures (was "Re: Sins")
References: <1264659011-22815940@hypernet.com>
Message-ID: <85bf22$r32$1@news.wrc.xerox.com>

"Gordon McMillan" <gmcm at hypernet.com> writes:

> > > Gordon McMillan wrote:

> > > >  for i in X:
> > > >    if X[i]==e:
> > > >      # do the "found" thing
> > > >      break
> > > >  else:
> > > >    print "Not Found"

> As one who regularly uses the searching algorithm above, I 
> find it highly useful. I find "if seq:" to be useful for testing for an 
> empty sequence. I'm not going to defend the spelling of the 
> construct, or it's description in the Language Ref, but it *is* 
> very useful.

The use of "else" as the keyword here is something I find jarring.  To
me, at least, it suggests the suite should be executed if the for loop
*doesn't* terminate normally (i.e. when there's a break) rather than
the opposite.  Same thing with while. . .else.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
    Faced with the choice between changing one's mind and proving
    there is no need to do so, almost everybody gets busy on the proof.
				- John Kenneth Galbraith




From roy at popmail.med.nyu.edu  Tue Jan 25 15:20:28 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Tue, 25 Jan 2000 15:20:28 -0500
Subject: re module -- case insensitive matching.
Message-ID: <roy-3281DF.15202825012000@netnews.nyu.edu>

Is there any difference between:

re.compile ('[a-zA-Z]*')

and

re.compile ('[a-z]*', re.I)

other than notation, i.e. are the state machines that get built 
identical in either case?


From m.faassen at vet.uu.nl  Fri Jan 14 18:29:29 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 14 Jan 2000 23:29:29 GMT
Subject: ODBC
References: <387A227E.EF84B01E@yahoo.com>
Message-ID: <85obgp$r1u$1@newshost.accu.uu.nl>

Tekir wrote:
>   At work I'm finishing up work on a equipment checkout database.  I've been
> using tab delimited files (from MS Access) for all of my work.  But I'm getting
> tired of all the work currently required.  So I'm going to make one master ODBC
> and do all my work from therer.  Ok now for my questions what files do this?  Do
> they come with Python or do I need to d/l them and if so from where?  And are
> there any good tutorials out there?

The Python-Win package comes with an ODBC interface. I've used that with
success with MS Access. There's also a seperate package called mxODBC.

This URL contains useful information:

http://www.python.org/windows/OdbcHints.html

For more about talking to databases from Python look here:

http://www.python.org/topics/database/

The PythonWin version of ODBC is at least DB-API 1 compliant, I think the
mxODBC one is DB-API 2 compliant.

Good luck!

Regards,

Martijn
-- 
History of the 20th Century: WW1, WW2, WW3?
No, WWW -- Could we be going in the right direction?
> there any good tutorials out there?

> Thanks,
> Jeremy

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


From wlfraed at ix.netcom.com  Wed Jan 26 23:29:44 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Wed, 26 Jan 2000 20:29:44 -0800
Subject: Newbie question
References: <R5Oj4.892$yY6.13414@news.goodnet.com>
Message-ID: <PsePOAN0wPRT=SF9fYxYjAROk10l@4ax.com>

On Thu, 27 Jan 2000 02:31:13 GMT, Scott Johnson  <sjohnson at csnet.net>
declaimed the following in comp.lang.python:

> My background is in MVS Cobol, CICS & Visual Basic.
> I have played a little with java & Perl.
> I have an understanding of classes & their properties & methods.
> Could someone explain to me the concept of .self ?

	Hmmm, Of those mentioned I though Java was supposed to be OO...
Doesn't it have a means of referring to the invoking instance from
inside a method?

	"self" is just a name, C++ commonly uses "this", and I'm sure
some language likes "me".

	However, that argument position in a class method is filled with
the actual instance that has been invoked. (Apologies if my syntax is a
tad off here, I don't do enough to recall it without a cheat book)

Class AClass:
	def dosomething(self, data):
	#some thing or other
		self.what = data


	a = AClass()
	b = AClass()

	a.dosomething(x)
	b.dosomething(x)

	The calls to dosomething effectively turn into (if this were a
language like Ada):

		dosomething(a,x)
		dosomething(b,x)

	Without "self", the methods would be sharing data too.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From parkw at better.net  Tue Jan 25 18:59:32 2000
From: parkw at better.net (William Park)
Date: Tue, 25 Jan 2000 18:59:32 -0500
Subject: vi and python?
In-Reply-To: <86l7iq024mg@news2.newsguy.com>
References: <85uc1m$5i8$1@pegasus.tiscalinet.it> <86l7iq024mg@news2.newsguy.com>
Message-ID: <20000125185932.A19619@better.net>

On Tue, Jan 25, 2000 at 04:22:14PM -0600, Ben Beuchler wrote:
> There seems to be a general preference for emacs in the python
> community.  I am personally much more comfortable working with vim.
> Does anyone have any particular tips for python development within
> vim? Particular shortcuts to test code, autoindentation settings,
> etc...  I would particularly like to be able to use some of the idle
> features with, perhaps, gvim...
> 
> Ben

This is what I have in ~/.vimrc relevant to *.py.  You can move the
global settings to 'au BufRead *.py set ...' if you like.

version 5.4
set nocompatible	" Use Vim defaults (much better!)
set backspace=2		" backspace over everything
set autoindent
set shiftwidth=4	" 4 space/indent
set smarttab		" 'shiftwidth' at beginning of line; <Tab> elsewhere
set textwidth=72	" 72 char/line
" Python indentations
au BufRead *.py set smartindent
au BufRead *.py set cinwords=if,elif,else,for,while,try,except,finally,def,class

--William



From fredrik at pythonware.com  Fri Jan 21 10:15:22 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 16:15:22 +0100
Subject: os._exit(), os.kill(os.getpid(), 9)
References: <20000121145429.A4320@stopcontact.palga.uucp> <slrn88gsj5.77g.hildeb@stahlw06.stahl.bau.tu-bs.de> <dt3888743di4ba0n79%milek@korweta.task.gda.pl>
Message-ID: <011c01bf6422$5a243050$f29b12c2@secret.pythonware.com>

Robert Milkowski <milek at task.gda.pl> wrote:
> how to exit with clean up?

    sys.exit()

or

    sys.exit(errcode)

or

    raise SystemExit

</F>




From sei at sunse.jinr.ru  Thu Jan 20 09:57:00 2000
From: sei at sunse.jinr.ru (Evguenii Smogailov)
Date: Thu, 20 Jan 2000 17:57:00 +0300 (MSK)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <slrn88e4cs.ie6.mgm@unpkhswm04.bscc.bls.com>
Message-ID: <Pine.SOL.3.91.1000120175408.21484B-100000@sunse>

On 20 Jan 2000, Mitchell Morris wrote:
> Except that Perl considers both '==' and 'eq' to be operators, and they
> do different things.

Yes, i know, but i'm not sure yet that it's a good idea for 'eq' to be 
different from '==' in Python

evguenii




From didier.rano at wanadoo.fr  Fri Jan 14 23:46:59 2000
From: didier.rano at wanadoo.fr (didier.rano)
Date: Sat, 15 Jan 2000 05:46:59 +0100
Subject: Built-in exception class not found: EnvironmentError
Message-ID: <85pe5p$n4$1@wanadoo.fr>

Hello,

i have try to embedded python in a tool. On Dec-Alpha, i have not problem,
but on HP-UX, i have this message when i execute my tool:

Built-in exception class not found: EnvironmentError.  Library mismatch?
Warning!  Falling back to string-based exceptions

This message is sent by Py_Initialize.

Someone knows this problem ?

I'm searching to generate a libpython.sl, because only the libpython1.5.a is
generated
on HP-UX. Someone knows it ?

I have a problem with TKinter to be generated a dynamic library on HP-UX.
Someone knows it ?

Thank you very much for your answers

Didier Rano





From tim_one at email.msn.com  Mon Jan 10 23:14:12 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 10 Jan 2000 23:14:12 -0500
Subject: Python plugin
In-Reply-To: <kizIvIAD2me4EwPp@jessikat.demon.co.uk>
Message-ID: <000e01bf5bea$512f6a40$d2a0143f@tim>

[Samuel A. Falvo & Robin Becker bump heads over the True
 Pukeworthiness or Lack Thereof of COM; then, in good
 Python fashion, make up before blood is spilt]

Admirably civilized, fellows!  Tip o' the hat.

[Robin]
> OK I guess we all need a bit of adrenalin now and then heh heh
> heh aaaaaarrrggghhhhh the Timbot is presumably busy.

Don't count on the timbot, Robin!  I've noticed with increasing concern an
increasing (albeit non-monotonic) tendency for it to act as bland as toast
this year.  If this keeps up, I'm going to have to slash its electricity.

it's-easy-to-be-nice-when-you're-not-starving-ly y'rs  - tim





From pinard at iro.umontreal.ca  Wed Jan 19 14:25:30 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 19 Jan 2000 14:25:30 -0500
Subject: Realtime scheduling??
In-Reply-To: Paul Winkler's message of "Wed, 19 Jan 2000 04:14:12 -0500"
References: <38858064.CDA4866C@yahoo.com>
Message-ID: <oqbt6hyic5.fsf@titan.progiciels-bpi.ca>

Paul Winkler <slinkp23 at yahoo.com> writes:

> I'm working on a module I eventually hope to use to generate musical
> events in realtime.  (These events will be sent to MIDI devices or
> to software synthesizers.)  I've found that on my Linux system (pretty
> much stock RedHat 5.2), scheduling actions with sched.scheduler is not as
> reliable as I'd like.  Events are anywhere from 2 to 20 milliseconds late.
> I'd like to improve the reliability, which brings me to my next question:

I did something similar, had similar findings, and did similar research.
However, being quite reluctant at forcing my things to be suid'ed to root,
I tried something else, which turned out to be very workable.  It seems that
even if the human ear is extremely picky about sound frequency, it is very
forgiving in rythmical details, as long as the overall rhythm is preserved.

So, I just set the scheduler so it runs after real-time, instead of using
the scheduler to insert theoretical delays.  That is, you maintain a logical
real time in your program, and when about to ask for a delay, you delay
the logical real time plus the wanted delay minus the actual real time.
The effect is that you automatically catchup on lags, and very surprisingly,
it works well.  The musical behaviour is often acceptable even under heavy
machine load.  Of course, it is perfect (or not far from) with light load.

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





From mkc at kc.net  Thu Jan  6 22:12:28 2000
From: mkc at kc.net (Mike Coleman)
Date: 06 Jan 2000 21:12:28 -0600
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <3dzouke2v7.fsf@amarok.cnri.reston.va.us>
Message-ID: <87vh56bmn7.fsf@kc.net>

akuchlin at mems-exchange.org (Andrew M. Kuchling) writes:
> The Essential Reference is just what its title says: a reference
> guide.  You'd have to be *really* motivated to learn Python from it.

Some people (e.g., yours truly) *like* learning a language from an elegant,
terse reference.  I have the original O'Reilly book, and although it has some
merit, I hate its by-the-pound style ("Here we are writing a bubble sort in
Python.  Okay, now let's see what an insertion sort would look like.  Here's
another one, only this time in descending order."  Ugh.).

Anyway, Beazley's reference is really good, IMO.  The only other language
reference I like as much is Steele's Common Lisp the Language.

--Mike

-- 
Any sufficiently adverse technology is indistinguishable from Microsoft.


From skip at mojam.com  Mon Jan  3 13:15:34 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 3 Jan 2000 12:15:34 -0600 (CST)
Subject: Splitting up classes between files
In-Reply-To: <roy-E337E8.09101731121999@netnews.nyu.edu>
References: <roy-E337E8.09101731121999@netnews.nyu.edu>
Message-ID: <14448.59206.519338.337830@beluga.mojam.com>

    Roy> I've got a top-level class, a, and a bunch of sub-classes, a1, a2,
    Roy> a3, etc.  To make it easier to manage the code, I'd like to put
    Roy> each in a separate file, but still have the user be able to import
    Roy> the whole bunch with a single import, i.e.  not have to do "import
    Roy> a, a1, a2, a3, etc".

    Roy> Can I just put "import a1, a2, a3, etc" in the bottom of my "a.py"
    Roy> file?  It seems like it should work fine, but are there any hidden
    Roy> gotchas?

Roy,

You might want to look at package support, but without it, I think you can
accomplish what you want.  At the end of a.py, simply add

    from a1 import a1
    from a2 import a2
    from a3 import a3

When your users import module a, they'll get access to

    a.a (class a in python file a.py)
    a.a1 (class a1 from python file a1.py)
    ...

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 dyeung at ust.hk  Wed Jan 26 10:09:42 2000
From: dyeung at ust.hk (David Yeung)
Date: Wed, 26 Jan 2000 23:09:42 +0800
Subject: To find a point inside a polygon
Message-ID: <388F0E36.6CE37135@ust.hk>

Does anyone know if any python implementation for finding whether a point
is inside a polygon? E.g., by giving a point(x,y) and an array of polygon's
points (x0,y0,x1,y1,x2,y2,...), it returns true or false to tell if the 
given (x,y) is inside the polygon.

Thanks

david


From ZakonS at eccelerate.com  Thu Jan  6 09:48:56 2000
From: ZakonS at eccelerate.com (Zakon, Stuart)
Date: Thu, 6 Jan 2000 09:48:56 -0500
Subject: bdb: which class?
Message-ID: <B9AA1D5DA95ED211900100805F6FFF9C02F3D8F6@exbhbhmb01.us.dbisna.com>

Hi,
So far I am using the following within a subclass of Bdb:

	def where_am_i (self, frame):
		name = frame.f_code.co_name
		filename = frame.f_code.co_filename
		lineno = frame.f_code.co_firstlineno
		self.here_i_am = '%s, %s, %d' % (name, filename, lineno)

I would like to have the class name for methods of classes which are being
called but I didn't see it in the documentation for code objects. 
Anybody know if this can be found?

Thanks,
Stuart Zakon





From Alex.Martelli at think3.com  Wed Jan 12 08:49:01 2000
From: Alex.Martelli at think3.com (Alex Martelli)
Date: Wed, 12 Jan 2000 14:49:01 +0100
Subject: RunPy (was py2bat (was: How do I make a Python .bat executabl
	e file?))
Message-ID: <6D8A17398E28D3119F860090274DD7DB4B3DC2@pces.cadlab.it>

Thanks to Piet Ostrum for this information...:


> AM> Haven't seen the program you mention, either, but I've had a go
> AM> at writing a version for Python, just in case some should find it
> AM> useful.  http://aleax.supereva.it/Python/runpy.zip (2477 bytes)
> 
> It's not found there.
> 
EEk, you're right -- supereva doesn't serve it, although I have
placed it there and it does serve other (htm) files from that
directory just fine.  Crazy free server.

I've put it in:
	http://www.magenta.com/~alex/runpy.zip
and a link to it on the
	http://aleax.supereva.it/Python/index.htm
until I clear things up with the supereva support folks (will
probably take another 1000 years or so).


> AM> a version that is a window app rather than a console one (and
> AM> which runs pythonw rather than python), options to look for
> AM> the script elsewhere than on the PATH (e.g., in the same
> AM> directory as the renamed runpy.exe itself, or on PYTHONPATH),
> 
> In the same directory at least I would think. And pythonw should be
> simple:
> First check if there is a .pyw file and if so, run pythonw.
> 
"Same directory" would be easier than "along the PATH".  Doing
both would be a bit harder of course, and, in what order should
the two options be tried (matters in case of duplicates)?

What's a .pyw file?  I thought pythonw.exe was the exact
equivalent of python.exe, except for not using a console.
Is there any doc about this anywhere...?


Thanks for the feedback & TIA for any more of it,


Alex




From donn at oz.net  Thu Jan 20 00:54:22 2000
From: donn at oz.net (Donn Cave)
Date: 20 Jan 2000 05:54:22 GMT
Subject: SUGGEST: os.getuser()
References: <Jylh4.3912$Ef6.1036319@news.shore.net> <20000116174004.A5995@stopcontact.palga.uucp> <000b01bf6049$36532140$f29b12c2@secret.pythonware.com> <20000116195710.A7700@stopcontact.palga.uucp>
Message-ID: <8667ue$o8o$0@216.39.151.169>

[... re whether os.getuser() should look at $LOGNAME or /etc/passwd first ...]

I'm with Michael on this one, if LOGNAME has been manipulated by the
user then it's just as well if the program goes along with it.
Things like filesystem permissions should protect everyone from any
resulting mischief.  If not, it would be something like a setuid 
situation, where Python probably just doesn't belong altogether.

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


From aahz at netcom.com  Mon Jan  3 10:43:50 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Jan 2000 15:43:50 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com> <84olef$5j1$1@news1.xs4all.nl> <84omaa$qln$1@nntp9.atl.mindspring.net> <84pg9v$rda$1@news1.xs4all.nl>
Message-ID: <84qg3m$u0v$1@nntp1.atl.mindspring.net>

In article <84pg9v$rda$1 at news1.xs4all.nl>,
Boudewijn Rempt <boud at rempt.xs4all.nl> wrote:
>Aahz Maruch <aahz at netcom.com> wrote:
>>
>> Mind expanding on this?  Are you asserting it's also a Bad Idea to slurp
>> multi-megabyte BLObs to a client?
>
>Well, what databases are good at, is picking one thing out of many.
>If you use a database as a flat file, and then do the picking of one
>out of many you've essentially written a distributed database. I've
>never used blobs myself - I know they can be useful, even necessary,
>but until now I've just stored an url in the database, and put the blobs
>behind a http server. That way, the database doesn't have to work so
>hard, and it becomes easy to distribute the load over two machines.

Oh, okay.  Let's suppose, for example, that you have a customer
database, and the client wants to view the entire order history for a
customer.  Would you say that it's inappropriate for the database to
return thousands of rows for an active customer?

Similarly, suppose you have millions of BLObs that change regularly.
How would you organize them on an HTTP server?  Incidentally, HTTP is
*not* an efficient transmission mechanism (less true if you're using
HTTP 1.1).
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From tim_one at email.msn.com  Thu Jan  6 20:19:07 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 6 Jan 2000 20:19:07 -0500
Subject: lazy (?) evaluation, request for advice
In-Reply-To: <CC59868A0A76D311B1F50090277B321C1F300E@VRTMAIL>
Message-ID: <000701bf58ad$31ad7120$932d153f@tim>

[Mike Fletcher]
> Minor note: Tim is actually one of the most helpful people
> on the group.

Like Alex didn't already know that <wink>.

> He has a quirky sense of humour sometimes.

To the contrary -- it's all of you who are quirky!  I live on a diet of pure
umbrage here.

> Instead of reading "sneer", read "playful banter among assumed
> friends and colleagues".

Close enough to "brazen ridicule" for me.

> Tim tends to play the "crotchety old man" role quite a lot
> for its entertainment value. It's not a slight on you, it's
> just part of his persona in this group.

Indeed, over on rec.sport.pro-wrestling I'm known as the Gentle Shepherd.
Guiding our youth is harder than it used to be.

> Enjoy yourself,

excellent-advice-from-a-crotchety-young-man-ly y'rs  - tim





From mwh21 at cam.ac.uk  Mon Jan 17 09:19:14 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 17 Jan 2000 14:19:14 +0000
Subject: Version history of python
References: <85v0ih$409$1@nnrp1.deja.com>
Message-ID: <m3wvp8g4rh.fsf@atrus.jesus.cam.ac.uk>

Hamish Lawson <hamish_lawson at yahoo.co.uk> writes:

> I'm looking for a version history of Python that gives details of when
> particular features were introduced into the language and when
> particular libraries were included in the standard library. I've tried
> looking on the python.org website and searching the web, but I've only
> found an essay by Guido tracing Python's influences, but not detailing
> when these features made it into the language.

Misc/HISTORY in the source distribution has more than 200k of what you
ask for; Misc/NEWS has the changes since 1.5.1.

HTH,
Michael


From ullrich at math.okstate.edu  Sun Jan 30 14:20:46 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Sun, 30 Jan 2000 13:20:46 -0600
Subject: Dynamic class construction?
References: <389309F7.4A35371F@math.okstate.edu> <8707c0$ua2$1@nnrp1.deja.com>
Message-ID: <38948F0E.36AA2548@math.okstate.edu>


Nic Williams wrote:

> In article <389309F7.4A35371F at math.okstate.edu>,
>   "David C. Ullrich" <ullrich at math.okstate.edu> wrote:
>
> Just for reference, this is Part i.
>
> >     Say I want to make a function that returns a new _class_.
> > I can do this:
> >
> > def NewClass(whatever):
> >     class new(baseClass):
> >         pass
> >     new.SomeAttribute = whatever.whatever
> >     return new
> >
> > and it seems to do the job. Is there another way to
> > do this that seems much better for some reason?
>
> and the next question is part ii.
> I'll answer part ii first (helps explain part i soln)
>
> >     I guess a related question would be whether
> > having distinct classes with the same __name__
> > is going to be trouble - does __name__ get used
> > in significant ways internally or is it just there
> > for the programmer's convenience?
>
> A class's name when the class is first defined is used in two ways:
> i) internal information (so you can ask a class object who it is)
> ii) binding information within the namespace
>
> A good way to demonstrate this is with the following example:
> >>> class FirstName:
> ...     pass
> ...
> >>> FirstName
> <class __main__.FirstName at e6fe0>
> >>> FirstName.__name__ = 'SecondName'
> >>> FirstName
> <class __main__.SecondName at e6fe0>
> >>> SecondName
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> NameError: SecondName
>
> Note here that when we change the class's internal name, it now thinks
> that it's name is SecondName, but the namespace's binding still calls it
> FirstName.
>
> See py doc: http://www.python.org/doc/current/ref/execframes.html
>
> So you can multiple classes who think internally they are the same name,
> but they must be named differently if they lie in the same namespace.

    Of course they need to have different names, I was wondering if
there would be an internal problem with identical __name__'s. I
gather not, thanks.
    (Here for example I'll have

G1 = NewGroup(something)
G2 = NewGroup(somethingelse)

and then G1.__name__ and G2.__name__ would both be
"new".)

>
> Ok, back to part i.
>
> With your example above and below we see that the binding name within
> the global namespace is determined by the attribute G. Note that there
> is also no class called 'new' anywhere at the end of the method (it was
> declared within the local namespace which disappeared at the end of the
> method).
>
> So we have an object whose internal name is 'new' but its binding is
> 'G'. If you want to change the class's internal name to that of its
> external binding, use something like:

    I don't know that I want to - I was wondering if I needed to.
Couldn't hurt.

>
> def newClass(name, ops, superclass):
>         class NewClass(superclass):
>                 pass
>         for op in ops.keys():
>                 setattr(NewClass, op, ops[op])
>         NewClass.__name__ = name
>         return NewClass
>
> def anOp(self):
>         return 'This is an instance of the class %s' % \
>                 self.__class__.__name__
>
> someOps={'__str__':anOp}
> G=newClass('G', someOps, Group) # Group defined elsewhere
> print G
> <class __main__.G at f92a8>
>
> BINGO. It's name binding and internal naming now match.
>
> Hope that answered the question.

    Yup, thanks.

<sigh>
    I tried simply setting the value of __name__ to what
I wanted, and I got an error about read-only attributes.
At least I _thought_ I'd tried that - works fine today.
It may be time-of-day-dependent.
</sigh>

    I take it that simply putting a "pass" in the inner
class definition and then setting attributes _is_ the
"right" thing? It wasn't the first thing I thought of,
but it was the first thing that worked.

Thanks,

DU



From wjk at wjk.mv.com  Tue Jan  4 18:48:09 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Tue, 04 Jan 2000 18:48:09 -0500
Subject: String Handling
References: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com>
Message-ID: <387286B8.F5482A9A@wjk.mv.com>

As an alternative from a newbie:

import re
replacement = re.compile(" ")
str1 = "Lucy by the carriage"
str2 = replacement.sub("+",str1)

.... should result in str2 as 'Lucy+by+the+carriage'

though I would opt for import string and string.replace()


Simon Faulkner wrote:

> Hi All,
>
> Can anyone give me the Python line/command to change all spaces in a
> string to +s?
>
> ie
>
> "Lucy by Railway Carraige"
>
> becomes
>
> "Lucy+by+Railway+Carraige"
>
> Muchos gratcias etc
>
> Simon
> Simon Faulkner



From niels at endea.demon.nl  Wed Jan 19 08:23:22 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Wed, 19 Jan 2000 14:23:22 +0100
Subject: Tkinter hangs again under Windows
References: <20000119064523.A15259@foxboro.com>
Message-ID: <3885BACA.C08F575B@endea.demon.nl>


Hans-Joachim Widmaier schreef:
> 
> Niels Diepeveen wrote:
> 
> > This seems to be the problem. If you show any dialogs before entering
> > mainloop, you won't receive events anymore.
> 
> Apparently this is the case. Is this a (known?) bug or just an undocumented
> feature, or is it even documented? The problem seems to only exist under
> Windows.

I'd say it's probably a bug in the Windows port of tk. WinSight shows
that after the dialog all mouse messages are sent to the containing
window, not to the controls. Also, the following tcl code shows the same
problem in Wish 8.0:
  button .b -text "Click here" -command {
    tk_messageBox -message "Button pressed"
  }
  pack .b
  tk_messageBox -title "Welcome" -message "Hello, bug!"

> 
> Thanks a lot. I'll find a way around it, probably a self.after() that
> shows the dialogue shortly after entering the mainloop.

I posted a different workaround yesterday in response to a similar
question. 

-- 
Niels Diepeveen
Endea automatisering



From jjlucsy at concentric.net  Sat Jan 15 22:04:59 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Sat, 15 Jan 2000 22:04:59 -0500
Subject: How to use Pythonwin's ability to access windows dll's???
References: <1264133143-1099206@hypernet.com>
Message-ID: <vzag4.2959$iy5.176986@typ12.deja.bcandid.com>

I have to disagree. I wrote a program that interfaces with another using an
existing dll library that worked wonderfully. Of course I ran into some
problems with different versions of calldll.pyd,npstruct.py, and
npstruct.pyd.

"Gordon McMillan" <gmcm at hypernet.com> wrote in message
news:1264133143-1099206 at hypernet.com...
> Sam Schulenburg writes:
> > In article <1264144125-438654 at hypernet.com>,
> >   gmcm at hypernet.com wrote:
> > > Sam Schulenburg writes:
> > >
> > > > I can access windows dll's using Sam Rushing's calldll.pyd and
> > windll.py. My
> > > > question is are thes features built into Pythonwin?
> > >
> > > Nope. Everything in the Win32 extensions is wrapped (much
> > > of it via SWIG).  Sam's stuff is regarded as dangerous,
> > > because it's too easy to crash Python.
>
> > Seriously what is the purpose of the LoadLibrary function if not to
> > access Dll's>
>
> You can load & use resources from a dll (for dialogs, views
> etc). Not much use, otherwise.
>
>
> - Gordon
>




From CJACKSON2 at prodigy.net  Tue Jan  4 23:37:00 2000
From: CJACKSON2 at prodigy.net (Charles)
Date: Tue, 4 Jan 2000 22:37:00 -0600
Subject: Quick Python v.s. Learning Python
References: <387152EB.10DD0E17@Lextek.com> <Fnu1K6.1nz.0.bloor@torfree.net>
Message-ID: <84uhgf$93lm$1@newssvr03-int.news.prodigy.com>

I have half-a-dozen Python books (I consider myself a newbie, Perl convert,
actually) and I think that Quick Python is the best.  It's coverage of
Tkinter is *LOUSY* but the rest of it is great.  The O'Reilly Python book
has a better intro/explanation to Tkinter.  I'm also using Tcl/Tk in a
Nutshell to help me with options etc.

Chuck
William Park <cv601 at torfree.net> wrote in message
news:Fnu1K6.1nz.0.bloor at torfree.net...
> Art Pollard (Pollarda at Lextek.com) wrote:
> : I just purchased Programming Python.  As I am
> : totally new to Python, I learned from this book
> : that I want to learn more.  Unfortunely, I don't
> : think this is the best book for a Python newbie.
> : (Though from what I've seen, it makes a good
> : language reference.)
>
> : So right now, I am looking for another learning
> : book.  The two which popout are:
>
> : Quick Python (In my local bookstore)
> : Learning Python (Not in stock but obtainable)
>
> : Does anyone have any preferences / recommendations???
>
> : I have extensive C/C++ experience but would like a good
> : book to help me break through the wall and reach that
> : Ahhh Haaaa experience when the language becomes clear.
> : (I still remember the Ahhhh Haaa from many years ago
> : when I figured out C's pointers... What a day. :-)
>
> : -Art
>
> I recommend "The Quick Python Book" --- the layout and the typesetting are
> better.  So is the content, but it's biased opinion.
>
> William Park
>
> --




From kuncej at mail.conservation.state.mo.us  Mon Jan 10 16:11:37 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Mon, 10 Jan 2000 15:11:37 -0600
Subject: Python Web Host List [Updated]
Message-ID: <TVre4.314$B5.6129@news.more.net>

A number of recent postings have asked/answered questions like:
" Does anyone know who would host a CGI Python site?"

I have just updated my "Python-Friendly Virtual Website Providers"
page: http://www.corrt.com/info/pyisp-list.html
with all the sites mentioned.

Let me know of any additions/corrections/updates.

  --Jeff







From tim_one at email.msn.com  Tue Jan  4 21:52:22 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 4 Jan 2000 21:52:22 -0500
Subject: how to copy a dictionary
In-Reply-To: <roy-5C6597.21015404012000@netnews.nyu.edu>
Message-ID: <000c01bf5727$e4153780$1a2d153f@tim>

[Roy Smith]
> If I do:
>
> d = {'hedgehog': 'spiney norman'}
> temp = d
> temp['scotsman'] = 'earnest potgorney'
>
> I end up changing the original dictionary, d.  It's obvious
> that what's going on is when I do temp = d I get a pointer
> to the same object instead of a new object.  My question is,
> how do I force a new object to be created, so when modify it,
> I don't also modify the original?

For dicts in particular,

    temp = d.copy()

makes a shallow copy.  For the general case, see copy.copy() and
copy.deepcopy() in the std library.

Note that

    copy() (dictionary method)

is the first "copy" entry in the Library Ref's index.

that's-a-hint<wink>-ly y'rs  - tim





From fredrik at pythonware.com  Wed Jan 12 07:35:13 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 12 Jan 2000 13:35:13 +0100
Subject: PIL 1.0 bugs? (TIFF, Image)
References: <387c6eb2$0$22474@senator-bedfellow.mit.edu>
Message-ID: <001401bf5cf9$7a8a9da0$f29b12c2@secret.pythonware.com>

Robert <soundhack at hotmail.com> wrote:
> looking at the code, apparently the plugin assumes that the SOURCEFORMAT tag
> is a scalar.
> again looking at the traceback, there was a problem in the _getdecoder and
> _getencoder "codec factories", specifically an apply call.

sounds like you're using different versions of
the PIL Python library and the _imaging module.

since there is no thing called SOURCEFORMAT
in the 1.0 release, you're probably using old
python files with a new _imaging module.

better check your installation.

</F>




From tye4 at yahoo.com  Wed Jan 19 19:43:43 2000
From: tye4 at yahoo.com (tye4)
Date: Wed, 19 Jan 2000 16:43:43 -0800
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com>
Message-ID: <865lsm$s3p@news.or.intel.com>

Ivan Van Laningham <ivanlan at callware.com> wrote in message >

> Don't code that way to start with.  It's dumb.  Treat yourself to a good
> editor.
>
Better yet. Fix the dumb language.


> You got about as much chance of getting this into Python as you do of
> getting hit by a meteor.  Less, actually, because the odds for the
> latter are measurable.  What you're asking for is for Python to not be
> Python.  You don't like it, design your own language, build the
> interpreter or compiler, and publish it.  Maybe you can get some people
> to use it.  One of them won't be me.
>
> You're saying to people who know and love Python, "Python is a great
> language except that it's Python.  Let's fix it."  We don't think it's
> broken.  Your insistence on swimming upstream on this issue is not only
> wrong-headed but boring.
>
Python is good but could be better.
Everything has flaws and nothing is perfect. If we can't fix it then its a
different
matter. But in this case the problem can be fixed. It just so happens that
there
are a lot of bull-headed close-minded people than I reckoned there would be.
I'm pretty certain now that this problem will never be fixed.

I suggest you go to other newsgroups and let them know how brilliant Python
is
because it has an invisible block terminator.

Since we are so lazy let's just skip the ending anything and just use
indentation.

Math:
Before:
    (2 + 3) * 5

After Python rule:
    (2 + 3
            * 5

C:
Before:
    if (x == 5)
    {
        printf("blah\n");
    printf("blah blah");

HTML:
Before:
    <h1> Python indentation: <i>Why it is such a pain</i>
   </h1>
    Blah blah blah

After:
    <h1> Python indentation
            <i> Why it is such a pain
    Blah blah blah

Go ahead and convince these guys to switch do it the "Python Way".
And Good luck. You will need it.

-tye4




From mikael at isy.liu.se  Mon Jan 24 09:49:59 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Mon, 24 Jan 2000 15:49:59 +0100 (MET)
Subject: Living without static member functions
In-Reply-To: <86hnac$r83$1@mordred.cc.jyu.fi>
Message-ID: <XFMail.000124154959.mikael@isy.liu.se>

On 24-Jan-00 Sami Hangaslammi wrote:
 >  Mikael Olofsson <mikael at isy.liu.se> wrote in message
 >  news:XFMail.000124133458.mikael at isy.liu.se...
 >  
 >  <snip>
 >  
 > > a=Milkman('Mikael')
 > > b=Milkman('David')
 > > a=Milkman('Mattias')
 > >
 > > then allMilkmen would be
 > >
 > > {'Mikael': <__main__.Milkman instance at e0f68>,
 > > 'David': <__main__.Milkman instance at e0de8>,
 > > 'Mattias': <__main__.Milkman instance at e0aa0>}
 > >
 > > Hence, allMilkmen still keeps a reference to the first instance
 > > (Mikael), which has been replace by the third (Mattias).
 >  
 >  Because objects use reference semantics, Mattias does NOT replace
 >  Mikael. The only thing changed is the local variable 'a' which now
 >  referes to another object. 

Which was exactly what I meant. 

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    24-Jan-00
Time:    15:46:59

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From bkc at murkworks.com  Mon Jan  3 16:34:15 2000
From: bkc at murkworks.com (Brad Clements)
Date: Mon, 3 Jan 2000 16:34:15 -0500
Subject: NetWare port of Python - not so hot pystone
Message-ID: <200001032133.QAA08724@anvil.murkworks.com>

Having completed an initial port of Python to NetWare, I've been testing 
it out to see how well it performs.

On a PII-233 with 128 meg ram I can run either NetWare 5.1, or 
Windows NT 4.0 SP3.

When running NT, Pystone 1.1 returns 3634

When running on NetWare in protected mode I get 2748 with 
setcheckinterval(<big number>), or 1628 using the default check interval

When running on NetWare in non-protected mode, I get 3086 with 
setcheckinterval(<big number>), or 2044 using the default check interval.

The NetWare version of python15.nlm was compiled using Pentium 
instructions and scheduling.

Because NetWare is a non-preemptive OS, the check interval also 
controls when Python15.nlm explicitely calls ThreadSwitch(). Setting 
the check interval to a higher number has a big impact.

--

Still, these numbers are rather crummy. What kinds of things in a CLIB 
could have a large impact on Python?

1. memory allocation routines
2. string manipulation
3. <other>???

I suspect that the memory allocation functions in clib aren't very good, 
and could be replaced with something better. Does anyone know of a 
free source of a very good dynamic memory allocator?

Perhaps there is a compile time option I'm missing here? I don't think 
my config.h has debugging or memory tracing enabled.

I'm using the Borland Compiler, if I used VC would that make a big 
impact too?

Suggestions welcome. Thanks

ps. Zope runs on NetWare too.


Brad Clements,                bkc at murkworks.com   (315)268-1000
http://www.murkworks.com                          (315)268-9812 Fax
netmeeting: ils://ils.murkworks.com               AOL-IM: BKClements



From smigowsky at dspace.de  Tue Jan 18 03:54:40 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Tue, 18 Jan 2000 09:54:40 +0100
Subject: WIN32 : how to list the current processes by their executable na
	me?
Message-ID: <326536345498D311B3BC00105A39802A07457E@newsintern.dspace.de>

Hi,

see the example given by Mark Hammond in the win32 package

...\Python\win32\scripts\killProcName.py

Stefan

>-----Original Message-----
>From: Yann SAMAMA [mailto:ysamama at cybercable.fr]
>Sent: Monday, January 17, 2000 8:47 PM
>To: python-list at python.org
>Subject: WIN32 : how to list the current processes by their executable
>name?
>
>
>Hi everybody!!
>
>Is there a simple way to implement an os.kill() that works on all WIN32
>platforms?
>I know about the FAQ article, but I want the kill function to 
>take the name
>of the executable file as parameter, not the Process ID.
>
>It should work like this :
>
>
>def kill(exe_file):
># assumes we're on a Win 32 platform
>    import win32api
>
>    mylist = getcurrentprocesses(exe_file)  # that's the key 
>function, it
>retrieves a list of corresponding PID
>
># if exec_file is "myprog.exe", we kill all instances of this 
>program that
>are actually running...
>
>    for pid in mylist:
>        handle = win32api.OpenProcess(1, 0, pid)   # that's the FAQ
>implementation of kill function
>        win32api.TerminateProcess(handle, 0)
>
>
>Any idea about how writing this function "getcurrentprocesses()" ?
>
>
>I've heard about a POSIX port of "kill" on NT that accepts a file name
>with -f option, but I'm afraid I won't be allowed to install 
>it on my NT
>platform...
>
>Thanks in advance,
>
>Yann.
>
>
>-- 
>http://www.python.org/mailman/listinfo/python-list
>



From grant at nowhere.  Mon Jan 17 12:35:17 2000
From: grant at nowhere. (Grant Edwards)
Date: Mon, 17 Jan 2000 17:35:17 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
Message-ID: <slrn886kml.284.grant@grante.comtrol.com>

I just found the http server page on Exchange server that
allows access to e-mails via a web browser.  What I'd really
like is to write a Python program that just sucks the mail out
of the inbox and delivers it locally.

I've never done any Python code that accesses an http server.
Are there any applications which would be good examples?

-- 
Grant Edwards                   grante             Yow!  Oh, I get it!! "The
                                  at               BEACH goes on," huh,
                               visi.com            SONNY??


From jbauer at rubic.com  Fri Jan 14 16:52:41 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 14 Jan 2000 15:52:41 -0600
Subject: class.__getattr__ behavior:  feature or bug?
References: <000801bf5ed7$6cd5e540$a0a2143f@tim>
Message-ID: <387F9AA9.1A1654DF@rubic.com>

Tim, Huayin:

Raising an AttributeError exception in the __getattr__()
method fixes Huayin's problem, assuming Tim's suggestion
isn't applied. <wink>:

class S:
    def __init__(self):
        self.a = 0
    def __getattr__(self, name):
        if name == 'c':
            self.a = self.a + 1
            return self.a
        else:
            raise AttributeError, name

if __name__ == '__main__':
    s = S()
    print "s.a =", s.a
    print "s.c =", s.c
    print "s =", s
    try:
        print "s.b =", s.b
    except AttributeError:
        "AttributeError 's.b' not found"

---
Jeff Bauer
Rubicon Research

Tim Peters wrote:
> 
> [Huayin Wang]
> > >>> s=S()
> > >>> s
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > TypeError: call of non-function (type None)
> > >>>
> 
> Feature.  Try sticking "print name" in __getattr__ to see 
> what's happening. Putting "s" on a line by itself implicitly 
> asks Python to invoke repr(s) (to build an output string), 
> so Python tries to look up the special method s.__repr__.  
> Your __getattr__ returns None, and Python complains that None
> isn't a sensible __repr__ method.
> 
> To fix that, don't do this <wink>, define S.__repr__, or 
> capture the name "__repr__" in your __getattr__ and return 
> something sensible for it.



From aa8vb at yahoo.com  Fri Jan 21 10:11:00 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 21 Jan 2000 10:11:00 -0500
Subject: Exiting program
In-Reply-To: <38873769.5048399@122.0.0.250>
References: <38873769.5048399@122.0.0.250>
Message-ID: <20000121101100.A184338@vislab.epa.gov>

volumeregeling:
 |I am making a menubar which uses functions from a cupple of COM
 |objects. I want to exit my program through a stop_program function not
 |from any COM object.
 |
 |def Exit_program():
 |    Editor.mnuFile_Exit()
 |    sys.exit()                        <-- for some reason this also
 |doesn?t work. but that on problem jet.

Keep in mind that this doesn't terminate your program.  It merely raises a
SystemExit exception.  If it is allowed to propagate up past the top level
of your program, the Python interpreter will terminate.  However, it can be
caught and ignored, which is most likely what is going on:

    >>> import sys
    >>> try:
    ...   sys.exit(0)
    ... except:
    ...   print "Request denied"
    ... 
    Request denied
    >>>

-- 
Randall Hopper
aa8vb at yahoo.com



From jasont at wellmed.com  Tue Jan 25 20:45:31 2000
From: jasont at wellmed.com (Jason Taylor)
Date: Tue, 25 Jan 2000 17:45:31 -0800
Subject: Python startup seems slow... (NT)
Message-ID: <MPG.12f7f19a8fa3be15989680@or.news.verio.net>

Under NT, it seems that starting up python to process even a trivial 
script is pretty slow.  With the following python program, 
helloworld.py:

  print "Hello World"

it takes a second or more for "python helloworld.py" to execute on my 
machine.  It doesn't get any faster the second or third time around.

Curious, I used the same program with Perl.  It had basically no delay.

This is a problem for me because I have a python script which needs to 
be called repeatedly from another language.  I could use Perl, but I 
would rather not :)

Is there anything I can do to improve the performance?  Is this a known 
problem?

Thanks!
Jason Taylor



From wlfraed at ix.netcom.com  Sat Jan 22 15:42:33 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Sat, 22 Jan 2000 12:42:33 -0800
Subject: Python dumps core with this.. how come?
References: <38891316.393423@news.erols.com> <005301bf64b5$d2f234d0$f29b12c2@secret.pythonware.com> <3889c4b6.242514@news.erols.com> <004401bf64ef$a8e1e0c0$f29b12c2@secret.pythonware.com>
Message-ID: <eBOKOJ5H=Z=+SzINCNiw4F0OiAfu@4ax.com>

On Sat, 22 Jan 2000 16:45:06 +0100, "Fredrik Lundh"
<fredrik at pythonware.com> declaimed the following in comp.lang.python:

> 
> 5. simple assignments don't copy any data,
>    they rebind the name
> 
> (here, a simple assignment is an assignment
> of the form "name = expression".  more com-

	This was something that took me a few minutes to understand when
I first encountered it myself.

	I finally had to resort to the old "intro to computing 101"
mailbox simile.

	The traditional example is a set of mailboxes each with a name
on the front. A statement of the form "A = <constant>" was represented
by taking out whatever paper was in the box named "A", and putting in a
paper with <constant> on it. "A = <variable name>" meant making a copy
of the paper in the box named <variable name> and putting that copy into
the "A" box.

	Python seems to reverse this.

	"A = <constant>" looks for the box with the name "A" on it (if
it exists), removes the name-label, and then sticks the label on
whatever box contains the paper with <constant>. Same for "A = <variable
name>" -- The box with the name "A" on it is moved to whichever box
already has the label <variable name>. Potentially, this can leave a box
with no labels attached (ready to be garbage collected), and does leave
the other box with more than one label.
--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From piers at cs.su.oz.au  Thu Jan 27 21:37:15 2000
From: piers at cs.su.oz.au (Piers Lauder)
Date: Fri, 28 Jan 2000 13:37:15 +1100
Subject: imaplib and fetch results
Message-ID: <949027567.96.869751374@cs.usyd.edu.au>

On 27 Jan 2000 17:22:29 +0100, Harald Terkelsen wrote:

 > I've been playing with the imaplib.py module and found some difference
 > in the returned data of the fetch function depending on the
 > message_part argument.
 > 
 > The command:
 > 
 > typ, data = M.FETCH('13:17', '(FLAGS envelope)')
 > 
 > returns a list where data[0] is the flags and envelope data from
 > message 13, data[1] for message 14 and so on. This is what i would
 > expect.
 > 
 > But the result of the following command is somewhat different:
 > 
 > typ, data = M.FETCH('13:17', '(FLAGS body.peek[header.fields (From Subject)])')
 > 
 > Now data[0] contains the information from message 13 as in the previous
 > fetch command. I would then assume data[1] to contain the requested
 > information for message 14, but it doesn't. Instead data[1] contains
 > the character ')'. data[2] contains the data for message 14 and
 > data[3] gives me again an ')' and so on.
 > 
 > Does anybody know why it is like this, or is it a bug?

This is a feature, not a bug :-)

The reason is that to fix the problem you are seing I would have to
enhance imaplib to parse all responses from all commands, and that's
something I'd prefer that the user do.

The `correct' way to deal with responses from an IMAP4 server is to
concatenate all the strings returned, and then parse the response per
the RFC.

The server's responses to multi-part commands is to return a series of
packets in the form "description [literal]" and the imaplib returned
data reflects this (you'll see a list item consisting of a tuple of 2
strings). So if the server sends a packet consisting of just a single
')' (presumably terminating a parenthesis in a previous description,
probably after some "literal" data), you'll see it as a single list
item following the one with the data.

Hope that helps!

Piers.







From fredrik at pythonware.com  Fri Jan 14 09:34:48 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 14 Jan 2000 15:34:48 +0100
Subject: file width and height
References: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com> <ubqf4.3018$hN5.92483@typhoon.nyroc.rr.com> <004b01bf5e68$92965a40$f29b12c2@secret.pythonware.com> <026a01bf5e98$12edfd90$0100a8c0@rochester.rr.com>
Message-ID: <002d01bf5e9c$82f9a090$f29b12c2@secret.pythonware.com>

> I screwed up here.

no problemo.

(side note: you don't need PIL's binary extensions to
just identify and measure image files; just include
Image.py, ImageFile.py, and whatever plugins you
need)

</F>




From reic0024 at ub.d.umn.edu  Mon Jan 31 13:30:12 2000
From: reic0024 at ub.d.umn.edu (Aaron J Reichow)
Date: Mon, 31 Jan 2000 12:30:12 -0600
Subject: Python1.5.2/Zope 2.1.3 on MacOS X Server notes
In-Reply-To: <B4BB2B2F.4F4F%jpshell@bigfoot.com>
References: <B4BB2B2F.4F4F%jpshell@bigfoot.com>
Message-ID: <Pine.SO4.4.02.10001311229590.9750-100000@ub.d.umn.edu>

Thanks much! I'll have try at it. :)

Aaron




From tlu at bmo.com  Wed Jan 12 09:28:04 2000
From: tlu at bmo.com (Tao Lu)
Date: Wed, 12 Jan 2000 09:28:04 -0500 (EST)
Subject: Perl speed vs. Python speed
Message-ID: <200001121428.JAA04066@trm.bmo.com>

Perl is fine-tuned for text processing. So try to compare with something else. 
IMHO, run-time performance should not be the main reason to choose Python.

Cheers,
_______________________________________________________________________
<name>Tao Lu</name><email>tlu at iname.com</email><phone>867-6722</phone>

> Date: Wed, 12 Jan 2000 10:02:23 GMT
> From: rwgk <rwgk at my-deja.com>
> To: python-list <python-list at python.org>
> Subject: Perl speed vs. Python speed
> 
> I just wrote a tiny Python program to do some text processing.
> When I ran it I was very surprised how long it took to finish.
> My first guess is that Perl would be a lot faster, maybe even a
> factor 10, but I did not do a direct comparison.
> 
> Could someone with more experience comment on Perl speed vs.
> Python speed?
> 
> Thanks!
>         Ralf
> 
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> -- 
> http://www.python.org/mailman/listinfo/python-list






From ray at commsecure.com.au  Wed Jan 12 06:43:25 2000
From: ray at commsecure.com.au (Ray Loyzaga)
Date: Wed, 12 Jan 2000 22:43:25 +1100
Subject: CGI problem
References: <855k3e$agm$1@nnrp1.deja.com> <Pine.WNT.4.21.0001100945530.162-100000@wks_3.e-shopwork.com>
Message-ID: <387C68DD.71D3F226@commsecure.com.au>

Sposhua wrote:
> 
> On Fri, 7 Jan 2000, sp00fD wrote:
> 
> > I'm creating a CGI program, and I keep getting an error from apache:
> >
> > "(8) Exec format error exec of /usr/local/apache/htdocs/...
> > Premature end of script headers: /usr/local/apache/htdocs/..
> >
> > I've put the print "Content-type: text/html\n" (with anywhere from 0 to
> > 8 newlines after it) in there.
> >
> > What am I doing wrong?
> 
> Either
> a) The script has errors
> or
> b) You have to add 'HTTP:/1.1 200 OK\n' at the begining of the header. Don't ask
> me why, but on my server I can't run anything without that.

I think the exec is failing .... is the path for the cgi file accessible
to "nobody", do you start with #!/path to python executable
The other gotcha with this sort of error is to do a system("/bin/date") or
similar and not flush stdout before to make sure the headers are not
buffered and come out after the output of date.


From tim_one at email.msn.com  Tue Jan  4 01:13:58 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 4 Jan 2000 01:13:58 -0500
Subject: Python not a Very High-Level Language?
In-Reply-To: <1265170846-519622@hypernet.com>
Message-ID: <000701bf567a$e3434a20$1a2d153f@tim>

[Tres Seaver]
> What reasonable use can you propose for an associative
> container which allows mutable keys?  You might as well revert
> to a list of (key,value) pairs, since you're going to do a
> linear search on each lookup anyway if the keys are allowed to
> change.

[Neel Krishnaswami]
> You can compute the hash function using the object's identity,
> rather than its value. IOW, hash on the object's address. Then
> you can mutate the contents of the object as needed and still
> access the hash. In fact, I believe this is the default behavior
> for Common Lisp hash tables (though of course you can create hash
> tables with different hash functions).
> ...

[Gordon McMillan]
> Well, you could, but then you have to pass around the one-
> and-only key to access the value. Seems more sensible, as
> Tres said, to pass around (key, value) instead and just forgo
> the hash.

Na, there are lots of uses for this.

Simple one:  you have a list of lists (or of any other kind of mutable
object) and want to weed out duplicates.  If it were a list of (say) ints,

def weed(bag):
    temp = {}
    for x in bag:
        temp[x] = 1
    return temp.keys()

is obvious and fast.  The same algorithm (although not this Python spelling
of it) is safe & effective for any kind of objects with hash functions that
don't themselves mutate the objects.

Fancy one:  In state space searches where the operators are reversible
and/or the objects have exploitable symmetries, the natural way to detect
cycles and reduce redundant search is to make the next-state generator
return canonical representatives (kinda like interning a string in Python
today).  Although Python won't let you spell it this way directly, keeping a
dict of representatives (viewed as a set) is the "obvious and fast" way to
do that in the generator.

Outside the generator, in this case you are indeed passing around "the
one-and-only key", so info can be stored in the object; but you still want
to do stuff like e.g. associate info with *pairs* of objects via

    dict[obj1, obj2] = info

notation.  I have yet to find a really slick way to do that in Python.  For
purposes of the generator, I wrap stuff in classes with

    def __hash__(self):
        return something based on all the \
            relevant members of self

    def __cmp__(self, other):
        likewise

but *outside* the generator wrap in classes with

    def __hash__(self):
        return id(self)

    def __cmp__(self, other):
        return cmp(id(self), id(other))

The problem is that these two views of the objects aren't that sharply
separated in real life, and stupid mistakes are common.

The "proper" ways to hash and compare really depends on the use to which
each dict is being put, not on the objects therein.

just-because-you-*can*-mutate-something-doesn't-mean-you-
    will-as-all-your-worst-habits-can-testify<wink>-ly y'rs  - tim





From dieter at handshake.de  Sun Jan 23 05:10:43 2000
From: dieter at handshake.de (Dieter Maurer)
Date: 23 Jan 2000 11:10:43 +0100
Subject: Python - SIGPIPE handled bug
References: <20000121162925.A189033@vislab.epa.gov>
Message-ID: <x7ya9h5c9o.fsf@lindm.dm>

Hallo Randall

Randall Hopper <aa8vb at yahoo.com> writes on Fri, 21 Jan 2000 16:29:25 -0500:
>     > python -c "for i in xrange(1000): print 'Some output'" | head -2
>     Some output
>     Some output
>     Traceback (innermost last):
>       File "<string>", line 1, in ?
>     IOError: [Errno 32] Broken pipe
>   ---------------------------------------------------------------------------
> 
> python getting a SIGPIPE is appropriate behavior.  "Whether" it gets a
> SIGPIPE is up to the shell, in UNIX:
It seems, that in your case Python does not get a SIGPIPE but
the "write" returns "-1" and "errno==EPIPE".

According to POSIX, this is the behaviour, if SIGPIPE is not handled
by the default action:
       EPIPE  fd is connected to a pipe or socket  whose  reading
              end  is closed.  When this happens the writing pro??
              cess will receive a SIGPIPE signal; if it  catches,
              blocks or ignores this the error EPIPE is returned.

The SIGPIPE handling under Python 1.5.2/Linux 2/Bash seems to be
to ignore the signal (possibly inherited from the bash!).

If you do not like this, use:
	import signal
	signal.signal(signal.SIGPIPE,signal.SIG_DFL)

Dieter


From rdudfield at my-deja.com  Sat Jan 29 00:17:29 2000
From: rdudfield at my-deja.com (rdudfield at my-deja.com)
Date: Sat, 29 Jan 2000 05:17:29 GMT
Subject: statements and blocks
References: <86t0ov$4en$1@news.psc.edu>
Message-ID: <86tt56$d4a$1@nnrp1.deja.com>

Speaking of blocks, is there a way to show where the block ends?

I have been caught a few times doing this:
class b():
  def asdf():
    print 'a'
    if bla:
      print 'a'
      print 'a'
      ... more than a page of stuff.

   # method is supposed to be in the classes scope.
    def qwer():
      print 'a'
      print 'a'
      print 'a'


Basically because the function asdf is more than a page long.

I find indentation is better in most cases except this.  This isn't too
much trouble, as I usually find the problem pretty quickly.  However
occasionally I waste a bit too much time on it than I'd like to :)

Rene.


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


From mgushee at havenrock.com  Mon Jan 17 15:49:20 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 17 Jan 2000 15:49:20 -0500
Subject: Tkinter Tab orders and keybindings
References: <3883775C.9369120@exceptionalminds.com>
Message-ID: <m3puv0todr.fsf@havenrock.com>

Hi, Tim--

Timothy Grant <tjg at avalongroup.net> writes:

> How does one control the tab order of Tkwidgets? Is it even possible?

AFAIK the only 'built-in' way to do this is by controlling the order
in which you *create* the widgets -- which, of course, is not
necessarily related to their positions on screen.

If that doesn't work for you, you might have to write your own 'next'
and 'prev' methods, which would use your own list of widgets. If you
don't need the Tab key for anything else, you could use the following
bindings in your toplevel window.

self.bind_all('<Tab>', self.next)
self.bind_all('<ISO-Left-Tab>', self.prev)

> also, if I bind the Tab key to a widget, how do I get it to perform its
> original function as well as my function?

Here, again, I think you would need your own traversal method. Then,
of course, you either include some additional actions in the traversal 
method, or bind <Tab> to a handler that invokes the traversal method + 
something else.

Hope this helps.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From greg at cosc.canterbury.ac.nz  Wed Jan 26 22:20:39 2000
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 27 Jan 2000 16:20:39 +1300
Subject: How to store a function pointer in class?
References: <867j5p$qvq$1@mordred.cc.jyu.fi> <vndaem01s1r.fsf@betelgeuse.ccs.neu.edu> <867nbb$1ho$1@mordred.cc.jyu.fi> <vnd66wo1nz2.fsf@betelgeuse.ccs.neu.edu> <867ssq$a3c$1@mordred.cc.jyu.fi>
Message-ID: <388FB987.BD516313@cosc.canterbury.ac.nz>

Sami Hangaslammi wrote:
> 
> Yeah, I figured this out myself, but it still seems like an
> unneccessary and ugly hack.

It might be less ugly to use just one wrapper object for
all the functions in a given class, e.g.

class FunctionHolder:
    pass

class MyClass:
    functions = FunctionHolder()

MyClass.functions.f1 = some_function
MyClass.functions.f2 = another_function

MyClass.functions.f1(42)

etc.

-- 
Greg Ewing, Computer Science Dept,
+--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg at cosc.canterbury.ac.nz	   +--------------------------------------+


From jasont at wellmed.com  Thu Jan 27 15:44:35 2000
From: jasont at wellmed.com (Jason Taylor)
Date: Thu, 27 Jan 2000 12:44:35 -0800
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com> <86q5vv$mci$1@nnrp1.deja.com>
Message-ID: <MPG.12fa4e10dbb016cb989683@or.news.verio.net>

I copied the following from the MSDN Library, October 1999.

-----------------
Delay-Loaded DLLs
-----------------

Earlier, we discussed the high expense of a page fault relative to a CPU 
cycle. We also pointed out loading a lot of unnecessary code can cause 
page faults during a program's operation as well as slowing the 
program's initial load. Aside from keeping your code small to begin with 
by careful design and intelligent use of the compiler's optimizations, 
you can help to avoid this problem by loading code only when needed.

Windows has always supported dynamic link libraries (DLLs) and dynamic 
function loading. Unfortunately, calling LoadLibrary and GetProcAddress 
to load a function and balancing LoadLibrary calls with FreeLibrary 
calls are somewhat tedious and error-prone tasks for developers.

The Visual C++ linker now supports the delayed loading of DLLs in a way 
that's almost transparent. You can call functions from your code as 
though they are statically linked and specify the /delayload:<dllname> 
linker switch for each DLL you do not want loaded with the .exe. You 
also must link to Delayimp.lib, which contains the delay load helper 
routines. The linker generates a load thunk for imported functions in 
delay-loaded DLLs.

A thunk is a short writeable code block that contains a function 
pointer; Windows uses thunks to implement code relocation and to bridge 
between 16-bit and 32-bit functions. The thunk we're talking about here 
is what enables transparent delayed loading. It is initially set to call 
the delay-load helper routine.

So, when your code first uses a DLL that has been marked for delayed 
loading, the delay-load helper function is called. It checks the stored 
handle for the DLL, sees it has not been loaded, and brings it into 
memory with LoadLibrary. On the first use of each function in the DLL, 
the helper function retrieves its entry point with GetProcAddress and 
saves it in the thunk. On subsequent uses of the function, the function 
is called directly from the stored pointer in the thunk.

You can explicitly unload delay-loaded DLLs after you are done with them 
by calling the following:

__FUnloadDelayLoadedDLL 

It is not necessary to do so, however.




From boud at rempt.xs4all.nl  Wed Jan 12 13:20:41 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 12 Jan 2000 18:20:41 GMT
Subject: Now What? :o)
References: <n2bd4.64$%Ph.18874529@news.randori.com>
Message-ID: <85iglp$qiv$1@news1.xs4all.nl>

-james <NoSpam at linkamericacorp.com> wrote:
<...>

> I'm almost through the online tutorial, and I'm hooked. I don't see any info
> on coding GUI  forms though. Can anyone recommend a website, sample code, or
> a good book that covers how to do this? Are GUI forms a platform dependant
> issue? I'm coming from a world of drag and drop and, from what I gather,
> this done programatically in python.

Being both a VB programmer (at work) and a Python programmer (at home),
I think I understand you. Well, you have to code out guis by hand,
coding every button your form shows. If you want easy examples of 
an easy gui, go to:

http://www.xs4all.nl/~bsarempt/python/tutorial.html.

There is the rudiments of a drag-and-drop gui-designer, vipyl, but I
fear the author has lost heart and quit the project. That might well be
because coding a Python gui can be really easy, and because drawing a
gui doesn't make much sense with layout managers anyway.



-- 

Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt


From P.J.W.S.VRIJLANDT at INT.azg.nl  Wed Jan 26 11:46:40 2000
From: P.J.W.S.VRIJLANDT at INT.azg.nl (P.J.W.S. Vrijlandt)
Date: Wed, 26 Jan 2000 17:46:40 +0100
Subject: Keywordargument in string.split
Message-ID: <1D9EA098C@nds-3.azg.nl>

Hi all,

I have this error I don't quite understand.
Why can't I give 'sep' as a keyword argument?

It seems that the string module I import, is not the string.py in the 
standard distribution but a not-quite-equivalent c-implementation of 
it.

Patrick

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


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 string
>>> string.split('abcd', sep = 'c') 
Traceback  (innermost last):
  File "<pyshell#1>", line 1, in ?
    string.split('abcd', sep = 'c')
TypeError: this function takes no keyword arguments
>>> def split(s, sep=None, maxsplit=0):  # = string.split
           print s, sep, 
maxsplit >>> split('abcd', sep = 'c')
abcd c 0



From mgushee at havenrock.com  Thu Jan  6 22:32:17 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 06 Jan 2000 22:32:17 -0500
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com> <38744512.377982379@news.isomedia.com> <m3n1qj5or5.fsf@havenrock.com> <20000107005230.23566@nms.otc.telstra.com.au>
Message-ID: <m31z7utv3y.fsf@havenrock.com>

Greg McFarlane <gregm at iname.com> writes:

> Tkinter is not "careful" enough, IMHO.  It exports, inter alia, the
> following names:
> 
>     At E EW N NE NO NS NW ON S SE SW Tk W Wm X Y
>     ALL END OFF SEL TOP YES
>     BOTH CHAR FLAT Grid LEFT Menu Misc NONE NSEW Pack TRUE Text WORD
> 
> I'm not sure how you can say that "X" and "Y" are not "commonly-used
> names".

Gosh, I've never used them ;-)

Good point, though. I was only thinking of the class names.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From hellan at acm.org  Tue Jan 25 09:53:02 2000
From: hellan at acm.org (Jon K Hellan)
Date: 25 Jan 2000 15:53:02 +0100
Subject: Embedding -- getting error/traceback information not to stderr?
References: <slrn88oujk.ebi.erw@erw-pc.dde.dk>
Message-ID: <87d7qqrynl.fsf@parus.parus.no>

erw at dde.dk (Erwin S. Andreasen) writes:

> In my application using embedded Python, I'd like to print out the error
> information as it occurrs (traceback, etc.). This is what PyErr_PrintEx()
> (called from PyErr_Print()) does -- unfortunately, it does it to sys.stderr.

I don't know if you want a solution in Python or C. Python solutions have
been given. Here's an example in C:
 
static char *
convert_py_exception_to_string ()
{
	char *header = _("Python exception");
	char *retval = header;
	char buf [256];
	int pos;
	
	PyObject *ptype = NULL, *pvalue = NULL, *ptraceback = NULL;
	PyObject *stype = NULL, *svalue = NULL;
	
	PyErr_Fetch (&ptype, &pvalue, &ptraceback);
	if (!ptype)
		goto cleanup;
	stype = PyObject_Str (ptype);
	if (!stype)
		goto cleanup;
	pos = snprintf (buf, sizeof buf, "%s: %s",
			header, PyString_AsString (stype));
	retval = buf;
	
	if (pvalue && (pos + 3 < sizeof buf))
		svalue = PyObject_Str (pvalue);
	if (!svalue)
		goto cleanup;
	snprintf (buf + pos , sizeof buf - pos , ": %s",
			PyString_AsString (svalue));
cleanup:
	Py_XDECREF (stype);
	Py_XDECREF (svalue);
	PyErr_Restore (ptype, pvalue, ptraceback);

	return strdup (retval);
}

The function returns a string without the traceback info. You'll
probably want to include the traceback. I call PyErr_Restore in case I
also want traceback to the console.

HTH

Jon


From apederse at siving.hia.no  Sun Jan 16 11:42:06 2000
From: apederse at siving.hia.no (Asle Pedersen)
Date: Sun, 16 Jan 2000 17:42:06 +0100
Subject: IMAP - fetch using imaplib
Message-ID: <85ssc4$8jk$1@snipp.uninett.no>

I have adressed this question to the group previoussly without any success
so I try once more :
I need help with the following IMAP command:
DF56 FETCH 1:3 (RFC822.SIZE BODY[HEADER.FIELDS (FROM SUBJECT)]).
I have tried using the imaplib for other fetch commands without trouble, but
this one I can't seem to get working utilising the imaplib.

-Asle Pedersen




From aloy at ctv.es  Sat Jan 29 15:38:30 2000
From: aloy at ctv.es (Antoni Aloy)
Date: Sat, 29 Jan 2000 21:38:30 +0100
Subject: Searching for text mode user interface
Message-ID: <86vj0f$h4d$1@lola.ctv.es>

Hi!
I'm looking for something like borland turbo vision. That is a framework to
make friendly user interfaces. Something like wxWindow but in text mode and
under MS-DOS (or portable text mode).

Any suggestions?

Regards,

Antoni Aloy




From mikael at isy.liu.se  Fri Jan 14 03:40:12 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 14 Jan 2000 09:40:12 +0100 (MET)
Subject: Newbie
In-Reply-To: <NVrf4.1234$bQ1.17968@dfiatx1-snr1.gtei.net>
Message-ID: <XFMail.000114094012.mikael@isy.liu.se>

On 13-Jan-00 Allen Cleveland wrote:
 >  ok, i'm new to this and want to get into programming.  What do i need to
 >  D/L?

Everything you need is at www.python.org

Go to "download".

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    14-Jan-00
Time:    09:39:06

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From tiddlerdeja at my-deja.com  Mon Jan 17 13:24:26 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Mon, 17 Jan 2000 18:24:26 GMT
Subject: MSSQL database access module?
Message-ID: <85vmo5$kia$1@nnrp1.deja.com>

Could someone recommend a module to use to access a MS-SQL database
from python?

Any help appreciated.

Thanks.


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


From cfollett at gac.edu  Thu Jan 13 13:49:30 2000
From: cfollett at gac.edu (Charles Follett)
Date: Thu, 13 Jan 2000 12:49:30 -0600
Subject: problem with Demos/sockets/mcast.py
Message-ID: <200001131849.MAA25767@giant-curled.mcs.gac.edu>

Whenever I attempt to select a multicast group, like when running
mcast.py, I get:

  File "mcast.py", line 89, in openmcastsock
    s.setsockopt(IPPROTO_IP, IP_ADD_MEMBERSHIP, mreq)
socket.error: (22, 'Invalid argument')

Multicast works on my machine, and 'mcast.py -s' sends just fine
(verified with a sniffer.) What is preventing me from listening? Also,
I have two ethernet interfaces on my machine, and I was wondering how
to manually choose what interface is used. Thanks in advanced.

charley



From tim_one at email.msn.com  Fri Jan 21 16:01:57 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 21 Jan 2000 16:01:57 -0500
Subject: In defence of the two-namespace rule
In-Reply-To: <869uel$suj$1@nntp1.atl.mindspring.net>
Message-ID: <000601bf6452$c123ada0$caa0143f@tim>

[Aahz Maruch]
> All right, I've seen it enough times this week, would somebody
> mind giving the Stupid Person's definition of currying?

In the extreme, splitting a function of N arguments into a composition of N
1-argument functions.  Commonly used to mean any gimmick for deriving a
function with fewer arguments that holds some number of F's arguments fixed.
For example, binary "+" can be viewed as a function F that takes a number,
and returns a function G that takes a number and returns the sum of that
number with the number passed to F.

Despite the verbal convolution of that example <wink>, it's a very general &
elegant technique used in some functional languages which have nothing *but*
1-argument functions (like Haskell, named in honor of the American logician
Haskell B. Curry -- and "currying" is also named in honor of him).

This can be emulated in Python but it "looks clumsy" here; it's very natural
in languages that support it directly.  For example, if "f" is any
(conceptual) 2-argument function in Haskell,

    g = f 42

defines a 1-argument function "g" such that (switching to Python notation
now!)

    g(x) == f(42, x)

for all x.  Haskell has several shorthand notations for currying of various
sorts.  For example,

    (- 1)

is the function that subtracts one from its argument, while

    (1 -)

is the function that subtracts its argument from 1.  Mix that into the usual
functional stew of higher-order functions and maps etc, and it's a darned
tasty meal!  Not to everyone's tastes, though.

Historical note:  Curry had nothing to do with computers.  He worked in the
foundations of mathematical logic, specifically the branch called
"combinatory logic".  This branch showed that the concept of (roughly
speaking) "variable names" isn't necessary, and, indeed, given all the
gotchas caused by bound and free vrbl names in formal logic, many complex
proofs become a heck of a lot more transparent without them.  An offshoot
showed that this is directly applicable to computational theory as well, and
leads to an amazingly easy and clean (tho not particularly efficient in its
naive form) implementation technique for functional languages, one that does
lazy evaluation and many forms of common subexpression elimination "for
free, by magic".

Raymond Smullyan wrote a strange and wonderful "general interest"
introduction to combinatory logic, in the form of a series of puzzles
involving birds(! when a bird hears another bird's song, it may respond with
a song of its own -- so a bird is a one-argument function over the universe
of songs <wink>).  It ends with a more-than-less rigorous proof of Godel's
incompleteness theorem, phrased in bird terms.

If you enjoy this kind of think (and, honestly, who doesn't <wink>?), check
it out:  "To Mock a Mockingbird".

every-bird-in-the-forest-is-flawed-ly y'rs  - tim





From darrell at dorb.com  Wed Jan 12 14:49:43 2000
From: darrell at dorb.com (Darrell)
Date: Wed, 12 Jan 2000 14:49:43 -0500
Subject: Ping in Python
References: <5650A1190E4FD111BC7E0000F8034D26A0F250@huina.oceanic.com>
Message-ID: <07b001bf5d36$2bef8d80$0100a8c0@rochester.rr.com>

From: "Doug Stanfield"
> Chapter 2, Networking, a section titled 'Emulating Ping'.  Good info that
> answers some of the "how does a PING work"
> Demo/sockets/udpecho.py".

AFIK udpecho.py isn't the same thing as ping.
I found some code to listen for a ping, but there's a lot I don't understand
about it.
It picks up pings coming from 127.0.0.1 but not from another machine ??

Ping uses an ICMP echo request, which requires some down an dirty socket
work.

--Darrell
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: ping.py
URL: <http://mail.python.org/pipermail/python-list/attachments/20000112/fadacfe7/attachment.ksh>

From super_banane at yahoo.com  Fri Jan 28 08:47:59 2000
From: super_banane at yahoo.com (super_banane at yahoo.com)
Date: Fri, 28 Jan 2000 13:47:59 GMT
Subject: diet Python
Message-ID: <86s6me$4mb$1@nnrp1.deja.com>


'freeze.py' is a nice utility for deployment, but it does
not get far enough, in the sense that it produces the C
files and the Makefile, but for the local machine only.
That is Ok if you want to ship binaries, but for source
code shipping that is not really it.

Would it be possible to create a 'diet' Python
distribution, without doc, Tools, etc. that would contain
only what is needed to compile the interpreter with maybe
a limited subset of modules? In that case, it would reduce
Python to a small tar file which you could include in
your own source code for distribution. This way, people
will not even notice that they installed Python.

Also good if Python changes in some incompatible ways
from one version to another, and you do not want to support
the changes.

All in all, what would be nice is a 'freeze' version that
supports source code deployment on multiple platforms as
well. Is there anything going in this direction? Any idea?

--
Nicolas


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


From thor at localhost.localdomain  Thu Jan  6 06:28:03 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 6 Jan 2000 11:28:03 GMT
Subject: Python ring propaganda
References: <slrn876erg.18q.thor@localhost.localdomain> <iacbt70r7he.fsf@vier.idi.ntnu.no>
Message-ID: <slrn8792lf.mk.thor@localhost.localdomain>

On 05 Jan 2000 14:08:45 +0100, Magnus L. Hetland <mlh at vier.idi.ntnu.no> wrote:
>thor at localhost.localdomain (Manuel Gutierrez Algaba) writes:
>
>> Python ring propaganda
>> -------------------------
>> 
>> This is a piece of information about the existing (La)TeX ring in:
>> http://www.ctv.es/USERS/irmina/pythonring.html
>> 
>> What is it?
>> From a functional point of view, it's a list of linked web sites
>> that share a common interest or information, in this case TeX
>> related stuff.
>
>This seems more like a TeX ring than a Python ring... (Is this a
>copy/paste error in your posting? ;)
>

Obviously, :), I think I'm a mess of ring master. Yeah, I run
the TeX/LaTeX ring too. :)


-- 
Manolo


From wpostma at ztr.com  Mon Jan 31 13:56:24 2000
From: wpostma at ztr.com (root)
Date: Mon, 31 Jan 2000 13:56:24 -0500
Subject: Python1-5.2/Demo/embed/demo.c core dump
Message-ID: <3895DAD8.2C434EF6@ztr.com>

I am having fun trying to get the basic "embedding" demo to compile,
link and run.

First, on Linux, I needed to add -lpthread to the Library options in the
makefile, and I had to remove -lsocket.   Any idea why that would be?
Hmm.  Do I need to install libsocket? Couldn't find any libsock*.rpm on
my RedHat 6.1 CD so I have no idea what libsocket is.

Anyways, I just removed -lsocket and it links okay.  Then I try to run
it and I get the "Hello Brave New World" message (from printf) and the
first attempt to access Python causes a core dump.   I ran inside gdb
and the crash more specifically appears to be:

Program received signal SIGSEGV, Segmentation fault.
0x400ca093 in _IO_vfprintf (s=0x80a8a78, format=0x8090c04 "(",
ap=0xbffffb5c)
    at vfprintf.c:1059

Any ideas what I'm doing wrong and why this is crashing?

--

Warren Postma
ZTR Control Systems





From hanche at math.ntnu.no  Sun Jan 23 13:03:55 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 23 Jan 2000 13:03:55 -0500
Subject: Bug in Calendar Module
References: <22E71DAEC504D111B78100805FFE9DC72F03CAD4@pfs21.ex.nus.edu.sg> <yttaelz2ldd.fsf@gilgamesh.cse.ucsc.edu> <388AD363.1D96FB45@camelot-it.com>
Message-ID: <pcoemb864xg.fsf@math.ntnu.no>

+ Michael Tiomkin <michael at camelot-it.com>:

|   Unfortunately, this is a bug.  The overflow exists only when you
| treat the time as a signed 32-bit integer.

But certainly every Unix system on which I ever bothered to check
this, time_t is a *signed* 32-bit quantity, allowing times from
1901-12-13 20:45:52 UTC to 2038-01-19 03:14:07 UTC inclusive.

| When you use time as unsigned integer, this will happen only around
| 1970+2**32/3e7, 2090?

Rollover will then happen at 2106-02-07 06:28:15 UTC, to be precise.
However, this solution offers no way to represent times prior to the
unix epoch.

| Ben Gertzfield wrote:
| 
| > There isn't a good solution, except to use a system that has a
| > 64-bit counter for time.

Indeed.  And I am sure all unixes will do so long before 2038.  That
way, rollover will not happen until 504140796-12-04 15:30:08 UTC, if
they are still using the Gregorian calendar at that time.

Which-is-a-safe-bet-they-wont-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From tjg at avalongroup.net  Sun Jan 23 03:32:52 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Sun, 23 Jan 2000 00:32:52 -0800
Subject: TKinter and focus events
Message-ID: <388ABCB4.A51BF817@exceptionalminds.com>

Hi,

Thanks to some kind assistance from the list I have been able to make
Tkinter do a pretty neat trick (at least I think so!)

Using Pmw.EntryFields, When the field does not have focus it displays
numbers formatted to a specification (e.g., $1,234.56). When the
EntryField gets the focus it reformats the number stripping out the
eye-candy (e.g., 1234.56). To make sure the user enters the correct
info, I turn on either 'integer' or 'real' validation when the
EntryField has focus, and turn it off before I set it to the prettified
entry.

It all work phenomenally well, and I'm tremendously pleased with how
easy it was to convert a number in to nicely formatted text with commas.
However, there is one minor annoyance, every time I move focus to the
field I get what sounds like an error beep, everything works correctly,
I just get this beep, I've been trying to track it down but haven't been
able to figure it out yet. It appears to occur when I do the setentry()
call with the unprettified number. Does anyone have any ideas what might
be happening?

    def FmtEstWt(self, event):
        # Event type 9 is FocusIn and type 10 is FocusOut
        if event.type == '9':   #FocusIn
            print 'EstWt FocusIn'
            self.EstWt.configure(validate='integer')
            self.tmp = string.replace(self.EstWt.get(),',', '')
            self.tmp = string.replace(self.tmp, '$', '')
            self.EstWt.setentry(self.tmp)	#Beep appears to go off here.
        else:                   #FocusOut
            print 'EstWt FocusOut'
            self.EstWt.configure(validate=None)
           
self.EstWt.setentry(commanumber(string.atoi(self.EstWt.get()),0,0))

OK, I just thought of something just before I hit my send button, so I
actually setup an "invalidcommand" option for the EntryField. Sure
enough, the invalid function gets called when I do the setentry. This
does not stop the setentry from working though. I've dumped the contents
of self.tmp, and it appears to be a clean string that if I keyed by hand
would pass the validation no problem.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<



From fredrik at pythonware.com  Wed Jan 19 10:08:20 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 16:08:20 +0100
Subject: no multithreading?
References: <864ih0$5is$1@nnrp1.deja.com>
Message-ID: <001f01bf628f$06630020$f29b12c2@secret.pythonware.com>

moonseeker at my-deja.com wrote:
> I have understood that the python interpreter switches to another thread if a
> fixed amount of python commands is completed. Is this true? That means if a
> task hangs on a command (i.e. file open), all tasks are locked?

nope.  python releases the interpreter lock whenever
it calls OS functions that may block.

> What happens if I use sleep() in a task?

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

    "Not all built-in functions that may block waiting for
    I/O allow other threads to run. (The most popular
    ones (time.sleep(), file.read(), select.select()) work
    as expected.)"

http://www.python.org/doc/current/api/threads.html
has more info on how this works.

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! -->




From insyte at emt-p.org  Tue Jan 25 23:17:17 2000
From: insyte at emt-p.org (Ben Beuchler)
Date: Tue, 25 Jan 2000 22:17:17 -0600 (CST)
Subject: vi and python?
In-Reply-To: <20000125185932.A19619@better.net>
Message-ID: <Pine.LNX.4.20.0001252215590.523-100000@marvin.squad51.net>

On Tue, 25 Jan 2000, William Park wrote:

> This is what I have in ~/.vimrc relevant to *.py.  You can move the
> global settings to 'au BufRead *.py set ...' if you like.

It looks like a useful set of settings, from what I've been able to
interpret...  What is accomplished by the section you mention above?

Thanks,
Ben
-- 
"There is no spoon"
	-- The Matrix




From aahz at netcom.com  Mon Jan 31 15:02:03 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jan 2000 20:02:03 GMT
Subject: list -> dict help
References: <3895BE15.A7CF569B@bam.com> <slrn89bidu.p04.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn89bipe.p04.scarblac-spamtrap@flits104-37.flits.rug.nl> <3895DEAE.574232D2@bam.com>
Message-ID: <874pnr$vr9$1@nntp9.atl.mindspring.net>

In article <3895DEAE.574232D2 at bam.com>, Bill Scherer  <scherbi at bam.com> wrote:
>
>My first inclination was to use the dictionary update method, but
>that overwrites existing keys.  I need something to add dictionaries
>together.  I'll hack this together on my own unless someone has
>somthing ready to go...

kjBuckets may do what you want.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From aa8vb at yahoo.com  Wed Jan  5 08:56:47 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 5 Jan 2000 08:56:47 -0500
Subject: Python and GUI options
In-Reply-To: <386fa6b1.58568155@news.telus.net>
References: <84lreq$hib$1@nnrp1.deja.com> <386fa6b1.58568155@news.telus.net>
Message-ID: <20000105085647.C9441@vislab.epa.gov>

guppy:
 |Have you looked at the wxPython interface to wxWindows?  It's
 |cross-platform compatible.  It'd be interesting to hear if it's quick
 |enough for you.  

 |Certainly at the rate its development is proceeding, it's set to become
 |the new GUI standard for Python (thank gods!) (or, rather, thank Robin!)

Dose of reality.  It has to take a big step in the cross-platform area
before that can happen.  Tcl/Tk/Python builds without a hitch on a slew of
UNIX boxes ("configure", "make", "make install" - done).  Sun, SGI, DEC,
FreeBSD, Linux, etcetc.  Tweaking a few lines in Python's Setup.py is the
only "manual drudgery" involved.

wxWindows/wxGTK/wxPython requires someone with experience and persistence
to help it along on a standard SGI UNIX box, in compilation and
configuration.

Maybe someday though.  wx* certainly sports some nice widgets.  I'd like to
see them standard in whatever toolkit Python uses next year (I'd use them),
but not at the expense of cross-platform support.

-- 
Randall Hopper
aa8vb at yahoo.com



From thamelry at vub.ac.be  Thu Jan 20 10:20:12 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 20 Jan 2000 15:20:12 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com> <8653jg$esi@news.or.intel.com> <002401bf62b8$c30da8a0$f29b12c2@secret.pythonware.com> <8656ne$h1j@news.or.intel.com> <slrn88c7qh.ja9.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>
Message-ID: <86793c$jqd$1@mach.vub.ac.be>

Stephen Coursen <coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com> wrote:
[snip]

: Of course, this assumes that people think that this is a good idea.  No one,
: other you, has expressed the need for an "end" block during this thread.

Not quite true....

I dislike them too, although I am a devoted Python user.
Also, don't forget that the people who have dropped Python because of the
indentation are not posting here...

But again: Python is still a great language. Nothing is perfect.

Cheers,

---
Thomas Hamelryck
Institute of Molecular and Structural Biology
Aarhus University
Gustav Wieds Vej 10C
DK-8000 Aarhus C
Denmark                                                                                            


From mwh21 at cam.ac.uk  Wed Jan 26 14:18:56 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 26 Jan 2000 19:18:56 +0000
Subject: In defence of the two-namespace rule
References: <000a01bf670e$cce49a00$632d153f@tim>
Message-ID: <m37lgwwsin.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim_one at email.msn.com> writes:

> > Of course, since the timbot is never wrong, this must have
> > been a test!  Do I pass, huh, huh?
> 
> Wrong conclusion, alas.  I (Tim) share an email address with the timbot, and
> that's how you can tell when I'm posting:  full of errurs an dumb misteakz.
> The timbot is temporarily offline to enhance its meanness chip.

Is this why your posts have started to thread properly?

USENET-and-gnus-are-two-steps-on-the-path-to-confusion-ly y'rs
Michael


From grant at nowhere.  Thu Jan 20 08:52:53 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 20 Jan 2000 13:52:53 GMT
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse> <38862948.5AEE964C@roguewave.com>
Message-ID: <slrn88e4pl.67b.grant@grante.comtrol.com>

In article <38862948.5AEE964C at roguewave.com>, Bjorn Pettersen wrote:
>Evguenii Smogailov wrote:
>
>> On 19 Jan 2000, Stephen Coursen wrote:
>> > When did this become a trait of Perl?  Gods, what's so hard about knowing
>> > that == is the equality operator?
>>
>> Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.
>
>I see... so you don't really want eq, gt, and ne; you want equal, greater, and
>not equal.

No, no, that would be too hard to read.  We really need the
operators 'is_greater_than' 'is_less_than' 'is_equal_to'.

Ideally, the underscores should be spaces, but that might be
problematic.  

As long as we're at, I think we should try to make Python code
not only more readable but also more polite. Rather than yet
another imperitive language, the world needs more polite
request languages!

I'm proposing the mandatory use of the keywords 'please',
'excuse_me', 'I'm_sorry' and 'thank_you'.  'Please' will be
used in most of the places where ':' is currently found, and
'thank_you' will be required syntax in the code path were an
exception doesn't occur, or an the end of any block.  'I'm
sorry' will be required in all exception handling code, and
'excuse_me' will be semantically meaningless, but programs will
be required to have at least 5% of the keywords to be
'excuse_me'

excuse_me, try please
  if A is_greater_than B please 
    return A
  else please
    return B
  thank_you
except please
  print "exception occurred"
  I'm_sorry
thank_you

For non-english speaking localities (or areas with high
concentrations of Modula-3 programmers), the same key words
will be used, but we'll just type the louder, slower, and with
exagerated pronunciation:

E_X_C_U_S_E__M_E, try P_L_E_A_S_E
  if A IS_GREATER_THAN B   P_L_E_A_S_E 
    RETURN A
  else  P_L_E_A_S_E
    return B
  T_H_A_N_K__Y_O_U
except P_L_E_A_S_E
  print "exception occurred"
  I'M_S_O_R_R_Y
T_H_A_N_K__Y_O_U

This latter option is based on the fact that everybody
understands English if you speak slowly and loudly (of course
using a bit of an accent to try to sound like the locals who
_do_ speak English is also recommended).

-- 
Grant Edwards                   grante             Yow!  RHAPSODY in Glue!
                                  at               
                               visi.com            


From wweston at worldnet.att.net  Thu Jan 13 12:27:57 2000
From: wweston at worldnet.att.net (wes weston)
Date: Thu, 13 Jan 2000 17:27:57 +0000
Subject: can't find tk.tcl
References: <387BB7D8.F1888924@worldnet.att.net> <387D4966.7BC156E6@charm.net>
Message-ID: <387E0B1C.3FB836F6@worldnet.att.net>

John,
    I think you have it right. That's what I had done; removed an old tcl install.
The only probem is which files. I looked in system and system32 and didn't see any
files starting with tc... or tk...
                wes

John Zouck wrote:

> I had a problem that sounds like this several weeks ago, and finally found that
> there were some tcl/tk .dll files in /windows/system that , when deleted, made
> things work right. Before I found this, I tried every other kind of clean-up and
> reinstall of python to no avail. I had installed tcl/tk months (or years) before,
> which probably caused these files to be there. I don't know why it worked, but it
> did, and this is just a suggestion.
>
> wes weston wrote:
>
> > Guys,
> >         At the command line prompt: "from Tkinter import *" then "w = Tk()" and
> > a "usable" tk.tcl can't be found in the several places looked at. But,
> > it's in one of them: C:\Progra~1\Tcl\lib\tk8.0.  The weirdest thing is
> > I've been running programs that use Tkinter recently on this
> > machine(WinNT Ws). It (Tkinter programs) works at work and on my
> > portable.
> >
> >         I tried a fresh install in several different places. Help!
> >
> >                 wes
>
> --
> ===================
> John H. Zouck
> jzouck at charm.net
> ===================



From haje01 at netsgo.com  Thu Jan 27 15:32:54 2000
From: haje01 at netsgo.com (Kim Jeong Ju)
Date: Fri, 28 Jan 2000 05:32:54 +0900
Subject: Echoing evaluation result in embeded Python?
Message-ID: <t%1k4.19$nR3.453@news.hananet.net>

When you run Python in console mode, you can see result string after
evaluation.

>>> 3 + 4
7
>>> dir
<built-in function dir>

But Embeded Python doesn't show its result string anymore..
You should specify 'print' command to see the result as in consol mode.
>>> 3 + 4
>>> dir
>>> print 3 + 4
7
>>> print dir
<built-in function dir>

Why this behavior happens?
Are there any way to let Python echoing result automatically in embeded
version?

Thanks.






From mgushee at havenrock.com  Sat Jan  8 10:47:37 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 08 Jan 2000 10:47:37 -0500
Subject: Looking for threading tutorial/HOWTO
Message-ID: <m3zougva3q.fsf@havenrock.com>

Hi, Folks--

I would like to learn how to program threads in Python. Or maybe I
should say, I'd like to use threads in a Python program, so I need to
learn how they work. Can anyone recommend a good online 
introduction to the subject (I like books but can't afford to buy one
just now)? Code examples might be good, too, if they are well commented.

Thanks!
-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From pj at sgi.com  Mon Jan 24 16:51:16 2000
From: pj at sgi.com (Paul Jackson)
Date: 24 Jan 2000 21:51:16 GMT
Subject: Python Rocks!
References: <000d01bf6312$77edd660$31a2143f@tim> <388BB4E1.974E63B7@prescod.net>
Message-ID: <86ihgk$tdiu3@fido.engr.sgi.com>

|> Wouldn't it be easier if we just added an optional, (but
|> compiler-recognized if present) block delimiter to Python 2?

argh.  I trust Guido won't do this.  A key selling
point of Python, in my view, is that it encourages
a uniformly plain and readable coding style.

If you belong to the "There is always more than one
way to do something" school of thought, try Perl.

-- 

=======================================================================
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 grant at nowhere.  Tue Jan 18 09:48:05 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 18 Jan 2000 14:48:05 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <034a01bf6154$dba16d80$01ffffc0@worldnet.att.net>
Message-ID: <slrn888v95.31r.grant@grante.comtrol.com>

In article <034a01bf6154$dba16d80$01ffffc0 at worldnet.att.net>, Emile van Sebille wrote:

>If you really want to write a program to pull the mail, then targeting
>the web interface probably makes as much sense as anything, but if what
>you want is to get your mail, why not use a rule or the out-of-office
>thingy to bounce your mail off the 'net to a *real* mailbox?

That's what I used to do, but there are two problems with that:

 1) MickySoft Exchange f@#ks up the headers on bounced mail so
    that you can't "reply" to it, and in some cases you can't
    even tell what the e-mail address of the sender was.

 2) The system admin shut that feature off to try to force
    everybody to use Outlook.

I've also considered setting up a rule to save mail in a local
mailbox that's on an SMB mountable filesystem and writing a
program to periodically suck mail out of that mailbox.  I've no
idea what the format of MS Outlook mailboxes is...

-- 
Grant Edwards                   grante             Yow!  Then, it's off to
                                  at               RED CHINA!!
                               visi.com            


From skip at mojam.com  Tue Jan 25 11:07:44 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 25 Jan 2000 10:07:44 -0600 (CST)
Subject: tabs do WHAT?
In-Reply-To: <86kgh4$s2v$1@mach.vub.ac.be>
References: <000301bf6706$99fca7c0$632d153f@tim>
	<86joac$l2s$1@mach.vub.ac.be>
	<86k78t$thg@newton.cc.rl.ac.uk>
	<86kgh4$s2v$1@mach.vub.ac.be>
Message-ID: <14477.51792.750927.480820@beluga.mojam.com>

    Thomas> C++ is a monster because of its origin (comptibility with C),not
    Thomas> because of the way it evolved. The evolution of python is quite
    Thomas> similar as far as I can see : people try out new stuff
    Thomas> (e.g. Tismer's stackless python), some of it ends up in the
    Thomas> official python distribution.

Christian Tismer's stackless python is more akin to someone modifying the
code generator for C's switch statement or porting C to a new instruction
set architecture than modifying the language programmers see.  Lots of stuff
happens under the covers that has nothing to do with the language definition
(John Skaller will object to my use of the term "language definition" ;-).
Deciding to admit visible block delimiters would be such a change.

Perhaps we should assemble a list of all the deja.com or egroups.com URLs
that pertain to Python's indentation style and stick 'em somewhere easy to
get at.  Then we can print it out and bludgeon people with the resulting
ream of paper when they raise the question (and, of course, add their URL to
the list for the next time)...

who-needs-trees?-ly y'rs,

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From aahz at netcom.com  Sat Jan 22 10:28:18 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Jan 2000 15:28:18 GMT
Subject: In defence of the two-namespace rule
References: <000a01bf646e$6b060500$caa0143f@tim>
Message-ID: <86ciai$7cv$1@nntp4.atl.mindspring.net>

In article <000a01bf646e$6b060500$caa0143f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>pondering-what-"guidoing"-will-mean-in-2037-ly y'rs  - tim

It will mean that the infamous time machine fixed the problem in 2003.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From thomas at bibsyst.no  Thu Jan 20 09:00:22 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Thu, 20 Jan 2000 15:00:22 +0100
Subject: Using glade to build guis
References: <38845051.5A9A1A49@bibsyst.no> <p6q3drsdi4j.fsf@informatik.hu-berlin.de>
Message-ID: <388714F6.3B0BE7B4@bibsyst.no>

Martin von Loewis wrote:
> 
> Thomas Weholt <thomas at bibsyst.no> writes:
> 
> > I`ve tried to use Glade to build guis and run them with the glade.py
> > module/script. What I don`t understand is how to make it react to
> > keypressing or other events. Does a manual/tutorial/example exists on
> > how to prototype a gui in glade then code events-handling and the
> > functional parts of the gui later ??
> 
> Documentation is largely missing, it appears. To connect to signals,
> you first have to declare a signal handler in glade. You can then
> connect the signal handler with a specific function. To do so, you can
> either use .connect(signalname, function), or .auto_connect(dictionary
> of signal/handler pairs).
> 
> Hope this helps,
> Martin

Yes, but do you have a small example to go with that ??

Thomas


From felixt at dicksonstreet.com  Sat Jan 22 21:48:09 2000
From: felixt at dicksonstreet.com (Felix Thibault)
Date: Sat, 22 Jan 2000 20:48:09 -0600
Subject: Suggestions for python 2
In-Reply-To: <m3ya9hvj46.fsf@atrus.jesus.cam.ac.uk>
References: <000501bf623e$18b56a20$9a2d153f@tim>
 <3.0.5.32.20000122162257.009af7f0@mail.dicksonstreet.com>
Message-ID: <3.0.5.32.20000122204809.009acae0@mail.dicksonstreet.com>

At 22:26 1/22/00 +0000, Michael Hudson wrote:
>Felix Thibault <felixt at dicksonstreet.com> writes:
>
>> How hard would it be to implement something like:
>> 
>> def like_a_list(*now_args):
>>     def return_func(!now_args, *later_args, **show):
>>         if show:
>>             return list(now_args+later_args)
>>         else:
>>             return apply(like_a_list, now_args + later_args)
>>     return return_func
>> 
>> where the ! (or whatever) means something like this happens:
>> 
>> def like_a_list(*now_args):
>>     """like_a_list(value0, value1,...) store some values in a function"""
>>     fun_str = """def return_func(*later_args, **show):
>> 	now_args = %s
>> 	if show:
>> 		return list(now_args+later_args)
>> 	else:
>> 		return apply(like_a_list, now_args + later_args)\n"""\
>> 		 % (now_args,)
>>     exec(fun_str)
>>     return return_func
>> 
>> when the inner function is defined ? (Except that it works when
>> eval(repr(object)) doesn't and it's not as ugly). 
>
>I don't really understand what you are trying to do, but you are
>invited to check out xapply.py in my bytecodehacks package which can
>be found at
>
>http://starship.python.net/crew/mwh/
>
>and is documented here:
>
>http://starship.python.net/crew/mwh/bch/module-bytecodehacks.xapply.html
>

Thanks! I'm reading it now...

>If that isn't what you're after, can you post an example of what you
>might want to do with this hypothetical function? I can probably whip
>something up for you.

All I was really trying to do was come up with an example of a function that 
would need access to the enclosing scope and also would need *args so it
couldn't just use default argument values, but here's what I did when I tested
it:

>>> from monkeybiz import like_a_list
>>> lal123 = like_a_list(1,2,3)
>>> lal123(show=1)
[1, 2, 3]
>>> more_lal = lal123(4, 5, 6, show=1)
>>> more_lal
[1, 2, 3, 4, 5, 6]

<take some values and store them in a function that takes some more values
and checks for **kwargs, if there are any it returns its values, otherwise
it stores them in a function like itself>  

I guess what I was wondering was if there could be a syntax for faking nested
scopes like what we have with default argument passing now, except that
whatever gets passed in can't be overwritten when the function is called,
so you can use *args, or whatever, too. That's what the '!' in the first
example and the 'exec(.... % args)' that emulated it was supposed to do.
But since "def takes a name, some more names and some code with those names
and ties 'em all together" sums up what I know about what's going on...all
I really know is I'm in full agreement with:

>> >>[Evan Simpson]
>> >>>
>> >>> I rather like the idea of having to explicitly bring names
>> >>> into an inner scope,
>> >>


as-long-as-we-don't-get-real-nested-scopes-I-can-muddle-through'ly yours,

Felix

>
>Cheers,
>Michael
>-- 
>http://www.python.org/mailman/listinfo/python-list
>
>




From fredrik at pythonware.com  Fri Jan 21 07:10:24 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 13:10:24 +0100
Subject: Tkinter coverage in books? [Was: Re: Python and Tkinter Programming by John E. Grayson
References: <3880792E.79A874B6@weihenstephan.org> <869fo6$jcs$1@srv38.cas.org>
Message-ID: <002d01bf6408$80654660$f29b12c2@secret.pythonware.com>

lvirden at cas.org wrote:
> Other than Programming Python and this new reference, what other books
> provide the Python/Tkinter programmer substantial information?  That is
> to say, does more than a few sentence mention that 'oh, by the way, there's
> this extension that a number of Pythonians use...' 

http://www.pythonware.com/people/fredrik/tkintrobook.htm

</F>




From pds at uberhacker.org  Fri Jan 28 13:27:41 2000
From: pds at uberhacker.org (Paul Schmidt)
Date: 28 Jan 2000 18:27:41 GMT
Subject: socket.recv() question
Message-ID: <slrn893nst.15sv.pds@uberhacker.org>

Hello all,
      I am trying to parse data from an rwhois server, and it sends back
lines terminated by \012.  whenever I do a:

      results = sock.recv(1024)

I only get the first line in results, but I want the entire output, so
I can split it by lines and do other parsing things with the data.

How would I go about doing this?

Thanks,
-Paul

--
[Paul D. Schmidt....................][..................pds at enteract.com]

   The grass may actually be greener on the other side of the fence, but it 
   still has to be mowed.  --Anonymous


From cjc26 at nospam.cornell.edu  Sun Jan 23 18:36:06 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: 23 Jan 2000 23:36:06 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de> <869uel$suj$1@nntp1.atl.mindspring.net> <slrn88hnpb.brp.quinn@auto.ugcs.caltech.edu>
Message-ID: <slrn88n40a.2qh6.cjc26@bridget.mindriot.net>

Pada 21 Jan 2000 22:33:56 GMT, Quinn Dunkan bilang:
| 
| A curried function always takes one argument.  If it needs more values to
| 'finish', it returns a closure which will accept the next arg.

My semantics prof always made us call currying "Schoenfinkelization",
because "Schoenfinkel came up with it first".
Do you know how hard it is to spell Schoenfinkelization? ;)


-- 
cliff crawford  -><-  http://www.people.cornell.edu/pages/cjc26/
                     "All I need is a small town
                      with good Internet connectivity" - Nadine


From jjlucsy at concentric.net  Wed Jan 26 17:35:16 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Wed, 26 Jan 2000 17:35:16 -0500
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net>
Message-ID: <nDKj4.747$UH3.67876@news4.usenetserver.com>

I have a version of python15.dll that I've compiled using the delayload
mechanism. Basically this means that it will only load a dll when it needs
it. Running a very basic test on my system (400Mhz celeron,128Mb,Win2K RC3)
the load time went from over 2 seconds to under half a second. If you wish I
can email it to you, or if someone wants, I'll upload it somewhere.

- Joel Lucsy (jjlucsy at concentric.net)






From nielsenjf at my-deja.com  Thu Jan  6 17:40:59 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Thu, 06 Jan 2000 22:40:59 GMT
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com> <3874F5DA.B73A6039@starmedia.net>
Message-ID: <8535lq$ikj$1@nnrp1.deja.com>

Thanks that works great.

john


In article <3874F5DA.B73A6039 at starmedia.net>,
  drew_csillag at geocities.com wrote:
> Darrell wrote:
> >
> > from filename import *
>
> Actually, that doesn't do it either if the other module wants to
> reference vars in the importing namespace.
>
> The way to make it so that the other code can access variables in the
> including files namespace is:
> exec open('filename').read()
>


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


From felixt at dicksonstreet.com  Sun Jan 30 05:24:44 2000
From: felixt at dicksonstreet.com (Felix Thibault)
Date: Sun, 30 Jan 2000 04:24:44 -0600
Subject: Creating a local variable with a parameterized name
In-Reply-To: <38937075.18B75753@rochester.rr.com>
Message-ID: <3.0.5.32.20000130042444.009e5100@mail.dicksonstreet.com>

At 17:57 1/29/00 -0500, Michael Ross wrote:
>I'm new to this, so I hope I'm not missing something obvious...
>
>I want to create a local variable with a passed in name.
>
>I realize I can do the following:
>
>>>> def f(varname):
>             cmd = '%s = 123' % varname
>             exec(cmd)
>             cmd = 'print %s' % varname
>             exec(cmd)
>
>>>> f('mike')
>123
>>>>
>
>But exec needs to invoke the parser, which seems to be an extremely
>costly thing to do.
>
>It seems local variables are stored in a special, inaccesible (from
>python) place in the frame object (I've been looking at
>PyFrame_FastToLocals and PyFrame_LocalsToFast). Do I need to write C
>code to do this?  A builtin to do this would be nice...
>
>Thanks in advance!
>Mike Ross
>
>-- 
>http://www.python.org/mailman/listinfo/python-list
>
>

Would something like one of these work?

>>> class Warehouse:
	def __init__(self, name, value):
		setattr(self, name, value)
	def __call__(self, name):
		return getattr(self, name)
		
		
>>> w = Warehouse('Mike', 123)
>>> w('Mike')
123

>>> class Empty:
	pass
>>> w = Empty()
>>> w.Mike = 123





From alex at mindlube.com  Mon Jan  3 03:01:01 2000
From: alex at mindlube.com (Alex Rice)
Date: Mon, 03 Jan 2000 01:01:01 -0700
Subject: doh! (was: cannot install wxPython on linux)
References: <387053D9.4EB9C9B8@mindlube.com>
Message-ID: <3870573D.D8AF29D@mindlube.com>

Alex Rice wrote:

> wxGTK-2.0.1-1
> wxPython-2.1.11-1

So I thought to myself-- hmm, what if I install wxGTK-2.1 instead of 2.0. 
That seems to have fixed it right up.


-- 
Alex Rice <alex at mindlube.com>


From drew.csillag at starmedia.net  Mon Jan 24 10:47:23 2000
From: drew.csillag at starmedia.net (Andrew Csillag)
Date: Mon, 24 Jan 2000 10:47:23 -0500
Subject: stupid FORM problem
References: <86hq3u$gmu$1@nnrp1.deja.com>
Message-ID: <388C740B.C2873645@starmedia.net>

sp00fD wrote:
> 
> I'm having a problem with a cgi program.  Basically some html calls a
> <FORM METHOD="POST" ACTION="/my.cgi?action=view&table=thistable">
> ...
> <SELECT NAME="thread">
> <OPTION VALUE="1">foo
> ....
> 
> None of my params {action : "view", table : "thistable", thread : 3(or
> whatever)} go through.  I've tried this using GET as the METHOD as
> well, but no luck.  Here's the thing though, if I type the path in as
> the url, it works (i.e http://myhost/my.cgi?
> action=view&table=thistable&thread=3)
> 
> What am I doing wrong.
> 

I don't think you can have the ?action... part as the ACTION url to
FORM.
Try putting these in your <FORM> block instead:
<INPUT TYPE=hidden NAME=action value="view">
<INPUT TYPE=hidden NAME=table value="thistable">


-- 
print(lambda(q,p):'pmt:$%0.2f'%(q*p/(p-1)))((lambda(a,r,n),t:(a*r/
t,pow(1+r/t,n*12)))(map(input,('amt:','%rate:','years:')),1200.0))



From niels at endea.demon.nl  Tue Jan 18 10:07:14 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Tue, 18 Jan 2000 16:07:14 +0100
Subject: Tkinter hangs again under Windows
References: <85v092$3n4$1@nnrp1.deja.com> <948188619.449404@cdwork.cvt.stuba.sk>
Message-ID: <388481A2.D233D165@endea.demon.nl>


Radovan Garabik schreef:
> class App(Frame):
>     def __init__(self, master=None):
...
>         if not have_gnuplot:
>             tkMessageBox.showwarning("Gnuplot missing", "Gnuplot.py not found, falling back to tkinter")

This seems to be the problem. If you show any dialogs before entering
mainloop, you won't receive events anymore.

-- 
Niels Diepeveen
Endea automatisering



From hartmut at oberon.noris.de  Tue Jan  4 12:35:07 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Tue, 04 Jan 2000 18:35:07 +0100
Subject: Pseudo-SQL: searching for Parser/Rewriter
References: <16600378@oberon.noris.de> <slrn86uroa.fph.neelk@brick.cswv.com>
Message-ID: <16600382@oberon.noris.de>

Neel Krishnaswami (neelk at brick.cswv.com) schrieb:

 >Try John Aycock's SPARK, the Scanning, Parsing, and Rewriting Kit:
 >
 >   <URI:http://www.csr.UVic.CA/~aycock/python/content.html>

Looks really great!

I'll use it even if it will be a bit to flexible :-)

 >Neel

+++hartmut


From tye4 at yahoo.com  Wed Jan 19 14:31:36 2000
From: tye4 at yahoo.com (tye4)
Date: Wed, 19 Jan 2000 11:31:36 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com>
Message-ID: <8653jg$esi@news.or.intel.com>

Ivan Van Laningham <ivanlan at callware.com> wrote in message
news:388521FB.BCC45087 at callware.com...
> Hi All--
>
> Aahz Maruch wrote:
> >
> > In article <8631qm$26p at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
> > >
> > >How hard can it be to add 'end' indicators to the language.. If these
end
> > >indicators are added, the language would still be backward compatible.
I
> > >can't believe nobody has fixed this for so long.
> >
> > Are you suggesting the addition of 'end' indicators IN ADDITION to the
> > indentation?
> >
>
> Yes, he is.  tye4, wake up and smell the tabs.  Most people who use
> Python like the indentation the way it is, and feel no need for the
> unnecessary clutter your hack would produce.  Python is not Perl and we
> don't want it to be.

You both are right. But that is not the whole story.
Say the 'end' thingies are added to Python. Now you have two types of Python
modules: ones with no 'end' (old legacy code) and others with 'end' (new
code).
A newer version of Python would compile both versions since it is backward
compatible (think of the end indicators as optional.)
As the new compiler supporting 'end' blocks comes out, the old style of
coding without 'end' will be declared as _deprecated_. As time goes on, the
code using 'end' will increase substantially while the legacy code without
'end' will reduce.

At that point, it will be safe to eliminate the _strict_ indentation rules
from the language. Without backward compatibility, all the people used the
old style will lose all their work if the language is changed.

-tye4






From kc5tja at garnet.armored.net  Sun Jan  9 20:36:20 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 10 Jan 2000 01:36:20 GMT
Subject: Python Host
References: <bJ3e4.8$78.764@news.uswest.net>
Message-ID: <slrn87ie93.sb7.kc5tja@garnet.armored.net>

In article <bJ3e4.8$78.764 at news.uswest.net>, Andrew Ferk wrote:
>Are there any web hosts that support Python?

Armored Internet, the ISP I work for, supports Python 1.5.2 on our Linux and
NT web servers.  We also support Zope on our Linux web servers.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From tjg at avalongroup.net  Sun Jan 23 08:45:47 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Sun, 23 Jan 2000 05:45:47 -0800
Subject: Tkinter and wxPython
Message-ID: <388B060B.D13F7B64@exceptionalminds.com>

OK, After just spending three weeks learning Tkinter on-the-fly, I have
to ask this question, what advantages/disadvantages are there to each
Tkinter and wxPython?

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<



From tye4 at yahoo.com  Fri Jan 21 16:38:40 2000
From: tye4 at yahoo.com (tye4)
Date: Fri, 21 Jan 2000 13:38:40 -0800
Subject: Python Rocks!
References: <200001211334.IAA28640@python.org>
Message-ID: <86ajlq$661@news.or.intel.com>

Yes, I think I'll let you guys wallow in your ignorance.

-tye4

PSU boss <secret at email.com> wrote in message
news:200001211334.IAA28640 at python.org...
> Martijn Faassen wrote on 948370683:
> > Someone get the Python Secret Underground to do something about him. :)
>
> He won't hurt us anymore.
>
> regards,
> PSU boss.
>
>




From nascheme at enme.ucalgary.ca  Sat Jan 22 15:48:19 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Sat, 22 Jan 2000 20:48:19 GMT
Subject: Python dumps core with this.. how come?
References: <38891316.393423@news.erols.com> <005301bf64b5$d2f234d0$f29b12c2@secret.pythonware.com> <3889c4b6.242514@news.erols.com>
Message-ID: <nIoi4.9746$up4.227638@news1.rdc1.ab.home.com>

Roey Katz <katz at glue.umd.edu> wrote:

>...why does this happen:
>
>     >>> a = 3 
>     >>> b = a 
>     >>> b
>     3 
>     >>> a = 5
>     >>> b
>     3

Let me try some ASCII art:

    a = 3
    b = a

gives you this (conceptually):

    +---+       +---+
    | a |------>| 3 |
    +---+    /  +---+
            /
    +---+  /
    | b |-/
    +---+

When you do:

    a = 5

now you have this:

                +---+
              ->| 5 |
             /  +---+
            /
    +---+  /    +---+
    | a |-/   ->| 3 |
    +---+    /  +---+
            /
    +---+  /
    | b |-/
    +---+

Variables do not contain values.  They refer to them.  In Python
everything is a reference.  Try using the id() function to see
what are the individual objects.

>I mean, what happens if I have a complicated path 
>to a variable and I want to alias it:
>
>    alias = long_name.hard_to_type[ 1 ].integer_value
>
>    # arbitrary calculations that would be 
>    # torturous with any label longer than 'alias'
>    if alias > someValue:
>       result = (alias * 5 ) / (alias + 4 )
>    else:
>       result = (alias * 3 ) / (alias + 2 )

In Python you need one more step:

     long_name.hard_to_type[ 1 ].integer_value = alias

When there is a tradoff between convience and explicitness,
Python chooses explicitness.  If you have a name that ugly you
are probably doing something wrong already.
    
>I understand that if I have a function and pass to it a parameter,
>I can modify it in-place (although that I cannot modify numbers,
>strings or tuples in-place seems to me inconsistent).  

You can modify objects if they are "mutable".  Integers, strings
and tuples are "immutable" for a reason.  Just use Python some
more and think about why it is the way it is.  The FAQ has some
more details if you are interested.

>So for example, these are roughly equivalent:
> 
>    void setValue(  int& x ) { x = 3  }    // C++
>    void setValue(  int *x  ) { *x = 3 }    // C
>    def  setValue( x ):    x = 3             # Python
>
>whoops, we cannot modfiy integers in-place, how invonvenient.

Python is not C or C++.  You must change your mindset.

>ok, here's a third question, and it builds on the above inquiry
>regarding a lack of an assignment operator.  What happens when I want
>to do convert this for loop:
>
>   for i,j in dest, src:
>        dest[ i ] = src[ j ]
>
>into a fast map() expression:
>
>   map( dest.assign_operator, src )

Who says that map() is faster?  It certainly is not any more
readable.  Why do you want to do such a thing?  Premature
optimization is the root of all evil (in programming).

Besides, if you want to copy the sequence why don't you just use
copy.copy()?

>I cannot use  filter() to narrow down the list to only the elements
>that I want and then copy it into another list with src.copy().  

Sorry, I don't understand what the problem is here:

    new_list = filter(my_predicate, my_list)

Why would you want to use copy()?  Why do you need an assigment
operator?


    Neil


From PClaerhout at CREO.BE  Fri Jan 21 06:49:00 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Fri, 21 Jan 2000 12:49:00 +0100
Subject: Screen depth
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA707A@MSGEURO1>

I'm working on Windows NT and 95/98.

Pieter

-----Original Message-----
From: Fredrik Lundh [mailto:fredrik at pythonware.com]
Sent: Friday, January 21, 2000 1:04 PM
To: Pieter Claerhout
Cc: Python Mailing List
Subject: Re: Screen depth


Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> how can I determine how many colors a screen can show?

it depends on the platform and what user interface
toolkit you are using.

</F>


-- 
http://www.python.org/mailman/listinfo/python-list



From josht at iname.com  Sat Jan 15 09:42:51 2000
From: josht at iname.com (Josh Tompkins)
Date: Sat, 15 Jan 2000 14:42:51 GMT
Subject: Nested while-s
References: <000301bf5f11$48fa28e0$b92d153f@tim>
Message-ID: <8EBC6C5A8joshtinamecom@209.208.0.7>

tim_one at email.msn.com (Tim Peters) wrote in 
<000301bf5f11$48fa28e0$b92d153f at tim>:

>Note that you never reset currentY to ypos-radius, so the inner loop won't
>be entered on any but the first iteration of the outer loop.  You probably
>want to change
>
>    currentY = (ypos - radius)
>    while currentX <= (xpos + radius):
>
>to
>
>    while currentX <= (xpos + radius):
>        currentY = (ypos - radius)
>
>(that is, just swap those two lines).
>
>feel-free-to-kick-yourself-now<wink>-ly y'rs  - tim
>
>
>
Ouch!  I guess that's what I get for trying to write this stuff on <enter 
small value here> hours of sleep.  <grin>  Thanks for the help.

Josh "Josue" Tompkins
________________________________________________________________

"Destined For Great Things -- but pacing myself."
- From a t-shirt.

E-Mail:  josht at iname.com 
ICQ:  21219667 
Web:  http://www.crosswinds.net/~josht
_________________________________________________________________


From cribbsjg at earthlink.net  Fri Jan 21 13:25:08 2000
From: cribbsjg at earthlink.net (Jamey Cribbs)
Date: Fri, 21 Jan 2000 18:25:08 GMT
Subject: Executing X-Windows app
Message-ID: <3888A53C.7B4CE735@earthlink.net>

Hi, I hope this doesn't end up being a stupid question.  I am trying to start an X app from within a Python script.

If I use os.system, my system freezes until I CTRL-C.  If I run os.system from interactive Python, it works ok.  I have tried popen, but it does the same thing.

What am I missing?  I am sure it is something very basic.  I have looked throught the ref manual, but cannot seem to find the answer.

By the way, I am running on top of RH Linux 6.1 with Python 1.52.

Any help would be greatly appreciated.

--
Jamey Cribbs
Sr. Technical Consultant


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000121/e9271e95/attachment.html>

From echuck at usit.net  Wed Jan  5 20:04:41 2000
From: echuck at usit.net (Chuck Esterbrook)
Date: Wed, 05 Jan 2000 20:04:41 -0500
Subject: trying again: CgiHttpServer workalike for Win/NT and Win/98?
References: <837o38$b65$1@serv1.iunet.it> <83b4a5$d5r$1@nntp3.atl.mindspring.net>
Message-ID: <3873EA29.C99706D2@usit.net>

A few weeks ago, I was in the exact situation that Alex was is in.

I found Medusa difficult to get along with for lack of documentation. And
CGIHTTPServer difficult because I'm using Windows 95.

I ended up downloading Apache (http://www.apache.org) and I'm quite happy with it.
It supports having "#!python" in your CGI scripts and everything works fine.

Regardless of what you use to serve your HTTP, I recommend you check out this
article. I found it quite useful and have cooked up my own version of the wrapper
script with better error reporting:

 http://www.webtechniques.com/archives/1998/02/kuchling/

My wrapper script also assumes a ".py" extension for the target script which makes
my URLs a bit cleaner.

-Chuck


Aahz Maruch wrote:

> In article <837o38$b65$1 at serv1.iunet.it>,
> Alex Martelli <alex at magenta.com> wrote:
> >
> >To test some CGI scripts with minimal hassle, I would like to
> >be able to run CgiHttpServer.py, or something similar to it,
> >on some Windows/NT and Windows/98 PCs.  Unfortunately
> >for this purpose, it seems that CgiHttpServer.py itself is oriented
> >to Unix (e.g., it wants to fork to run the script).
>
> The /F-bot gave you an answer, but it needs a bit of fleshing: unless
> you specifically need to run CGI scripts, you may be better off running
> them as Python modules underneath Medusa (which would take some
> rewriting on your part); this would make OSes completely transparent.
> --
>                       --- Aahz (@netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>     http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6
>
> TEOTWAWKI -- 16 days and counting!



From aahz at netcom.com  Sat Jan 22 21:40:54 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jan 2000 02:40:54 GMT
Subject: Python online docs are great!
References: <slrn88kouc.pej.nobody@isis.visi.com>
Message-ID: <86dpnm$dtg$1@nntp9.atl.mindspring.net>

In article <slrn88kouc.pej.nobody at isis.visi.com>,
Grant Edwards <nobody at nowhere.nohow> wrote:
>
>I just want to compliment GvR on the Python online documentation files.
>They are amazingly well done. They are without a doubt the best online
>documentation I've used for anything, anywhere, ever.  Often online
>documentation has been chopped up into pages that are way too small,
>and there's no useful index or navigation aids.  The result is you
>spend all your time thrashing around because the needed information
>is spread over a half-dozen different pages.  I've found the Python
>documents very well organized, comprehensive, and easy to use.

Fred Drake deserves much of the credit here. I can say from personal
experience that the docs have improved greatly over just the last year.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From kc5tja at garnet.armored.net  Fri Jan 14 16:54:45 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 14 Jan 2000 21:54:45 GMT
Subject: XML-RPC Server
References: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de> <slrn87ne3i.qla.kc5tja@garnet.armored.net> <xqxr9fkk94u.fsf@skewer.ca.boeing.com>
Message-ID: <slrn87v75j.gt3.kc5tja@garnet.armored.net>

In article <xqxr9fkk94u.fsf at skewer.ca.boeing.com>, Harry G. George wrote:
>What is the python community's position re SOAP and XML-RPC?  As I
>understand it, XML-RPC is standards-base, platform-independent, while
>SOAP uses MS-sponsored extensions (including, e.g., XML Data).  Can I
>assume that *both* will be supported in the future?

Having a copy of the XMLRPC specification right here, I can say that XML-RPC
doesn't reference a single standards document.  OK, the timestamps are in
ISO8601 format.  But everything else is anything but standards based.  In
fact, ALL of XML-RPC seems to be entirely Dave Winer's creation.  I've never
seen any third-party recommendation be included into XML-RPC except for the
Base64 datatype.

SOAP, in contrast, IS based on existing open standards.  Just look at its
bibliography.  Also, XML-Data is not a Microsoft "extension" -- it's
something the W3C organization has been working on for quite a long time.
Just because Microsoft had some input in the standards creation process does
mean it's suddenly a Microsoft extension.  I remember hearing about a desire
to replace the SGML DTD notation with XML back when XML was first
introduced.  And now, they're working on its successor: XML-Schemas.  I have
the specification for XML-Schemas.  I can see **nothing** in it that, in any
way, shape, or form, that makes me believe that Microsoft had anything to do
with the standards approval process -- there is nothing in EITHER
specification which makes them implicitly or explicitly Windows-specific.

For the record, that which is now SOAP happens to be the ORIGINAL XML-RPC
specification, developed long before Microsoft even got involved, and
developed by the same people who designed the modern XML-RPC standard.  The
ONLY reason XML-RPC standard exists today is because Microsoft wanted to
make sure everything was correct before releasing the standard, which would
have blown Dave Winer's schedule.  He released XML-RPC as a quick kludge
(and it shows!) that was both capable and simple to implement -- thus
guaranteeing Dave's software release schedule.  Are you aware that most of
the software supporting XML-RPC is now also supporting SOAP?  I can see the
market trends already -- XML-RPC will die a slow (but painless) death, and
SOAP will replace it.  XML-RPC has served its purpose; it's time to move on.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From kelly at poverty.bloomington.in.us  Sun Jan 30 20:28:04 2000
From: kelly at poverty.bloomington.in.us (Kelly Lynn Martin)
Date: Sun, 30 Jan 2000 20:28:04 -0500
Subject: Why Perl do But Python can't? 
In-Reply-To: Your message of "Mon, 31 Jan 2000 08:48:09 +0800."
             <872m6v$2pdg$1@news.cz.js.cn> 
Message-ID: <200001310128.UAA02858@poverty.bloomington.in.us>

On Mon, 31 Jan 2000 08:48:09 +0800, "????" <jfzhou at 188.net> said:

>Someone know the difference of Perl and Python in the CGI running
>mode?

Scripts (of any sort) are inherently unsafe when run setuid or setgid.
See the perlsec manpage for why.

Kelly



From herzog at online.de  Thu Jan  6 18:22:53 2000
From: herzog at online.de (Bernhard Herzog)
Date: 07 Jan 2000 00:22:53 +0100
Subject: Reading tar/gz-files
References: <3874DDAA.E14CC23B@bibsyst.no> <p6qzoujouig.fsf@informatik.hu-berlin.de>
Message-ID: <m3u2kqvl82.fsf@greebo.nodomain.de>

Martin von Loewis <loewis at informatik.hu-berlin.de> writes:

> Alternatively, you could try to live without an external tar command,
> and implement that yourself. The gzip module is already there and
> provides you with an uncompressed stream. All you'd need to do is to
> interpret the format of tar yourself. See /usr/include/tar.h for a
> starting point.

Note also, that the Python Imaging Library comes with a TarIO module
that implements at least some of the functionality.


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


From gmcm at hypernet.com  Wed Jan 19 00:00:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 19 Jan 2000 00:00:32 -0500
Subject: **kwargs
In-Reply-To: <86387a$7op$1@nnrp1.deja.com>
Message-ID: <1263864371-2295577@hypernet.com>

sp00fD writes:

> In article <1263886656-955174 at hypernet.com>,
>   gmcm at hypernet.com wrote:
> > sp00fD writes:
> >
> > > my.foo(foo="something", bar="something else", this_dict)
> >
> > Not quite. To call foo with the named args wrapped up as a
> > dictionary, you need to use apply:
> >   apply(foo, ("something", "something else"), this_dict)
> 
> Using the above apply(), how does foo know that foo(should have named
> that something else) == "something" and bar == "something else", they
> appear to be passed as normal args.

Because they're named in the parameter list, they always 
appear as locals (so "foo" and "bar" are always bound to 
something). The Python function-calling machinery takes care 
of it for you. You have to pull stuff out of this_dict.

The fully general (pass me anything) declaration would be
 def foo(*args, **kwargs)

So when I'm trying a new callback with <name a GUI>, the 
first shot will be
 def callback(*args, **kwargs):
   print "callback(%s, %s)" % (repr(args), repr(kwargs))

Beyond that, the **kwargs stuff is of limited use. You might 
look at how Pmw uses it in the __init__ methods of compound 
(Tk based) widgets. But for most cases, it's more work to 
suck stuff out of kwargs than it is to provide named 
parameters.
 
> Also, how is this dictionary represented after the foo method receives
> it?  ie. this_dict["wow"] = 1
> apply(foo, (....), this_dict)
> 
> foo then sees kwargs["wow"] == 1?
> 
> > or
> >   this_dict['foo'] = "something"
> >   this_dict['bar'] = "something else"
> >   apply(foo, (), this_dict)
> >

>>> def foo(arg1=None, **kwargs):
...  print 'arg1=%s, kwargs=%s' % (`arg1`, `kwargs`)
...
>>> foo(1)
arg1=1, kwargs={}
>>> apply(foo, (1,))
arg1=1, kwargs={}
>>> apply(foo, (1,), {'a':2})
arg1=1, kwargs={'a': 2}
>>> apply(foo, (), {'arg1':3,'a':2})
arg1=3, kwargs={'a': 2}
>>>

- Gordon



From s986001 at mailserv.cuhk.edu.hk  Thu Jan 20 13:41:28 2000
From: s986001 at mailserv.cuhk.edu.hk (Kenny)
Date: Fri, 21 Jan 2000 02:41:28 +0800
Subject: stupid question !
Message-ID: <867kic$kb1$1@eng-ser1.erg.cuhk.edu.hk>

Dear all,
     I am now trying to use python for CGI programming!
But I don't know how to wake up the python programming by using HTML!

        <form method=post action=" ....path to the python source code....">
      It seems to me that the program is not be activated!
 Thank you!




From gmcm at hypernet.com  Thu Jan  6 14:16:28 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 6 Jan 2000 14:16:28 -0500
Subject: no return values for __init__ ??
In-Reply-To: <3874E118.1C062D2C@mdlink.de>
Message-ID: <1264936306-6137156@hypernet.com>

Helge Hess writes:

> "Fred L. Drake, Jr." wrote:
> > Helge,
> >   Yes:  It doesn't make sense. 
> 
> Well, IMHO it does make sense, see below.
> 
> > The job of the __init__() method is to
> > initialize the instance which has already been created. 
> 
> Why should we have this distinction ? I find it inconsistent to
> declare '__init__' to behave differently from other methods.
> *This* doesn't make sense to me.
> 
> > There is no
> > way for a Python program to receive the return value.
> 
> In the current interpreter not. That is what annoys me, since it
> is an completly articial limitation I don't understand. As
> mentioned it took only 5 minutes to 'correct' this limitation.

What happens when a derived class chains to the base class 
__init__ and the base class __init__ changes "self"?

How about MI?
 
> >   To implement the singleton pattern you appear to be
> >   interested in,
> > try this:
> 
> [singleton using function]
> >         def C():
> >             global _C_singleton
> 
> This example covers only the most basic case and is a hack (eg
> how do I do 'o instanceof C' ?). 

Make it obvious: rename the factory function to "getC()" and 
rename "_C" to "C".

> The idea behind returning a
> value different from 'self' or 'None' is to customize the class
> based on the parameters.

If really you want to customize the class, you can assign to 
the instance's __class__ attribute.
 
> eg:
>   class ImmutableDictionaryClass: # this is abstract
>     def __init__(self, keys, values):
>       if len(keys) == 0:
>         # always return the same empty dict to save memory
>         if emptyImmutableDict is None: 
>           emptyImmutableDict = EmptyImmutableDict()
>         return emptyImmutableDict
>       elif len(keys) < 10:
>         return SequentialDictionary(keys, values)
>       else:
>         return HashTableDictionary(keys, values)
> 
> This is called a 'class-cluster' in Objective-C and has quite
> some advantages. Another useful application for clusters are
> string-objects (different classes for different char widths, eg
> Py8bitString, PyUnicodeString, ..)

And Fred's pattern is called a "factory" in a wide variety of 
languages.
 
> The nice thing is that you only need a single (abstract) class to
> interact with. Defining a function like you did above seems
> unnecessarily hackish to me.

Hackishness is in the eye of the beholder.
 
> Notably removing the constraint in the interpreter doesn't affect
> existing Python code ! This is what I wonder about. The
> interpreter was apparently restricted intentionally in this
> regard and I wonder what lead to this decision.

Having "self" be a complete object when __init__ runs  means 
the language doesn't have to come up with a bunch of arcane 
rules about default constructors, order in which base class 
constructors are called etc.

It's what makes MI useful in Python (vs a feature that is nearly 
impossible to use, as in some other languages).


- Gordon



From mwh21 at cam.ac.uk  Sat Jan 22 15:34:19 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 22 Jan 2000 20:34:19 +0000
Subject: Looking for Solaris 2.6/Python 1.5.2 make test fix
References: <0001221518.AA16327@cas.org>
Message-ID: <m3k8l1voac.fsf@atrus.jesus.cam.ac.uk>

"Larry W. Virden" <lvirden at cas.org> writes:

> From: Thomas Wouters <thomas at xs4all.net>
> > Something probably went wrong with building your Python binary. Can you
> > start the python executable ? do you get the prompt ? Can you execute code ?
> 
> Yes, I can start the executable and I get the prompt.  I don't know about
> the execute code - as far as I can tell.  Only the one Lib test fails -
> I assume there's more python code than that that executes during make test.

When you cd into Lib/test and run ../../python regrtest.py, do you get
output along the lines of

test_fcntl
test_gdbm
test_gl
test test_gl skipped -- an optional feature could not be imported
test_grp
test_gzip
test_imageop
test test_imageop skipped -- an optional feature could not be imported
test_imgfile
test test_imgfile skipped -- an optional feature could not be imported
test_long

? If so, what is it saying before it blows up?
 
> I typed make distclean; configure; make all.  If anything 'made statically'
> it would be something that configure wanted built that way.  Same for the
> info about 'make dynamically' - I did nothing to the Makefiles - so however
> they were built, that's what got compiled.  As for 'not importing that
> specific module' - I'm a novice to Python, so I don't know what 'module'
> that test is invoking.  Seems to be related to regular expressions (perhaps
> GNU's rx library?)
> 

Hmm. Which compiler? I seem to recall tales of regexp's failing due to
buggy optimizations in some compilers. You might want to try turning
off optimizations or using something like gcc-2.95.

Cheers,
Michael


From mikael at isy.liu.se  Tue Jan 25 11:04:55 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 25 Jan 2000 17:04:55 +0100 (MET)
Subject: tabs do WHAT?
In-Reply-To: <Thjj4.6690$S64.195053@nnrp1.uunet.ca>
Message-ID: <XFMail.000125170455.mikael@isy.liu.se>

On 25-Jan-00 Warren Postma wrote:
 >  Rather bloody minded of me to point this out, but hey "to each his own"...
 >  
 > #define BEGIN {
 > #define END     }
 >  
 >  Can I do the same somehow in Python?  ;-)

Someone pointed this out not long ago:

BEGIN = END = 0

Then use them. However, you cannot indent between them.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    25-Jan-00
Time:    17:03:13

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From sekter at matavnet.hu  Mon Jan 10 16:51:00 2000
From: sekter at matavnet.hu (Arpad Kiss)
Date: Mon, 10 Jan 2000 22:51:00 +0100
Subject: Problem distributing a Gadfly app
References: <85dah2$ip8$1@nnrp1.deja.com>
Message-ID: <387A5444.56A0EAEA@matavnet.hu>

calishar at my-deja.com wrote:
> 
> Hello folks,
> 
>   I am just in the finishing off stage of an application for work that
> uses the gadfly database package. I am trying to build an installable
> application using installer (from Gordon Macmillan?) and running into a
> problem.
> 
>   The application builds alright, but when I install it on the other
> machine, and go to run it, it fails.
> 
>   After re-building with the debug flag on, use runw off, and support
> on, I saw that it is trying to load the sql.mar file from the c: drive.
> Well, it ain't there. If I try to add the sql.mar file to the
> dependencies line of the cfg file, builder fails, if I add it to the
> includes line it fails.
> 
>   Has anyone run into this before?
> 
Hi,
My workaround: 
-i have added a carchive section to the cfg file. This carchive contains
the sql.mar file.
-i have added the sqlwhere.py to the excludes line of APPZLIB
-the unpack script puts the sql.mar file in the install dir and creates
a sqlwhere.py in the install dir too. The sqlwhere.py only contains the
path
to the sql.mar file(in this case this is the install dir).

Here is a sample cfg file:

[MYINSTALL]
type= INSTALL
name= Install_ehk.exe
bindepends= ehk.py
zlib = INSTALLZLIB
misc= MYSTANDALONE, ROOT
debug = 0
excludes = PyWinTypes15.dll, win32api


[MYSTANDALONE]
type= STANDALONE
icon= e:\newehk\images\ehk.ico
name= ehk.exe
script= ehk.py
zlib = APPZLIB
userunw = 1
support = 0
debug = 0


[ROOT]
type = CARCHIVE
name = root.pkg
misc = sql.mar
support = 0
installscript = createroot.py


[APPZLIB]
name= ehk.pyz
dependencies= ehk.py
excludes= dospath, posixpath, macpath, c:\Program
Files\Python\kwp\sqlwhere.py


[INSTALLZLIB]
name = installzlib.pyz
dependencies = installutils
includes = installutils
excludes = dospath, posixpath, macpath         


and the createroot.py:

installutils.installCArchive(os.path.join(idir,'root.pkg'), idir,
'.'+os.sep)
f=open(os.path.join(idir,'sqlwhere.py'),'w')
f.write("filename = '%s'" % os.path.join(idir,'sql.mar'))
f.close()            

Best Regards,
Arpad Kiss



From alex at magenta.com  Thu Jan  6 07:10:15 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 6 Jan 2000 13:10:15 +0100
Subject: trying again: CgiHttpServer workalike for Win/NT and Win/98?
References: <837o38$b65$1@serv1.iunet.it> <83b4a5$d5r$1@nntp3.atl.mindspring.net> <3873EA29.C99706D2@usit.net>
Message-ID: <011b01bf583f$65d97fc0$c02b2bc1@martelli>

Chuck Esterbrook writes:

> A few weeks ago, I was in the exact situation that Alex was is in.
>
> I found Medusa difficult to get along with for lack of documentation. And
> CGIHTTPServer difficult because I'm using Windows 95.
>
> I ended up downloading Apache (http://www.apache.org) and I'm quite happy
with it.
> It supports having "#!python" in your CGI scripts and everything works
fine.

Just for completeness (not sure if I mentioned it in this group) -- I
ended up with a similar solution, except I downloaded Xitami instead
of Apache (looks simpler/smaller to me, but I could be wrong).  It
offers a similar "shebang-despite-being-on-Windows" approach to
runnin Python CGI scripts.


Alex






From kuncej at mail.conservation.state.mo.us  Wed Jan 12 10:54:11 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Wed, 12 Jan 2000 09:54:11 -0600
Subject: Python and ArcView GIS
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca> <85hs33$4t9$2@newsserver.rrzn.uni-hannover.de>
Message-ID: <ks1f4.325$B5.6597@news.more.net>

> Why do you need to embed python in ArcView?
> Would it not be enough to just use the COM objects with python's
> win32com.client?

I haven't seen Bruce's work, but I'm hoping that it would allow
distribution of python code to ArcView users who do not have
python installed on their machine. win32com is great, but you
really need to have the full python and win32 extensions
installed on the local machine to make it work.

Anyway, I'm eager to see it. I've often wished that ESRI
had chosen python as the basis for Avenue, rather
that pseudo-Java[script]-garbage.

--Jeff





From hartmut at oberon.noris.de  Thu Jan  6 15:48:36 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Thu, 06 Jan 2000 21:48:36 +0100
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com>
Message-ID: <16600399@oberon.noris.de>

Grant Edwards (grant at nowhere.) schrieb:

 >Seconded.  If you already know a few languages, and already now
 >OOP, then PER is fine to learn from.

Does it contain some usefull "pattern", too?

+++hartmut


From aahz at netcom.com  Wed Jan  5 16:28:08 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 5 Jan 2000 21:28:08 GMT
Subject: TypeError: slice index must be int ?
References: <20000105111525.C1172@lexgen.com>
Message-ID: <850d18$lop$1@nntp4.atl.mindspring.net>

In article <20000105111525.C1172 at lexgen.com>, Hao Zhou  <hz at lexgen.com> wrote:
>
>Is it possible to use long int (say 2000000000L) to 
>slice a string with gigabytes size?

You may be able to use UserList or the arrays in NumPy to simulate that,
but I don't think you can do it directly.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From dcalvelo at lailp2pc2.univ-lille1.fr  Fri Jan 28 11:46:55 2000
From: dcalvelo at lailp2pc2.univ-lille1.fr (Daniel Calvelo)
Date: 28 Jan 2000 16:46:55 GMT
Subject: ole structured storage documents under *nix
References: <88256874.00095CFE.00@avnet.com>
Message-ID: <86sh5v$r9o$1@netserv.univ-lille1.fr>

Lance Ellinghaus (Lance.Ellinghaus at avnet.com) wrote:

: I have been looking for either Python or C code that will allow me 
: to read and write OLE structured documents..

: Anyone know of any??

: Please let us know!!

: Thank you,

: lance

1. There is wvlib, under the GPL, compiles like a charm, available at

http://www.wvware.com

It is used within AbiWord (www.abisource.com) for Word import, and from what 
was called MSWordView to convert .doc into .html.

2. Otherwise, you can try to rewrite in Python <grin> the OLE::Storage 
module from Perl.

3. Or you can use the format docs. The wv site has pointers and instructions 
to get'em. 

HTH

Daniel.


From shang.remove_edu at st.jyu.fi.edu  Wed Jan 26 17:50:29 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Thu, 27 Jan 2000 00:50:29 +0200
Subject: Tuples
References: <20000126221606.A962@stopcontact.palga.uucp>
Message-ID: <86nton$ob6$1@mordred.cc.jyu.fi>

Gerrit Holl <gerrit.holl at pobox.com> wrote in message
news:20000126221606.A962 at stopcontact.palga.uucp...

> With tuples, some things aren't really Pythonic. In some cases, even
> the Perl-like TMTOWTDI thing is true!

There's more than one way to write almost anything in almost any
language. You can ALWAYS add more parethesis if you like. For example
in C:

a = b + c is equivalent to a = (b + c)

> a = {}
> a[1,2] = 'f'
>
> is equivalent to:
> a = {}
> a=[(1,2)] = 'f'

The parenthesis are NOT part of the tuple syntax (except for zero and
one element tuples). They are just needed sometimes for different
grouping (just like parethesis usually are).

> And:
> "aaa %s bbb" % 'foo'
>
> is equivalent to:
> "aaa %s bbb" % ('foo',)

This is not a feature of tuples. The string format operator is defined
like this to make code prettier (i.e. to avoid the ugly (x,)-syntax).


--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi

Get paid to surf the Web!
http://www.alladvantage.com/home.asp?refid=BOT021




From mhagger at blizzard.harvard.edu  Wed Jan  5 22:53:27 2000
From: mhagger at blizzard.harvard.edu (Michael Haggerty)
Date: 05 Jan 2000 22:53:27 -0500
Subject: How to with **kw
References: <850vrc$v3g$1@nnrp1.deja.com>
Message-ID: <xj0hfgrc0ug.fsf@cyclone.harvard.edu>

wiktor1325 at my-deja.com writes:

> Is there a way to pass **kw args
> from one function to another,like:
> 
> def func1(**kw):
>   for key in kw.keys():
>      print key
> 
> def func2(**kw):
>   a_way_to_call_func1_with(kw)

The canonical way would be

    def func2(**kw):
        apply(func1, (), kw)

Michael

-- 
Michael Haggerty
mhagger at blizzard.harvard.edu


From tim_one at email.msn.com  Sun Jan 23 22:55:26 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 23 Jan 2000 22:55:26 -0500
Subject: Lexing in Python 2
In-Reply-To: <388B855A.78D6E46B@prescod.net>
Message-ID: <000301bf661e$d8dedf00$902d153f@tim>

[Paul Prescod]
> At the last XML conference I told someone that the reason
> that re doesn't take a stream instead of string parameter
> was because anyone sane working on a large file would use
> a proper tokenizer. Shouldn't such a tokenizer come with
> Python? With all due respect, what the hell is shlex and
> how did it get into the standard distribution?

I've wondered that myself <0.9 wink>.

> I mean the standard distribution alone must contain half a
> dozen hand-coded lexers and in a few places, the weirdness
> you need to apply regular expressions to streams. Surely we
> can do better for Python 2?

If nobody was motivated enough to write the code for Python 1, I don't know
why that would change for Python 3000 (that's what Guido insists on calling
it now <wink>).  If you want a *fast* Python lexer today, mxTextTools is
your best hope.

> It is my unconsidered, uneducated opinion that lexers do not
> vary as widely as parsers (LL(1), LR(1), LR(N) etc.) so we
> could just choose one at random and start building modules
> around it.

Curiously, mxTextTools is nothing like lex/flex.  Flex does such a good job
it's hard to get motivated to duplicate all that effort (it's not easy)
solely to get something releasable under a more Python-like license.  I
don't know how Marc-Andre would feel about folding mxTextTools into the
distribution.

> All in favor? Opposed? Carried.

I'm almost never opposed to someone else doing work <wink>.

BTW, there are several interesting parsing projects going on in the Java
world; at least JPython should be able to exploit them.

IDLE-adds-at-least-two-more-hand-crafted-lexers-ly y'rs  - tim





From sholden at bellatlantic.net  Mon Jan  3 11:11:01 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Mon, 03 Jan 2000 16:11:01 GMT
Subject: Python on NT systems (help a Unix guy, please)
References: <19991218113757.A637@Ridcully.home>
Message-ID: <3870CA2B.DD09F50E@bellatlantic.net>

Don't suppose anyone has implemented freeze() on NT, have they?
I only have the binary 1.5.2 on my NT box, and I don't see it
there :-(

regards
 Steve

Malcolm Tredinnick wrote:
> 
> I recently protyped a program in Python that is destined to run on a Windows
> NT box. As usual, the Python version runs well enough that I have no real wish
> to rewrite it in C. So all that remains is to convince the owner of the NT box
> to install Python on his machine.
> 
> The problem: disk space is a little tight on the target machine. Since I don't
> have any Windows machines at home at the moment, could somebody please let me
> know (just roughly) the disk space requirements for Python on NT? (private
> email is ok if you don't want to spam the list).
> 
> Just for comparison purposes, a reasonably standard Python install on my Linux
> box (the only additional thing is the LLNL Numeric package) takes up just
> under 24MB, the requirements for tkinter is another 5.5MB with wxPython
> (source code and all) being 18MB (plus whatever wxWindows requires).
> 
> Thanks in advance,
> Malcolm Tredinnick
> 
> --
> Despite the high cost of living, have you noticed how popular it remains?


From jcollins at pacificnet.net  Thu Jan 27 14:36:38 2000
From: jcollins at pacificnet.net (Jeffery D. Collins)
Date: Thu, 27 Jan 2000 11:36:38 -0800
Subject: compiling python programs
References: <389068A2.CD0DFCE0@mail.bulkley.net>
Message-ID: <38909E46.E8E95F7E@pacificnet.net>

Download the python source distribution, then look in Tools/freeze.
It explained quite well there and an example is included.

Jeff


Peter Schopfer (aka Dr. J. - magician) wrote:

> Hello
>
> i read somewere that i could "freez" my python program into a c
> executable. how can i do this? could someone show me the process for
> doing this?
> Thanks



From piet at cs.uu.nl  Wed Jan 12 09:27:40 2000
From: piet at cs.uu.nl (piet at cs.uu.nl)
Date: 12 Jan 2000 15:27:40 +0100
Subject: RunPy (was py2bat (was: How do I make a Python .bat executabl e file?))
References: <6D8A17398E28D3119F860090274DD7DB4B3DC2@pces.cadlab.it>
Message-ID: <uu2kjjrfn.fsf@cs.uu.nl>

>>>>> Alex Martelli <Alex.Martelli at think3.com> (AM) writes:

AM> What's a .pyw file?  I thought pythonw.exe was the exact
AM> equivalent of python.exe, except for not using a console.
AM> Is there any doc about this anywhere...?

A .pyw file is a .py file with the only difference that on MS Windows when
you start it e.g. by double-clicking, pythonw is used rather than python to
run it.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From aahz at netcom.com  Sat Jan 15 12:14:21 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Jan 2000 17:14:21 GMT
Subject: Help me improve my Python Skills, please
References: <m2g0vzsrte.fsf@excalibur.peernet> <m3u2kfabkp.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <85q9td$vrs$1@nntp6.atl.mindspring.net>

In article <m3u2kfabkp.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>
>Also, you're using re where string would do. This is also pretty
>common... remember the Tom Christiansen (I think) quote:
>
>"When faced with a problem, soem people think 'I'll use regular
>expressions.' Now they have two problems."

I think that's actually the Timbot.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From kevinmck at infi.net  Wed Jan 26 18:18:49 2000
From: kevinmck at infi.net (Kevin McKinley)
Date: Wed, 26 Jan 2000 18:18:49 -0500
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com> <86fl0m$gmj$1@inputplus.demon.co.uk> <TVPi4.2485$K5.22193@newsfeed.slurp.net>
Message-ID: <l40v8skdl4mdo6j98bq2pfaja7kr430adu@4ax.com>

I'll be starting Python soon.

Which tutorial are you referring to?


From ngps at madcap.dyn.ml.org  Mon Jan 24 10:12:58 2000
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 24 Jan 2000 15:12:58 GMT
Subject: Need XML-RPC help
References: <B4B0ED62.316E%tony.mcdonald@ncl.ac.uk>
Message-ID: <388c6bfa.0@news.cyberway.com.sg>

According to Tony McDonald  <tony.mcdonald at ncl.ac.uk>:
>I can't get XML-RPC to work through our proxy.

Try this patch:

*** xmlrpclib.py.orig	Sun Sep 19 16:00:47 1999
--- xmlrpclib.py	Sun Sep 19 16:01:02 1999
***************
*** 553,564 ****
      # client identifier (may be overridden)
      user_agent = "xmlrpclib.py/%s (by www.pythonware.com)" % __version__
  
      def request(self, host, handler, request_body):
          # issue XML-RPC request
  
          import httplib
!         h = httplib.HTTP(host)
!         h.putrequest("POST", handler)
  
          # required by HTTP/1.1
          h.putheader("Host", host)
--- 553,576 ----
      # client identifier (may be overridden)
      user_agent = "xmlrpclib.py/%s (by www.pythonware.com)" % __version__
  
+     def __init__(self, proxy_host=None, proxy_port=8080):
+         self.use_proxy=0
+         if proxy_host is not None:
+             self.proxy_host=proxy_host
+             self.proxy_port=proxy_port
+             self.use_proxy=1
+ 
      def request(self, host, handler, request_body):
          # issue XML-RPC request
  
          import httplib
!         req='http://%s%s' % (host, handler)
! 
!         if self.use_proxy:
!             h = httplib.HTTP(self.proxy_host, self.proxy_port)
!         else:
!             h = httplib.HTTP(host)
!         h.putrequest("POST", req)
  
          # required by HTTP/1.1
          h.putheader("Host", host)

-- 
Ng Pheng Siong <ngps at post1.com>


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



From tjg at avalongroup.net  Wed Jan 19 00:44:26 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Tue, 18 Jan 2000 21:44:26 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <oqpuuystm6.fsf@titan.progiciels-bpi.ca>
Message-ID: <38854F3A.4FBA74F9@exceptionalminds.com>

Fran?ois Pinard wrote:
> "tye4" <tye4 at yahoo.com> writes:
> 
> > I don't need to design a language to detect flaws in it.  This problem
> > needs to be fixed.  [...] (I don't want to type 4 spaces for every line.)
> 
> My humble feeling is that you need your editor fixed, and that you report
> the problem you have as being one of Python language design.  We should
> be much wiser, and keep things in better proportion.  Let me suggest that
> you pick some better editor, and just enjoy Python like most of us do! :-)

You know Francois, this was my first reaction to this also. A good
editor makes such a huge difference that I can barely work in one that
doesn't have a "python" mode and syntax highlighting. However, it does
take a little work to configure and editor to make it work.

Tye, have you taken a look at an editor with a Python mode? or used Idle
or Pythonwin?

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From shang.remove_edu at st.jyu.fi.edu  Mon Jan 24 02:19:12 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Mon, 24 Jan 2000 09:19:12 +0200
Subject: [Newbie] Python as a frontend to other apps.
References: <iJKi4.25111$pb2.1977836@tw11.nn.bcandid.com> <pgLi4.25161$pb2.1983916@tw11.nn.bcandid.com>
Message-ID: <86guct$ovq$1@mordred.cc.jyu.fi>

Patrick H. Mullins <pmullins at gatecom.com> wrote in message
news:pgLi4.25161$pb2.1983916 at tw11.nn.bcandid.com...
> > import os
> > password="secretstuff"
> > infile="test.idea"
> > outfile="test.zip"
> > os.system("idea -dk %(password) %(infile) %(outfile)")
>
> Found out that I just had to move a few things around.  :-)
>
> os.system("idea -dk %s %s %s" % (password, infile, outfile))

You could use your original syntax with a few small additions, i.e.

os.system("idea -dk %(password)s %(infile)s %(outfile)s" % vars())

You have add the format specifier 's' (string) after the variable name
or you'll get an 'incomplete format' error. Also, you must provide a
dictionary that contains the variables (returned by vars() ).

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi




From news at dorb.com  Sat Jan  1 13:50:01 2000
From: news at dorb.com (Darrell)
Date: Sat, 1 Jan 2000 13:50:01 -0500
Subject: pickle problem:cant reload object saved with binary option
References: <386e1de1.998928692@news-server> <YHpb4.914$Sr1.104202@typhoon.nyroc.rr.com> <386e31c2.1004019140@news-server>
Message-ID: <3Zrb4.971$Sr1.107689@typhoon.nyroc.rr.com>

import cPickle

class X:
    def __init__(self):
            self._x=1
            self._y=2

    def save(self,file='crcc.sav'):
            blah=open(file,'w+b')
            cPickle.dump(self,blah,1)
            blah.close()

    def load(self,file='crcc.sav'):
            blah=open(file,'r+b')
            val=cPickle.load(blah)
            blah.close()
            return val

x=X()
x._x=111111
x._z=333333
x.save()

x1=X()
x1=x1.load()
for n in x1.__dict__.keys():
    print n,'=', getattr(x1,n)

######################## Output
_x = 111111
_y = 2
_z = 333333


Replaced all tabs with spaces so this code should survive the mail.
--
--Darrell

"Marshall" <python at scoobysnacks.com> wrote in message
news:386e31c2.1004019140 at news-server...
> Thanks, now I have a new problem.  It still works without the binary
> option.





From mark at chem.uwa.edu.au  Fri Jan 21 19:58:38 2000
From: mark at chem.uwa.edu.au (Mark C Favas)
Date: 22 Jan 00 00:58:38 GMT
Subject: Python - SIGPIPE handled bug
References: <20000121162925.A189033@vislab.epa.gov>
Message-ID: <mark.948502718@declan>

Randall Hopper <aa8vb at yahoo.com> writes:


>  ---------------------------------------------------------------------------
>    > python -c "for i in xrange(1000): print 'Some output'" | head -2
>    Some output
>    Some output
>    Traceback (innermost last):
>      File "<string>", line 1, in ?
>    IOError: [Errno 32] Broken pipe
>  ---------------------------------------------------------------------------

On my system (1.5.2, latest from CVS, Tru64 (nee Digital) Unix 4.0F), the
above works perfectly under both csh and sh, viz:

120> python -c "for i in xrange(1000): print 'Some output'" | head -2
Some output
Some output
121> sh
$ python -c "for i in xrange(1000): print 'Some output'" | head -2
Some output
Some output
$ echo $?
0
$

However, if I put it into a file as:

117> cat junk2.py
def my_program():
  for i in xrange(10000): 
    print 'Some output'

my_program()
118> python junk2.py | head -2
Some output
Some output
Traceback (innermost last):
  File "junk2.py", line 5, in ?
    my_program()
  File "junk2.py", line 3, in my_program
    print 'Some output'
IOError: [Errno 32] Broken pipe

I get your result. No solution, just another data point.

Mark
--
Email  - mark at chem.uwa.edu.au      ,-_|\                           Mark C Favas
Phone  - +61 9 380 3482           /     \               Department of Chemistry
Fax    - +61 9 380 1005      ---> *_,-._/   The University of Western Australia
                                       v                               Nedlands
Loc    - 31.97 S, 115.81 E                               Western Australia 6009


From zakons at objectsbydesign.com  Thu Jan 13 13:36:07 2000
From: zakons at objectsbydesign.com (zakons at objectsbydesign.com)
Date: Thu, 13 Jan 2000 18:36:07 GMT
Subject: class methods?
References: <20000112145448.29699.qmail@web804.mail.yahoo.com>
Message-ID: <85l5ul$bea$1@nnrp1.deja.com>

After looking at the alternatives, it is clear that it is much cleaner
to just create an instance to print out the statistics, as in the
following:


class Company:

    # class-level statistics
    field_counts = { }
    company_count = 0

    # some code to read in the company fields
    def read ( self, file ):
        # <...>

        # for every name/value, get statistics
        self.do_stats ( name, value )

        # <...>

        # keep track of count of companies
        Company.company_count = Company.company_count + 1

    # record the statistics
    def do_stats ( self, name, value ):
        counts = Company.field_counts
        if not counts.has_key ( name ):
            counts [ name ] = 1
        else:
            counts [ name ] = counts [ name ] + 1

    #print out the statistics
    def print_stats ( self ):
        counts = Company.field_counts
        list = counts.keys()
        list.sort()
        for name in list:
            percent = ( 100 * counts[name] ) / Company.company_count
            print "%s, %d" % ( name, percent )


# to print out the statistics
Company().print_stats()


Of course, the question: "Is this really an instance of a Company" comes
up. Does it really matter?

It would still be interesting to know if having 'class methods' is
problematic within a Python environment or if it is just a simplifying
assumption...

Thanks for the responses,
Stuart


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


From backov at nospam.csolve.net  Tue Jan 11 14:32:54 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Tue, 11 Jan 2000 19:32:54 GMT
Subject: Please, someone tell me what's going wrong here. ;>
Message-ID: <GzLe4.54805$Ke.42098@news21.bellglobal.com>

I posted a while back on embedding, and after scouring deja and reading the
demos, I can't even get this simple example to work. It's essentially a C++
Builder'ized version of the simple embedding demo.. PyRun_SimpleString
returns -1 and foo is never executed.

TForm1 *Form1;

__fastcall TForm1::TForm1(TComponent* Owner)
    : TForm(Owner)
{
}

void initxyzzy(); /* Forward */

/* A static module */

static PyObject * xyzzy_foo(PyObject *self, PyObject *args)
{
    Form1->Memo1->Lines->Add("Blah");

 return PyInt_FromLong(42L);
}

static PyMethodDef xyzzy_methods[] = {
 {"foo",  xyzzy_foo, METH_VARARGS},
 {NULL,  NULL}  /* sentinel */
};

void initxyzzy()
{
 PyImport_AddModule("xyzzy");
 Py_InitModule("xyzzy", xyzzy_methods);
}

void __fastcall TForm1::Button1Click(TObject *Sender)
{
    int a;
    Py_SetProgramName(Application->ExeName.c_str());
    Py_Initialize();
    initxyzzy();
    a=PyRun_SimpleString("xyzzy.foo()");
    Py_Finalize();

}

void __fastcall TForm1::FormDestroy(TObject *Sender)
{
    Py_Finalize();
}





From fcy at enteract.com  Sun Jan  9 12:33:28 2000
From: fcy at enteract.com (Fred Yankowski)
Date: 9 Jan 2000 17:33:28 GMT
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <85agp8$ks0$1@eve.enteract.com>

Web2010.com offers Python 1.5.2 for CGI, at least on their "Premier"
UNIX accounts (about $25/month).  They offer PHP3 and MySQL (1
database) on the same accounts, which is why I use them.

--
Fred Yankowski



From skaller at maxtal.com.au  Sat Jan  8 09:54:52 2000
From: skaller at maxtal.com.au (skaller)
Date: Sun, 09 Jan 2000 01:54:52 +1100
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
		<14452.45436.641890.197509@beluga.mojam.com>
		<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com>
Message-ID: <38774FBC.EB8281CF@maxtal.com.au>

Skip Montanaro wrote:
> 
>     Darrell> Don't know if this if fair. But exceptions are slow when raised.
> 
> Agreed in the case of your two-branch case.  If you have a state machine
> with a bunch of states, try/except/except/... appears to perform about as
> well as if/elif/elif/.../else:

	Yeah, but this is an artefact, since Python does not
have a proper switch statement. If/elif ... is clearly
linear time .. a switch is constant time. The Knuth
construction should support constant time loop exits,
as if it had been coded with gotos instead of breaks.

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From m.faassen at vet.uu.nl  Mon Jan 10 06:16:26 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 10 Jan 2000 11:16:26 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net> <slrn87cvv9.gcu.kc5tja@garnet.armored.net>
Message-ID: <85cf2a$dsq$1@newshost.accu.uu.nl>

Samuel A. Falvo II <kc5tja at garnet.armored.net> wrote:
[snip]
> more stable NT version.  People have been saying, "Mozilla is almost as
> stable as..." for the last three years.

These people are good at looking into the future then. The netscape
sources were opened up in March '31, 1998. I think only at the end of '98
or so the decision was made to go with the new codebase, so the maximum
amount of time which people could've been saying this must be about
a year. :)

Not that Mozilla is stable in my experience (getting better); it's pre-alpha
though. I have good hopes for it, but patience is definitely required. :)

Regards,

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


From parkw at better.net  Mon Jan 17 16:04:54 2000
From: parkw at better.net (William Park)
Date: Mon, 17 Jan 2000 16:04:54 -0500
Subject: Removing tags from a html-file
In-Reply-To: <3882A004.A5F150F@bibsyst.no>
References: <3882A004.A5F150F@bibsyst.no>
Message-ID: <20000117160454.B327@better.net>

On Mon, Jan 17, 2000 at 05:52:20AM +0100, Thomas Weholt wrote:
> Hi,
> 
> I want to remove all tags from a html-file, so that only the plain-text
> remains. How, if so, can this be done using htmllib.py?
> 
> If not, is there any other way to do this ??

I use a text browser called 'lynx':
    lynx -dump http://... > x.txt

William



From pinard at iro.umontreal.ca  Sat Jan 15 15:51:22 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 15 Jan 2000 15:51:22 -0500
Subject: xpot version 0.3
Message-ID: <oq66wv5a9h.fsf@titan.progiciels-bpi.ca>

Hi, people.  A few days ago, I told you about a program able to extract
a POT file out of a series of Python sources.  I made it available as:

   http://www.iro.umontreal.ca/contrib/po/xpot-0.3.tar.gz

Your comments are welcome if, by any chance, you have some use for it. :-)

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





From boud at rempt.xs4all.nl  Sun Jan 30 03:43:27 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 30 Jan 2000 08:43:27 GMT
Subject: Python MySQL.connect() question
References: <3893CD31.B17E8B84@regnoc.com>
Message-ID: <870tjf$514$1@news1.xs4all.nl>

Jim Conger <jimc at regnoc.com> wrote:
> I have mySQL running on a RH Linux 6.1 box, and have added the
> MySQLmodule.so to my python library.  I can't figure out how to connect
> within python.  My attempts look like:

>>>> import MySQL;
>>>> db = MySQL.connect();
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> MySQL.error: Can't connect to local MySQL server
>>>>

I'm not really surprised - I think you need to supply
some parameters:

    hostname="essle"
		username="boud"
		password=""
		databasename="languages"
    self.dbConnection = MySQL.connect(hostname,username,password)
	  self.dbConnection.selectdb(databasename)
				
Which is what I did when I used MySQLmodule; however, I've since
began using the DB-API II compliant MySQLdb, but I miss some of
the conveniences of MySQLmodule.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From seanb at home.com  Mon Jan 24 12:07:50 2000
From: seanb at home.com (seanb at home.com)
Date: Mon, 24 Jan 2000 09:07:50 -0800 (PST)
Subject: Usability reports (was RE:tabs do WHAT?)
In-Reply-To: <000201bf6640$d154d100$022d153f@tim>
Message-ID: <200001241707.JAA14172@seanb.sttln1.wa.home.com>

On 24 Jan, Tim Peters wrote:
> [seanb at home.com]
>> Out of curiosity, are there any online copies of these
>> usability studies that were so formative to pythone?  Such
>> studies might make interesting reading and some people may
>> have more trust in the results of a study they could examine
>> online.
>>
>> Failing online access, is there any public access to write-ups
>> of these studies?
> 
> Agreed it would be interesting!  I don't have any links for you -- I knew
> ABC only as a user, but read several papers that came out of the project.
> That was over a decade ago, though, and certainly not on the web; the
> specific claim about indentation is one I don't recall reading about then,
> but is one Guido has mentioned on c.l.py several times since (but he hasn't
> bothered to jump into a "whitespace thread" in years -- you'd have to search
> hard for that).
> 
> The ABC project is no longer active.  Its sponsoring institution still has
> *some* info available (http://www.cwi/nl/).  Leo Geurts, Lambert Meertens
> and Steven Pemberton were ABC's primary designers.  A quick web search on
> their names turned up this citation:
> 
>     "An Alternative Simple Language and Environment for PC's"
>     S. Pemberton, IEEE Software 4(1):56-64 (Jan 1987)
> 
> Presumably (a) that contains more info, and (b) is not available online.
> 
> Here's a snazzy quote from SP's homepage (http://www.cwi.nl/~steven/):
> 
>     We did requirements and task analysis, iterative design,
>     and user testing.  You'd almost think programming languages
>     were an interface between people and computers.
> 
> today's-generation-would-probably-find-3D-rotating-
>     metallic-braces-easier-to-grasp<wink>-ly y'rs  - tim
> 
> 
> 
	I found a number of reports available for ordering at 
http://www.cwi.nl/~steven/abc/publications.html.  Among the more
interesting items in this listing are:

Designing a Beginners' Programming Language

Leo Geurts and Lambert Meertens, 18 pages. This was the first article on B. It
clarifies some of the design objectives and describes the result of the first 
iteration in the defining process. Published in New Directions in Algorithmic 
Languages 1975, ed.S.A. Schuman, IRIA, Rocquencourt (1976). Available from CWI
 as report IW 46, price Dfl.4.00. 

Program Text and Program Structure

Lambert Meertens, 11 pages. Proposes the method of stepwise refinement as a 
means to make the structure of program development explicit in the program text. Published in Constructing Quality Software, ed. S.A. Schuman, North-Holland Publ.
Co. (1978). Available from CWI as report IW 78, price Dfl.4.00. 

	These both seem interesting and I am tempted to order them.  My
biggest stumbling block is finding out what a DFL is (sorry, clueless
American here, I think in USD).  Hazarding a guess that Dfl ==
Netherlands Guilder == NLG, I plugged4 into the Universal Currency
converter to discover that 4 NLG == $1.82256 USD.  In other words, I
would probably end up paying more for shipping than for the reports.

	I have heard many claims about Python's usability and the ease
of learning python.  While my own experience strongly concurs with
these claims, I am eager to get my hands on some concrete information
aout this.  Perhaps eventually such information would be available from
the python.org website?


--
Sean Blakey
(206)297-7123
quine = ['print "quine =",quine,"; exec(quine[0])"']; exec(quine[0])




From andrew.henshaw at gtri.gatech.edu  Mon Jan 24 10:58:15 2000
From: andrew.henshaw at gtri.gatech.edu (Andrew Henshaw)
Date: Mon, 24 Jan 2000 10:58:15 -0500
Subject: Whitespace delimiters suck - Seen Occam?
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <86cmom$5pp$1@news-int.gatech.edu> <86h7jm$o7i@newton.cc.rl.ac.uk> <388C760A.1E66C20@gtri.gatech.edu>
Message-ID: <388C7697.FCF141EB@gtri.gatech.edu>


Andrew Henshaw wrote:

Whoops, sorry for the odd line breaks of the last message.  Tried a
different newsreader.


From mok at imsb.au.dk  Tue Jan  4 07:51:46 2000
From: mok at imsb.au.dk (mok)
Date: Tue, 04 Jan 2000 13:51:46 +0100
Subject: RPM-interface/module
References: <386B3CDB.FF843280@bibsyst.no>
Message-ID: <3871ECDB.9120CE7F@imsb.au.dk>


Thomas Weholt wrote:
> 
> Is there a RedHat package interface/module available for Python??

I wrote a simple wrapper for rpm. It calls the program itself, and does
not rely on shared modules. There are classes for an rpm file, and for a
database entry. The classes act like a Python dictionary, with the rpm
tags as keys.

See http://imsb.au.dk/~mok/linux/doc/rpmfile.py

/Morten

-- 
Morten Kjeldgaard   <mok at imsb.au.dk>             | Phone : +45 89 42 50 26
Institute of Molecular and Structural Biology    | Fax   : +45 86 12 31 78
Aarhus University                                | Home  : +45 86 18 81 80
Gustav Wieds Vej 10 C, DK-8000 Aarhus C, Denmark | icq   : 27224900


From shang.remove_edu at st.jyu.fi.edu  Tue Jan 25 06:07:19 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Tue, 25 Jan 2000 13:07:19 +0200
Subject: delete files?
References: <86jtm2$7cs$1@news1.sunrise.ch>
Message-ID: <86k067$ol1$1@mordred.cc.jyu.fi>

nathan <nnever50 at yahoo.com> wrote in message
news:86jtm2$7cs$1 at news1.sunrise.ch...

> somebody knows how can I quickly delete files (without using
> os.system..) ? Working under WinNT, v.1.5.2.

You can use os.remove

http://www.python.org/doc/current/lib/os-file-dir.html

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi

Get paid to surf the Web!
http://www.alladvantage.com/home.asp?refid=BOT021




From nickb at earth.ox.ac.uk  Tue Jan 18 08:12:52 2000
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Tue, 18 Jan 2000 13:12:52 +0000
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de>
Message-ID: <388466D4.CC797CFC@earth.ox.ac.uk>

>
> Get used to it, and don't use tabs.
>

! - whats wrong with tabs?

:-)))))))))))))




From breiter at usf.Uni-Osnabrueck.DE  Thu Jan 13 05:25:18 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 13 Jan 2000 10:25:18 GMT
Subject: Python and ArcView GIS
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>
    <85hs33$4t9$2@newsserver.rrzn.uni-hannover.de>
    <XXcf4.16787$2x3.303971@newscontent-01.sprint.ca>
Message-ID: <85k96e$hn1$1@newsserver.rrzn.uni-hannover.de>

In article <XXcf4.16787$2x3.303971 at newscontent-01.sprint.ca>,
	"Bruce Dodson" <bruce_dodson.nospam at bigfoot.com> writes:
> Bernhard Reiter wrote in message
> <85hs33$4t9$2 at newsserver.rrzn.uni-hannover.de>...

>>Would it not be enough to just use the COM objects with python's
>>win32com.client?
> 
> Sorry, I forgot who I was talking to.  :-)  I didn't mean to imply that
> ArcView would bestow this benefit on Python.  I meant the opposite: Without
> Python, ArcView doesn't have any way to deal with COM objects.  With Python
> (and win32com), it does have a way.  So your suggested alternative is
> exactly what I meant in the first place!

There is still confusion in my head. :)
I am not exactly sure if ArcView exposes COM objects right now.
If is does, python could control ArcView and then you can also control
other COM objects with python.

If you embed python in ArcView, then python might get more control about 
ArcView. Of course when inside python you can also control other COM objects.

So the question is:
	Does your extension enable python to have more control about ArcView
	as it would have controling it over ArcView's COM objects from outsite?

	Do you need the python extension to expose ArcView capabilities
	to other programs?

If the answer is no, then I assume that you can access other COM objects
from within the python script, but you would not need to embed python
but you could also control it and the other programs from an outsite python
script.

Do I make any sense?

>>I hope you will license it as free software.
> It is definitely free software.  The license will be BSD-style.

Great.
	Bernhard
-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From ngps at madcap.dyn.ml.org  Thu Jan 13 09:49:36 2000
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 13 Jan 2000 14:49:36 GMT
Subject: MySQLdb, blobs, and inserting
References: <85itgj$mlc$1@nnrp1.deja.com> <85k0fd$hv2$1@news1.xs4all.nl>
Message-ID: <387de600.0@news.cyberway.com.sg>

According to Boudewijn Rempt <boud at rempt.xs4all.nl>:
> >>>> c.execute(r"INSERT INTO items (story) VALUES (%s)" % file )
> 
> I haven't tried it - but wouldn't 
> ...." % file.read() )
> work? I mean, are you trying to insert a file object or the contents
> of the file? 

Assuming the original poster is trying to insert the content of the 
file, try:

  c.execute('insert into items (story) values ("%s") % file.read())
                                               ^^^^
                                               Note this!

Cheers.

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



From djc at itga.com.au  Thu Jan 20 22:19:34 2000
From: djc at itga.com.au (Dave Cole)
Date: 21 Jan 2000 14:19:34 +1100
Subject: Attributes acting as arrays (in C extensions)?
References: <slrn88f7le.ke5.joakim@login1.styx.net>
Message-ID: <87puuw9kmx.fsf@heresy.itga.com.au>

>>>>> "Joakim" == Joakim Ziegler <joakim at login1.styx.net> writes:

Joakim> I'm implementing a Python module in C. I've already done
Joakim> object attributes which invoke C functions to get and set the
Joakim> values, but is it possible to have object attributes that act
Joakim> as arrays? So that you can do

Joakim> object.attribute[n]

Joakim> in Python, and call a C function with object, the name of the
Joakim> attribute, and n as arguments? I've not been able to find docs
Joakim> on how to do this.

I suppose you could implement that particular attribute as another C
object which implements the sequence interface.

- Dave


From felixt at dicksonstreet.com  Sat Jan  8 11:53:38 2000
From: felixt at dicksonstreet.com (Felix Thibault)
Date: Sat, 08 Jan 2000 10:53:38 -0600
Subject: control structures (was "Re: Sins")
In-Reply-To: <38774EBA.40716D1E@maxtal.com.au>
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
 <14452.45436.641890.197509@beluga.mojam.com>
Message-ID: <3.0.5.32.20000108105338.007ba100@mail.dicksonstreet.com>

At 01:50 1/9/00 +1100, skaller wrote:
>Skip Montanaro wrote:
>> 
>>     Emile> Doesn't this do that now? or is there some objection to this
usage?
>> 
>>     [ ... snip try/except used to implement state machine ... ]
>> 
>> I find nothing wrong with it.  Some people object to using try/except for
>> anything but handling errors however.
>
>My objection is twofold. First, the syntax is unweildy:
>
>	try:
>		for i in range(3):
>			if ... : raise X
>	except X:
>
>requires at least one extra level of indentation.
>
>[I'm interested in a 'generic' way to combine
>python statements which avoids this (like 'elif' does)
>Any ideas?]
>
>	My second objection is more serious:
>try .. raise .. except is too dynamic (for this usage): 
>there is no way to match up the exceptions actually 
>thrown with the handlers.
>
>	The constructions under consideration
>(variants on a construction proposed by Knuth)
>would support compile time verification that
>every loop exit had a corresponding handler
>lexically following the loop.
>
>	In particular, 'falling through'
>such a loop is impossible: all exits must be explicit.
>

What would be the problem with doing something like this:

def Foundfunc(...):
    ...
def NotFoundfunc(...):
    ...
exitState = 'NotFound'
exitFuncd = {'Found':Foundfunc, 'NotFound':NotFoundfunc}
for i in X:
    if i == e:
        exitState = 'Found'
        break
exitFunOut = exitFuncd[exitState](...)
...



>	Current looping constructions are so bad,
>that the simplest of uses cannot be expressed
>using them : searching a list for an element
>cannot be written correctly:
>
>	for i in X: 
>		if X[i]=e: goto Found
>	print "Not Found"
>	goto next
>	Found: ...
>	next
>
>YUKKKKKK!!! There is no natural way to do this.
>We fall through the loop on failure, instead
>of success. (using a flag, or testing anything
>afterwards, is not acceptable as a solution).
>
>Using exceptions:
>
>	try:
>		for i in X:
>			if X[i]==e: raise Found
>		raise NotFound
>	except Found: ..
>	except NotFound: ..
>
>is much cleaner, but exceptions are excessively
>dynamic for what is essential a static control
>flow problem. A solution will resemble the
>above exception handling implementation .. but
>with the matching up of exits and handlers enforced
>at compile time.
>
>-- 
>John (Max) Skaller, mailto:skaller at maxtal.com.au
>10/1 Toxteth Rd Glebe NSW 2037 Australia
>homepage: http://www.maxtal.com.au/~skaller
>voice: 61-2-9660-0850
>
>-- 
>http://www.python.org/mailman/listinfo/python-list
>
>




From pmullins at gatecom.com  Sun Jan 23 16:51:11 2000
From: pmullins at gatecom.com (Patrick H. Mullins)
Date: Sun, 23 Jan 2000 21:51:11 GMT
Subject: [Newbie] Python as a frontend to other apps.
Message-ID: <iJKi4.25111$pb2.1977836@tw11.nn.bcandid.com>

Hello.  I'm just starting out with Python and I've already hit a
road block.  I'm trying to write a frontend for a commandline
app called idea, which is a basic encryption program..  This
app takes 3 arguments which I've stored as variables.
The problem is that I can't seem to get python to output the
contents of the variables on the commandline:

import os
password="secretstuff"
infile="test.idea"
outfile="test.zip"
os.system("idea -dk %(password) %(infile) %(outfile)")

Everything works fine up until it hits %(password) and so on.
What am I doing wrong here??????

Thanks!

Patrick Mullins
pmullins at gatecom.com





From thierry at acm.org  Fri Jan 28 11:23:36 2000
From: thierry at acm.org (Thierry Thelliez)
Date: Fri, 28 Jan 2000 09:23:36 -0700
Subject: Gif charts
Message-ID: <3891C288.DC7A85FB@acm.org>

Thanks for everybody who answered the charts thread.

I forgot to mention 3 things:
- I need a gif (the chart will be put in a web page),
- I imperatively need scattered plots (as well as histogram, line chart,
possibly pie charts)
- I want a  nice legend system. There will possibly be several series in
one chart.


Thanks !
Thierry



From aloy at ctv.es  Mon Jan 31 16:04:31 2000
From: aloy at ctv.es (Antoni Aloy)
Date: Mon, 31 Jan 2000 22:04:31 +0100
Subject: C, C++ to python
Message-ID: <874t8o$3lm$1@lola.ctv.es>

Hi!
Anybody knows if exits a tool to convert C or C++ code to Python. I like to
make a port for TVision and such a tool would be a great help.

Best regards

Antoni Aloy




From warlock at eskimo.com  Fri Jan 21 00:37:28 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Thu, 20 Jan 2000 21:37:28 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net> <8631bt$1nu@news.or.intel.com>  <slrn88d63f.13f.wtanksle@hawking.armored.net>
Message-ID: <slrn88fs4o.2d0.warlock@raven.myth>

On 20 Jan 2000 04:58:00 GMT, 
 William Tanksley, in the persona of <wtanksle at hawking.armored.net>,
 brought forth the following words...:

>On Tue, 18 Jan 2000 16:41:06 -0800, tye4 wrote:
>
>>I have offered you many reasons why we need 'end' block support..
>
>No, you haven't.  All you've done is claim that other languages don't use
>indentation.  So what?  Python uses assignment.  ML doesn't.  Does that
>mean ML is bad and Python is good, or is it the other way around?

what does ML do then? this has me puzzled...



-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From gmcm at hypernet.com  Wed Jan 19 00:00:31 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 19 Jan 2000 00:00:31 -0500
Subject: Python Rocks!
In-Reply-To: <8631bt$1nu@news.or.intel.com>
Message-ID: <1263864370-2295671@hypernet.com>

"tye4" wrote:

> So you don't think this is a problem?? Hello!, read all the messages in this
> thread. Seems to me a lot of programmers have a problem with this.

Go to DejaNews and look for "indentation" or "whitespace" in 
any period over the last 5 years. Most Pythoneers do not 
agree with you. Most say "<yawn> this crap again?" and go 
on to a different thread.

This particular horse stopped responding to beatings before 
c.l.py was a newsgroup. It's not going to wake up now.

> Thanks to this quirk, I now have to worry how an editor saves tabs to files.

I use 4 different editors with Python code, all of which do the 
right thing with a little bit of configuration. If your editor doesn't, 
there are tools which will reformat whitespace to your 
choosing (tabcleaner in the System directory of the contrib ftp 
site).

If your editor line wrapped at 72 characters, would you argue 
that any language which allows a line length of > 72 
characters was flawed?

> I have offered you many reasons why we need 'end' block support..
> You have not offered a single reason why this is not a problem..

All arguments on both sides have been presented hundreds of 
times already.


- Gordon



From nobody at nowhere.nohow  Sun Jan 23 19:36:47 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 24 Jan 2000 00:36:47 GMT
Subject: Python compiler?
References: <86duuq$8a1@OM9.omantel.net.om> <20000123092040.A2010@stopcontact.palga.uucp>
Message-ID: <slrn88n7kv.cbl.nobody@isis.visi.com>

In article <20000123092040.A2010 at stopcontact.palga.uucp>, Gerrit Holl wrote:

>> Does it have a compiler for Linux and/or windows that can produce native
>> executables?
>
>Yes, it has one. It's included in the Python distribution, Tools/freeze.
>It's called freeze, and with it, you won't need any Python interpreter.

Does freeze actually produce a native executable, or does it bundle up
bytecode with the bytecode interpreter so that all the user sees is a
single "executable" file?

-- 
Grant Edwards                   grante             Yow!  NOT fucking!! Also
                                  at               not a PACKAGE of LOOSE-LEAF
                               visi.com            PAPER!!


From MSteed at altiris.com  Wed Jan 19 16:33:06 2000
From: MSteed at altiris.com (Mike Steed)
Date: Wed, 19 Jan 2000 14:33:06 -0700
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
Message-ID: <65118AEEFF5AD3118E8300508B124877073D23@webmail.altiris.com>

> From: Fredrik Lundh [mailto:fredrik at pythonware.com]
> Sent: Wednesday, January 19, 2000 1:52 PM
> To: Evguenii Smogailov
> Cc: Python Mailing List
> Subject: Re: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
> 
> Evguenii Smogailov <sei at sunse.jinr.ru> wrote:
> > New operators:
> > 1) 'eq' (like '==');
> > 2) 'ne' (like '!=','<>');
> > 3) 'last' (like 'continue');
> > 4) '..' for generating sequences (like '1..20' or 'a..z'),
> > like in Perl.
> > 
> > Generally for improving readability,
> > (and Perl compatibility).
> 
> oh, come on.  since when is "Perl compatibility" a
> guiding principle for Python's design?
> 
> what's wrong with you folks?

Seminar posters.  I think that's what we're seeing.  They attend seminars
where they are carefully taught how to write posts that distract and annoy
the readers of the target news group/mailing list/whatever.  The seminars,
of course, are sponsored by vast conspiracies that aim to popularize severe
hackery in its various forms, such as four-letter languages beginning with
P.

Okay, I'm kidding, don't get upset.

--
M.



From Alessandro.Bottoni at think3.com  Thu Jan 27 05:47:17 2000
From: Alessandro.Bottoni at think3.com (Alessandro Bottoni)
Date: Thu, 27 Jan 2000 11:47:17 +0100
Subject: Off-topic: Looking for Ennio Simoni (a special Python specialist.
	..)
Message-ID: <6D8A17398E28D3119F860090274DD7DB498453@pces.cadlab.it>

[Excuse me for the off-topic post, but I really have no alternatives...]

I'm trying to get in touch with an old friend, and a very good Python
specialist, who worked with me in a CAD software house, Eurograph, in
Padova, Italy, in 1995/1996: Software Engineer Ennio Simoni.

Ennio, can you please contact me?

Should anybody knows how can I contact Ennio, please let me know.

(My e-mail addresses are listed below)

Many thanks for your attention.

--------------------------------
Alessandro Bottoni
(Corporate address: Alessandro.Bottoni at Think3.com)
(Personal address: alessandro.bottoni at libero.it)
Web Programmer
Think3 inc.
(www.think3.com)




From george_planansky at harvard.edu  Sun Jan  9 03:29:37 2000
From: george_planansky at harvard.edu (George Planansky)
Date: 09 Jan 2000 03:29:37 -0500
Subject: python Could not find platform dependent libraries <exec_prefix>
Message-ID: <w3yvh53d4we.fsf@gigue.eas.harvard.edu>

Using python 1.5.2, from the freeware.sgi.com Nov 99
collection, on irix 6.5, gets:

% python
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
Python 1.5.2 (#1, Aug 26 1999, 00:12:37) [C] on irix6-n32

Python is in /usr/freeware/bin, various libs are in 
/usr/freeware/lib and /usr/freeware/lib/python, 
so prefix and exec_prefix would seem to be /usr/freeware.  

Par however shows python looking under /usr/freeware/bin/lib 
for some libraries.  

Setting PYTHONHOME to various plausible values seems to make
no difference.

what's the cure?
-- 
george_planansky at harvard.edu


From nobody at nowhere.nohow  Mon Jan 24 22:52:12 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Tue, 25 Jan 2000 03:52:12 GMT
Subject: Python Rocks!
References: <000d01bf6312$77edd660$31a2143f@tim> <388BB4E1.974E63B7@prescod.net> <slrn895cqh.rar.kc5tja@garnet.armored.net> <388CDABB.EF07BA23@callware.com>
Message-ID: <slrn88q7fc.la9.nobody@isis.visi.com>

In article <388CDABB.EF07BA23 at callware.com>, Ivan Van Laningham wrote:

>> >Wouldn't it be easier if we just added an optional, (but
>> >compiler-recognized if present) block delimiter to Python 2? Then people
>> 
>> Sometimes you just have to stick up for what you believe in.  I say, "No,
>> let's not do this."
>> 
>
>I don't think we get a vote.  After all, we have Guido the Benevolent
>Dictator, don't we?  And we all know how he's going to vote (if you can
>call it vote--it's more like fiat).

Of course we get a vote.  You vote by using the language. Or not. It's
that simple. There's nothing stopping somebody from creating a
Python-like language with block delimiters.  Everybody is free to stop
using Python and start using the new lanuage.

-- 
Grant Edwards                   grante             Yow!  Will the third world
                                  at               war keep "Bosom Buddies"
                               visi.com            off the air?


From usenet-mm-yyyy at julesallen.com  Sun Jan  2 20:32:51 2000
From: usenet-mm-yyyy at julesallen.com (Jules Allen)
Date: Mon, 03 Jan 2000 01:32:51 GMT
Subject: New To Python
References: <1cRb4.10160$zp.149897@news3.mia>
Message-ID: <386FFC40.6685FD09@julesallen.com>

Hello Gary.

I can only speak for myself, of course. The best, most robust code and the
most fun I've had as an accidental programmer has come from the need to
scratch a technological itch. I think part of your problem is you're looking
to learn a language without a clear need to solve a computing problem.

That's a problem in itself.

If you're interested in building Web applications, perhaps you might want to
put together a dynamic resume password-protected Web site. Before you know it
you'll be hooking up databases or building Zope products (check out
www.zope.org!). Or write yourself an address book using Tk. Before you know
it, you'll have written a bunch of code and will be itching to make it smell
more Python-like. Or whatever -- you'll know what's missing from your making
your techno-world ideal.

Thankfully this list/newsgroup is your best friend. Newbie questions that
would be ignored by other programming language-centric support systems are
gleefully answered here. Just watch out for that Tim Peters fellow.

Or maybe you're not a programmer. Not everybody can be. It takes a particular
brand of kook to want to sit at a keyboard and work on mind bending problems.
I started programming back in the dark-ages (1980-ish) and didn't have support
network like the net to get me though the harder times. So go scratch and itch
and come back with your questions and see if this programming lark is for you!

Good luck!
Jules

--
Put the current month and year in my address to reply via email.




From s.schwarzer at ndh.net  Thu Jan  6 14:40:44 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 06 Jan 2000 20:40:44 +0100
Subject: Naming conventions
References: <3874EBF7.8317091D@ndh.net>
Message-ID: <3874EFBC.F33D03E8@ndh.net>

Hi,

Stefan Schwarzer wrote:
> are there common naming conventions for Python (functions,
> classes, variables ...)?
> [...]
> Have I forgotten something; do you have more remarks or hints?

Er, ..., one hint I found myself is the Python Style Guide,
http://www.python.org/doc/essays/styleguide.html . Any more
hints? :-)

Stefan


From thomas at bibsyst.no  Mon Jan 24 04:02:09 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Mon, 24 Jan 2000 10:02:09 +0100
Subject: Dictionary from a file
References: <388BF0C4.F0E08005@mail.bulkley.net>
Message-ID: <388C1511.ADFADFD7@bibsyst.no>

"Peter Schopfer (aka Dr. J. - magician)" wrote:
> 
> Hello
> 
> I have a dictionary that is too big to fit right in my program, so i was
> going to have it as a seperat file, labeled dic.py. how would i be able
> to retrieve this dictionary. wuld this work:
> 
> from dic.py import dictionary
> mydic=dic.dictionary
> 
> i am a newbie, so any help would be appreciated
> thanks
> 
> keagan

Or you could use a key-based database. Look into the shelve-docs in
www.python.org.

Thomas


From tiddlerdeja at my-deja.com  Wed Jan 26 19:28:41 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Thu, 27 Jan 2000 00:28:41 GMT
Subject: Date and Time
Message-ID: <86o3fj$73i$1@nnrp1.deja.com>

I'm using Python and COM. One of my 3rd party COM objects returns a Date
time thing like this:

<PyTime:"11/09/99 11:53:23 AM">

I'd like to create a Python class that returns the Date/Time like this.
How would I best mimick this? Also what is PyTime? Is it related to the
Python/COM interface?

Any help greatly appreciated.

Thanks.


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


From paul at prescod.net  Sun Jan 23 21:11:45 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jan 2000 20:11:45 -0600
Subject: Python Rocks!
References: <000d01bf6312$77edd660$31a2143f@tim>
Message-ID: <388BB4E1.974E63B7@prescod.net>

Tim Peters wrote:
> [Haskell] also has an
> optional notion of explicit block closers (in fact, the all-whitespace form
> is defined as syntactic sugar for the explicit form), but it's *so* optional
> <wink> that I don't recall ever seeing a Haskell program use explicit block
> markers (except in the language reference manual).

Wouldn't it be easier if we just added an optional, (but
compiler-recognized if present) block delimiter to Python 2? Then people
would have to stop complaining that Python is broken and instead admit
that Python programmers choose to program without delimiters.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Earth will soon support only survivor species -- dandelions, roaches, 
lizards, thistles, crows, rats. Not to mention 10 billion humans.
	- Planet of the Weeds, Harper's Magazine, October 1998





From wware at world.std.com  Mon Jan 24 11:28:05 2000
From: wware at world.std.com (Will Ware)
Date: Mon, 24 Jan 2000 16:28:05 GMT
Subject: global variable problem
References: <388c496a.9657176@122.0.0.250>
Message-ID: <FouLqt.LzD@world.std.com>

volumeregeling (volumeregeling at hotmail.com) wrote:
: I have the following :
: Debugging = 0
: def remote_debug():
:       Debugging = 1
: can you tell me how i can update the global variable from
: within a function ????  

Do this:

Debugging = 0
def remote_debug():
	global Debugging
	Debugging = 1

When a variable is used within a function definition, it is
global as long as you don't assign to it. If you assign to it,
Python assumes you want it to be local, and you must explicitly
declare it to be global. I forget why, but there is a reason
that this is a good thing.
-- 
   People say that everyone has a few skeletons in their closet. Not me.
   Well, not yet anyway. I mean, the bodies are still decomposing.
Will Ware		email: wware[at]world[dot]std[dot]com


From calishar at *NOSPAM*home.com  Tue Jan 11 00:17:36 2000
From: calishar at *NOSPAM*home.com (Calishar)
Date: Tue, 11 Jan 2000 05:17:36 GMT
Subject: Problem distributing a Gadfly app
References: <85dah2$ip8$1@nnrp1.deja.com> <387A5444.56A0EAEA@matavnet.hu>
Message-ID: <15fl7s8ho4ub8ip2ggl6sa0459k8dhmf3d@4ax.com>

On Mon, 10 Jan 2000 22:51:00 +0100, Arpad Kiss <sekter at matavnet.hu>
wrote:

>Hi,
>My workaround: 
>-i have added a carchive section to the cfg file. This carchive contains
>the sql.mar file.

Thanks Arpad, worked like a charm. I'm also using it to include a few
other things that I forgot about.

Now back off to re-read the installer docs and see about adding an
icon to the app....

Calishar



From controld at hotmail.com  Mon Jan  3 22:22:06 2000
From: controld at hotmail.com (Mike Apted)
Date: Mon, 03 Jan 2000 19:22:06 PST
Subject: OT: OpenSource Python Books?
Message-ID: <20000104032206.36493.qmail@hotmail.com>

That's funny... he has "nospam" in his e-mail address, but he seems to 
forget he has already graced us with his post and obviously does not read 
the list.

I wonder what "portion" would go back to the Python project... 1%?

>From: poet at linuxports.com
>To: python-list at python.org
>Subject: OT: OpenSource Python Books?
>Date: Tue, 04 Jan 00 03:09:01 GMT
>
>Hello,
>
>I represent LinuxPorts.Com (http://www.linuxports.com/) and I have been
>approached by a printer to possibly publish OpenSource books. To start
>these books would just be paper versions of the actual documentation.
>In this case, the Python Documentation and we were wondering if this
>would be of interest to the Python Community.
>
>I spoke with the printer and we both have agreed that if we do this
>that we will be providing a portion of the Gross Profits back to the
>individual projects. In this case the Python project.
>
>Anyway, if this of interest to you please go to
>http://www.linuxports.com/ and make your vote. It will help us
>determine the demand for such a book.
>
>Thank you for your time,
>
>Joshua Drake
>Webmaster Linux Documentation Project
>Owner LinuxPorts.Com
>poet.nospam at linuxports.com
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
>
>--
>----------- 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
>------------------------------------------------------------
>--
>http://www.python.org/mailman/listinfo/python-list

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com




From leonar_n at epita.fr  Fri Jan 14 17:31:29 2000
From: leonar_n at epita.fr (Nicolas Léonard)
Date: Fri, 14 Jan 2000 23:31:29 +0100
Subject: os.popen under Win32 Using Apache
References: <013101bf5e46$b2d83a20$d8c9fea9@desktop>
Message-ID: <85o7r0$2mup$1@news6.isdnet.net>

Ok, thanks

It works perfectly with win32pipe.popen()


Sameer Chowdhury <sameerc1 at yahoo.com> wrote in message
news:013101bf5e46$b2d83a20$d8c9fea9 at desktop...
> I came across some reference indicating that popen does not work in all
> conditions under win32.  I believe Mark Hammond includes popen2 in the
> win32all package that is supposed to replace popen under Windows.  I would
> check Python's windows page for more information.
>
> -----Original Message-----
> From: Nicolas L?onard <leonar_n at epita.fr>
> Newsgroups: comp.lang.python
> To: python-list at python.org <python-list at python.org>
> Date: Thursday, January 13, 2000 4:33 AM
> Subject: os.popen under Win32 Using Apache
>
>
> >Hit there
> >
> >
> >I have a problem when I try to use os.popen() in a cgi with Apache,
> >under Win32.
> >
> >It works in command-line mode, but don't work through Apache.
> >
> >Any idea ?
> >
> >Thanks.
> >
> >--
> >Nicolas L?onard
> >T?l : (33) 06-61-40-30-31
> >E-Mail : leonar_n at epita.fr
> >
> >
> >
>
> __________________________________________
> NetZero - Defenders of the Free World
> Get your FREE Internet Access and Email at
> http://www.netzero.net/download/index.html
>




From nickb at earth.ox.ac.uk  Fri Jan 28 07:21:22 2000
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Fri, 28 Jan 2000 12:21:22 +0000
Subject: database connectivity
References: <s90ihpl47q511@corp.supernews.com>
Message-ID: <389189C2.4F56701E@earth.ox.ac.uk>

David Metzler wrote:

> work on Windows 98/NT and on Linux. The idea is I can slip the programs in

I can help with a little info on Linux but not NT

> I also am interested on comparing MySql with Postgres. It looks like
> Postgres will only work on NT. Now I'm thinking that I would be better off
> using Linux as the server from the start and make the client program cross
> platform.

As far as I knew, I thought Postgres only ran on Unix systems !!
It certainly runs well on Linux - very impressively. So if it runs on NT aswell
!!!!

> The only problem is that now I have to find out how to make the connection
> with both windows and Linux to the Linux server. And of course I cannot take
> the project with me on a single Win 98 notebook to work on, when on the
> road.

I am playing with Postgres on Linux - accessed via Zope which lets it run on a
single Server
with access from anywhere. But of course Postgres is a distributed system
anyway so.....

> Is what I want to do possible or is Python just for "C" programmers or text
> parsing?
> For example is Python the right tool to create an inventory program?

I haven't touched C (other than for little twiddly bits) since I found Python.
It is great for large and small jobs - very neat  (with its invisible block
delimiters and indented syntax).

For an inventory prog? - you might find Zope rather neat...standard browser
frontend which everyone understands now.

just some random thoughts - hope it helps

cheers
Nick

--
Dr Nick Belshaw                                    Dept of Earth Sciences
Tel : 01865 272000                                 Oxford University
Fax : 01865 272072                                 Parks Rd, Oxford, OX2 3PR





From billtut at microsoft.com  Mon Jan 31 19:14:29 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Mon, 31 Jan 2000 16:14:29 -0800
Subject: GetObject instead of CreateObject
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BCC8E@RED-MSG-50>

What VB's GetObject does:
Well, it does two things, so I'll explain both. ;)

1) GetObject(Class := "ProgID") will connect you to an already existing
instance of a COM object.
2) GetObject("c:\blah.xls") is the COM moniker syntax.

In the latest Win32 extenstion bits at:
http://starship.python.net/crew/skippy/

There are three new functions in win32com.client to help you mimic the 
required behavior.

GetObject(): Cheesy VB double purpose crap.
GetActiveObject("ProgID"): Connects you to a running instance ONLY if it
exists.
  I think it throws a com_error exception if the ProgID isn't currently
running.
Moniker("c:\blah.xls"): Instantiates the appropriate COM object given the
passed in   
  moniker.


Bill



From m.faassen at vet.uu.nl  Thu Jan 20 08:16:20 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 20 Jan 2000 13:16:20 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com>
Message-ID: <8671r4$77k$1@newshost.accu.uu.nl>

tye4 <tye4 at yahoo.com> wrote:
> Python is good but could be better.
> Everything has flaws and nothing is perfect. If we can't fix it then its a
> different matter. But in this case the problem can be fixed. It just so 
> happens that there are a lot of bull-headed close-minded people than I
> reckoned there would be.

Excuse me? You're calling me bull-headed and closed-minded? *you* are the
person so closed-minded you can't accept indentation marks block structure.
Most *python* programmers have used other programming languages which have
explicit block delimiters, and often still use these languages in
addition to Python.

So who here is bull-headed and closed-minded about something?

> I'm pretty certain now that this problem will never be fixed.

it-won't-because-it-is-not-a-problem-ly yours,

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


From clarence at beach.silcom.com  Mon Jan 31 13:04:15 2000
From: clarence at beach.silcom.com (Clarence Gardner)
Date: Mon, 31 Jan 2000 18:04:15 GMT
Subject: Why Perl do But Python can't?
References: <200001310128.UAA02858@poverty.bloomington.in.us>
Message-ID: <z8kl4.783$S3.31411@newsfeed.avtel.net>

Kelly Lynn Martin <kelly at poverty.bloomington.in.us> wrote:
: Scripts (of any sort) are inherently unsafe when run setuid or setgid.
: See the perlsec manpage for why.

This is not true, and is a myth people need to get over.  The danger
Kelly referred to exists on some Unix-like systems and not on others.
On a system where they are safe, there is no reason whatsoever to use
a setuid wrapper.

-- 
Clarence Gardner
Software Engineer
NetLojix Communications    NASDAQ:NETX
clarence at netlojix.com


From a.eyre at optichrome.com  Mon Jan 31 13:13:11 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 18:13:11 -0000
Subject: Why Perl do But Python can't?
In-Reply-To: <z8kl4.783$S3.31411@newsfeed.avtel.net>
Message-ID: <001401bf6c16$d5657660$3acbd9c2@peridot.optichrome.com>

> : Scripts (of any sort) are inherently unsafe when run setuid or setgid.
> : See the perlsec manpage for why.
> 
> This is not true, and is a myth people need to get over.  The danger
> Kelly referred to exists on some Unix-like systems and not on others.
> On a system where they are safe, there is no reason whatsoever to use
> a setuid wrapper.

Just out of interest, what is the reason why this in unsafe. I couldn't
see the difference between a script and a binary.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From tomh at po.crl.go.jp  Wed Jan  5 20:41:16 2000
From: tomh at po.crl.go.jp (Tom Holroyd)
Date: Thu, 6 Jan 2000 10:41:16 +0900
Subject: Portable Authentication? WinNT and UNIX
In-Reply-To: <84vqlj$2e8$1@nnrp1.deja.com>
References: <84vqlj$2e8$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.10.10001061034090.6489-100000@holly.crl.go.jp>

On Wed, 5 Jan 2000 charliehubbard76 at yahoo.com wrote:

> I'm writing a simple, portable server for both WinNT and Unix in python,
> and I need to authenticate my users with the system users and passwords.
> I know that Unix provides these mechanisms, but what about WinNT?  Does
> anyone know how I can authenticate WinNT users thru Win32 api?

SRP Authenticated Sockets in Python

http://www.aist.go.jp/NIBH/~tomh/srpsocket.html

SRP stands for "Secure Remote Passphrase" -- it's an authentication
protocol that's safe to use over the network, it's patent-free, and there
are no export restrictions.  As an added bonus it creates a 160 bit shared
secret key you can use to encrypt the data sent over the socket.

The home page for SRP is http://berlin.arcot.com/srp/ or this mirror
http://srp.stanford.edu/srp/.

Dr. Tom Holroyd



From fdrake at acm.org  Mon Jan 10 13:13:28 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 10 Jan 2000 13:13:28 -0500 (EST)
Subject: Using PyString_AsString? Very Simple.
In-Reply-To: <85d6g3$f74$1@nnrp1.deja.com>
References: <85d6g3$f74$1@nnrp1.deja.com>
Message-ID: <14458.8520.61450.105303@weyr.cnri.reston.va.us>

Arinte writes:
 > Should I delete the char* that I get from PyString_AsString()???  Don't
 > readily see this in docs.

  I've already made changes so that the next documentation release
will be clear about this.
  The return values of PyString_AsString() and PyString_AS_STRING()
should *never* be freed or written to (outside of string creation);
they point to internal data which is actually part of the string
object.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From pf at artcom-gmbh.de  Mon Jan 17 05:25:19 2000
From: pf at artcom-gmbh.de (Peter Funk)
Date: Mon, 17 Jan 2000 11:25:19 +0100 (MET)
Subject: fintl.py -- a python module to aid i18n (was Re: International Python)
Message-ID: <m12A9Lb-000CxTC@artcom0.artcom-gmbh.de>

In <oqk8ldg2jp.fsf at titan.progiciels-bpi.ca> 
	=?ISO-8859-1?Q?Fran=E7ois_Pinard?= <pinard at iro.umontreal.ca> writes:
>Things are improving in the area of Python internationalisation! :-) After
>having made PO mode able to mark strings for translation in a collection
>of Python sources, I finally got `xpot' to extract those strings into a
>neater POT file.  I'm grateful to Barry Warsaw for his `pygettext' tool,
>which has been useful to cross-check the results of 'xpot' and debug it.

Seeing this subject I like to use it as a shameless plug to release
the following module, which I've written to aid the i18n of my Python apps.

I'm still seeking for hints how to mimic the locale functionality on the
WinXX and Mac platforms.  However on Linux it works as is.  Have fun.

Regards, Peter
-- 
Peter Funk, Oldenburger Str.86, D-27777 Ganderkesee, Germany, Fax:+49 4222950260
office: +49 421 20419-0 (ArtCom GmbH, Grazer Str.8, D-28359 Bremen)
---- 8< ---- 8< ---- cut here ---- 8< ---- schnipp ---- 8< ---- schnapp ----
#!/usr/bin/env python
"""i18n (multiple language) support.  Reads .mo files from GNU gettext msgfmt

If you want to prepare your Python programs for i18n you should 
add the following lines to the top of a BASIC_MAIN module of your py-program:
    try:
	import fintl
	gettext = fintl.gettext
	fintl.bindtextdomain(YOUR_PROGRAM, YOUR_LOCALEDIR)
	fintl.textdomain(YOUR_PROGRAM)
    except ImportError:
	def gettext(msg):
	    return msg
    _ = gettext
and/or also add the following to the top of any module containing messages:
    import BASIC_MAIN
    _ = BASIC_MAIN.gettext
	    
Now you should use _("....") everywhere instead of "...." for message texts.

Once you have written your internationalized program, you can use
the suite of utility programs contained in the GNU gettext package to aid
the translation into other languages.  

You ARE NOT REQUIRED to release the sourcecode of your program, since 
linking of your program against GPL code is avoided by this module.  
(Although it is possible to use the GNU gettext library by using the 
intl.so module written by Martin von L?wis if this is available, it is 
not required to use it in the  first place)
"""
# Copyright 1999 by pf at artcom-gmbh.de (Peter Funk)
#  
#                         All Rights Reserved
#
# Permission to use, copy, modify, and distribute this software and its
# documentation for any purpose and without fee is hereby granted,
# provided that the above copyright notice appear in all copies.

# ArtCom GmbH AND Peter Funk DISCLAIMS ALL WARRANTIES WITH REGARD TO
# THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
# AND FITNESS, IN NO EVENT SHALL ArtCom GmBH or Peter Funk BE LIABLE
# FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
# WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN
# AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
# OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.

_default_localedir = '/usr/share/locale'
_default_domain = 'python'

# check out, if Martin v. L?wis 'intl' module interface to the GNU gettext
# library is available and use it only, if it is available: 
try:
    from intl import *
except ImportError:
    # now do what the gettext library provides in pure Python:
    error = 'fintl.error'
    # some globals preserving state:
    _languages = []
    _default_mo = None # This is default message outfile used by 'gettext'
    _loaded_mos = {}   # This is a dictionary of loaded message output files

    # some small little helper routines:
    def _check_env():
        """examine language enviroment variables and return list of languages"""
	languages = []
	import os, string
	for envvar in ('LANGUAGE', 'LC_ALL', 'LC_MESSAGES', 'LANG'):
	    if os.environ.has_key(envvar):
		languages = string.split(os.environ[envvar], ':')
		break
        # use locale 'C' as default fallback:
	if 'C' not in _languages:
	    languages.append('C')
	return languages

    # Utility function used to decode binary .mo file header and seek tables:
    def _decode_Word(bin):
        # This assumes little endian (intel, vax) byte order.
	return  ord(bin[0])        + (ord(bin[1]) <<  8) + \
	       (ord(bin[2]) << 16) + (ord(bin[3]) << 24)

    # Now the methods designed to be used from outside:

    def gettext(message):
        """return localized version of a 'message' string"""
	if _default_mo is None: 
	    textdomain()
	return _default_mo.gettext(message)

    _ = gettext

    def dgettext(domain, message):
        """like gettext but looks up 'message' in a special 'domain'"""
	# This may useful for larger software systems
	if not _loaded_mos.has_key(domain):
	    raise error, "No '" + domain + "' message domain"
	return _loaded_mos[domain].gettext(message)

    class _MoDict:
        """read a .mo file into a python dictionary"""
        MO_MAGIC = 0x950412de # Magic number of .mo files
	def __init__(self, domain=_default_domain, localedir=_default_localedir):
	    global _languages
	    self.catalog = {}
	    self.domain = domain
	    self.localedir = localedir
	    # delayed access to environment variables:
	    if not _languages:
	        _languages = _check_env()
	    for self.lang in _languages:
		if self.lang == 'C':
		    return
		mo_filename = "%s//%s/LC_MESSAGES/%s.mo" % (
						  localedir, self.lang, domain)
		try:
		     buffer = open(mo_filename, "rb").read()
		     break
		except IOError:
		     pass
	    else:
		return # assume C locale
            # Decode the header of the .mo file (5 little endian 32 bit words):
	    if _decode_Word(buffer[:4]) != self.MO_MAGIC :
		raise error, '%s seems not be a valid .mo file' % mo_filename
	    self.mo_version = _decode_Word(buffer[4:8])
	    num_messages    = _decode_Word(buffer[8:12])
	    master_index    = _decode_Word(buffer[12:16])
	    transl_index    = _decode_Word(buffer[16:20])
	    buf_len = len(buffer)
	    # now put all messages from the .mo file buffer in the catalog dict:
	    for i in xrange(0, num_messages):
		start_master= _decode_Word(buffer[master_index+4:master_index+8])
		end_master  = start_master + \
		              _decode_Word(buffer[master_index:master_index+4])
		start_transl= _decode_Word(buffer[transl_index+4:transl_index+8])
		end_transl  = start_transl + \
		              _decode_Word(buffer[transl_index:transl_index+4])
		if end_master <= buf_len and end_transl <= buf_len:
		    self.catalog[buffer[start_master:end_master]]=\
				 buffer[start_transl:end_transl]
		else: 
		    raise error, ".mo file '%s' is corrupt" % mo_filename
		# advance to the next entry in seek tables:
		master_index= master_index + 8
		transl_index= transl_index + 8

	def gettext(self, message):
	    """return the translation of a given *message*"""
	    try:
		return self.catalog[message]
	    except KeyError:
		return message
	# _MoDict instances may be also accessed using mo[msg] or mo(msg):
	__getitem = gettext
	__call__ = gettext

    def textdomain(domain=_default_domain):
        """Sets the 'domain' to be used by this program. Defaults to 'python'"""
	global _default_mo
	if not _loaded_mos.has_key(domain):
	     _loaded_mos[domain] = _MoDict(domain)
	_default_mo = _loaded_mos[domain]

    def bindtextdomain(domain, localedir=_default_localedir):
	global _default_mo
	if not _loaded_mos.has_key(domain):
	    _loaded_mos[domain] = _MoDict(domain, localedir)
	if _default_mo is not None: 
	    _default_mo = _loaded_mos[domain]

def _testdriver(argv):
    message   = ""
    domain    = _default_domain
    localedir = _default_localedir
    if len(argv) > 1:
        message = argv[1]
	if len(argv) > 2:
	    domain = argv[2]
	    if len(argv) > 3:
		localedir = argv[3]
    # now perform some testing of this module:
    bindtextdomain(domain, localedir)
    textdomain(domain)
    info = gettext('')  # this is where special info is often stored
    if info:
	print ".mo file for domain %s in %s contains:" % (domain, localedir)
	print info
    else:
	print ".mo file contains no info"
    if message:
        print "Translation of '"+ message+ "' is '"+ _(message)+ "'"
    else:
        for msg in ("Cancel", "No", "OK", "Quit", "Yes"):
	    print "Translation of '"+ msg + "' is '"+ _(msg)+ "'"

if __name__ == '__main__':
    import sys
    if len(sys.argv) > 1 and (sys.argv[1] == "-h" or sys.argv[1] == "-?"):
        print "Usage :", sys.argv[0], "[ MESSAGE [ DOMAIN [ LOCALEDIR ]]]"
    _testdriver(sys.argv)


Gru?, Peter
-- 
Peter Funk, Oldenburger Str.86, 27777 Ganderkesee, Tel: 04222 9502 70, Fax: -60
Wer sich zu wichtig f?r kleine Arbeiten h?lt,
ist meist zu klein f?r wichtige Arbeiten.     --      Jacques Tati



From islandbase at yahoo.com  Sat Jan 29 00:16:42 2000
From: islandbase at yahoo.com (Joseph J)
Date: Sat, 29 Jan 2000 05:16:42 GMT
Subject: Web Board/Forum CGI in python.  Does it Exist?
Message-ID: <s94tshnfq0s78@corp.supernews.com>

Hi, I'm wondering if anyone has developed web board/forum type CGI scripts in 
Python.  If so, can anyone point me in the right direction.  Thanks.

Joe J.
islandbase at yahoo.com


From shae at tripoint.org  Sun Jan 30 13:10:01 2000
From: shae at tripoint.org (Shae Erisson)
Date: Sun, 30 Jan 2000 18:10:01 GMT
Subject: was Stackless 1.01 Homepage: another bug in stackless?
References: <388BE68E.B26985CA@tismer.com>
	 <$WXiIAAO+Dj4Ew2W@jessikat.demon.co.uk> <388C7D4A.D873AA@tismer.com>
	 <zMBtEAA$jIj4Ewm7@jessikat.demon.co.uk> <388C9733.B4F60DC0@tismer.com> <7PUz+AAC6aj4Ewne@jessikat.demon.co.uk>
Message-ID: <38947CE6.5ADEE391@tripoint.org>

<snip>

I installed the stackless update on win95, and suddenly hitting the
enter key in IDLE gives me two carriage returns.

Any ideas?
-- 
Shae Matijs Erisson - shae at lapland.fi
Programmeur-savant  - Python, Java, JPython


From stefan.andreatta at uibk.ac.at  Thu Jan 20 16:28:11 2000
From: stefan.andreatta at uibk.ac.at (Stefan Andreatta)
Date: Thu, 20 Jan 2000 22:28:11 +0100
Subject: read from binary file
Message-ID: <38877de8.0@sia.uibk.ac.at>

I have encountered severe problems when trying to read large amounts of data
(unsigned short integer) from a binary file under NT. I use something like:

f = 'C:\\ai.xxx'
file = open(f,'r')
buffer = file.read()

But the read() function always returns just some 1000 bytes of data and the
stops.

Does anybody know, what could be the reason for that, and how I could read
the whole file into a string?

Thanks You

Stefan




From ivanlan at callware.com  Tue Jan 25 20:52:56 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 25 Jan 2000 18:52:56 -0700
Subject: tabs do WHAT?
References: <86h5ep$iq$1@mach.vub.ac.be> <4.2.0.58.20000125114146.00c1a740@mail.aspectdv.com>
Message-ID: <388E5378.F45FD7F9@callware.com>

Hi All--

Jim Althoff wrote:
> 
> In case this thread gives the impression that
> it is Guido against the rest of the world ...
> 
> I LOVE "the whitespace stuff"!  I'm thrilled every time
> I type code and don't have to put in nasty
> syntactic delimiters (maybe I get my thrills
> too easily  :-)  ).
> 

[A glowing testimonial to the joys of clutter-free code snipped]

I couldn't agree more.  I was skeptical until I wrote my first program. 
Which took me a grand total of 23 minutes (it was not hello world,
either, but a Mayan calendar program).  I've been madly in love ever
since.

<almost-better'n-sex-and-a-dam-site-better'n-chocolate>-ly y'rs,
Ivan;-0
----------------------------------------------
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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From nolan_d at bigfoot.com  Mon Jan  3 20:03:54 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Mon, 03 Jan 2000 19:03:54 -0600
Subject: Continued attempts at Yacc-ification :)
Message-ID: <E125IOA-0001pG-00@ethereal>

I'm continuing to try writing a Yacc grammar for Python.

Today I finished a large subset of the grammar. I am now trying to
pound out any bugs before I write a lexical analyzer. I am
particularly having problems with the following two rules which I
wrote for single_input and file_input:

Grammar rules:
single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
file_input: (NEWLINE | stmt)* ENDMARKER
My whacked-up interpretation :)
single_input: '\n' | simple_stmt | compound_stmt '\n';

file_input: '\n'
| stmt
| file_input
| END;

The file_input rules don't reduce, though. My reasoning for each
branch of the rule is:

line 1: eliminate newlines in file.
line 2: Reduce statements and consider them as a file_input.
Line 3: Producing one file_input instead of lots of separate ones per
statement or newline.
Line 4: Matching END, which I've declared as a %token and my lexical
analyzer will return, and to signal the end of the file.

When I run byacc, these 4 rules don't reduce. The best explanation I
can come up with is that single_input somehow conflicts, since I can
comment it out and everything reduces. I want single_input though, since I'm also working on
an interpretter.

Can someone please show me how I've gone wrong here? Or, better yet,
if there are any pre-written Yacc grammars available for Python, which
I doubt but I thought I'd at least ask :), I'd love to see them to
 shamelessly ste...err, examine and learn from. :) Thanks.



From stiller at phil.uni-erlangen.de  Tue Jan 25 16:44:46 2000
From: stiller at phil.uni-erlangen.de (Jan Stiller)
Date: Tue, 25 Jan 2000 22:44:46 +0100
Subject: newbie: py1.52 + expat, nt
Message-ID: <86l5cm$mit$1@rznews.rrze.uni-erlangen.de>

Hi

I have sgml-files (created by Wordperfect) and want to convert them to html
via a dom.

When I write something like &uuml; (e.g.) in the sgml-file, the parser stops
at the '&' and leaves
the program gentle. I found out that &amp;uuml; works (!?) - I don't know
that much about ?ml, but
in my point of view this behavior is quite strange...

I could convert the whole file before processing it, but I'm really
interested in the 'official' way to
handle this.

thanks in advance  - Jan




From chega at my-deja.com  Sun Jan  9 17:21:24 2000
From: chega at my-deja.com (Vadim Chugunov)
Date: Sun, 09 Jan 2000 22:21:24 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <011501bf58a4$7b0db540$0100a8c0@rochester.rr.com> <3877513B.73B762F4@maxtal.com.au> <004901bf5a08$11a35be0$0100a8c0@rochester.rr.com> <387890F6.2D1136F8@maxtal.com.au>
Message-ID: <85b1kp$ts8$1@nnrp1.deja.com>


skaller <skaller at maxtal.com.au> wrote in message
news:387890F6.2D1136F8 at maxtal.com.au...
> Darrell wrote:
> >
> > From: "skaller" <skaller at maxtal.com.au>
>> ...
> > def one(arg):
> >     print 'ARGV:', arg,
> >     return 'RetVal'
> >
> > def two(arg):
> >     print 'ARGV:', arg,
> >     return 'RetVal'
> >
> > states={1:one, 2:two, 3:3}
> >
> > def switch(state, args, stateTable, default=None):
> >     val=stateTable.get(state,default)
> >     if callable(val):
> >         return val(args)
> >     return val
>
> Yes, that provides the required functionality, but has a serious
> problem: cases have to be defined 'out of line', and therefore
> do not have access to the current scope.
>
> So I'm looking for a syntax for providing the above
> functionality _inline_.

Well, you could do this:

states= {
    1: compile("x=x+1","<string>","single"),
    2: compile("x=x+2","<string>","single")
}
# and later...
x = 0
exec states[y]  # executes in locals AND modifies the calling context
print x


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


From aahz at netcom.com  Sat Jan  8 01:06:05 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jan 2000 06:06:05 GMT
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <3875099A.2CF9333C@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net> <3876B79F.30439283@mdlink.de>
Message-ID: <856k4d$7e9$1@nntp5.atl.mindspring.net>

In article <3876B79F.30439283 at mdlink.de>,
Helge Hess  <helge.hess at mdlink.de> wrote:
>Aahz Maruch wrote:
>> 
>> Not directly, no.  I think it breaks compatibility with Python's
>> extremely simple object model and I would be surprised if Guido
>> disagrees with me.
>
>Hm, I wonder why it is simpler to have two separate kinds of methods
>instead of a consistent behaviour for all methods ?

Because we're looking at it from two different perspectives.  You're
looking at the process from the bottom and I'm looking at it from the
top.  Providing your capability would make it too easy for an
inexperienced programmer to screw things up royally.

I think one point that you're missing is that double-underscore methods
in general do not follow the straight standard method paradigm; they all
have restrictions on their interface.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From tim_one at email.msn.com  Sat Jan 15 15:26:31 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 15 Jan 2000 15:26:31 -0500
Subject: Help me improve my Python Skills, please
In-Reply-To: <m3u2kfabkp.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <000301bf5f96$cefa3360$132d153f@tim>

[Michael Hudson]
> ...
> Also, you're using re where string would do. This is also
> pretty common... remember the Tom Christiansen (I think)
> quote:
>
> "When faced with a problem, some people think 'I'll use
> regular expressions.' Now they have two problems."

Fredrik Lundh first posted that quote here, attributed to Jamie Zawinski on
comp.lang.emacs.  tchrist would more likely have written "When faced with a
problem, some people think 'but if I use regular expressions, I'll have two
problems!' Script babies aren't welcome in the Republic of Perl." <wink>.

To be fair, the timbot would have written "When faced with a problem, some
people think.  Now they have three problems."

or-four-including-the-timbot<wink>-ly y'rs  - tim





From tye4 at yahoo.com  Wed Jan 19 15:24:55 2000
From: tye4 at yahoo.com (tye4)
Date: Wed, 19 Jan 2000 12:24:55 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com> <8653jg$esi@news.or.intel.com> <002401bf62b8$c30da8a0$f29b12c2@secret.pythonware.com>
Message-ID: <8656ne$h1j@news.or.intel.com>

Fredrik Lundh <fredrik at pythonware.com> wrote in message
news:002401bf62b8$c30da8a0$f29b12c2 at secret.pythonware.com...
AC "tye4" <tye4 at yahoo.com> wrote:

> As time goes on, the code using 'end' will increase substantially while
the
> legacy code without 'end' will reduce.

so you mean that if you make 'end' optional, everyone
will add extra end statements to their code?

yeah, right...

Oh yes, they will. Failure to do so will result in barrage of 'warning'
messages from the compiler that they are using a 'deprecated' language
construct. (BTW, that's how Java implements gradual change in the language
and its library.)

-tye4




From robert.meegan at murl.com  Tue Jan 18 14:18:21 2000
From: robert.meegan at murl.com (Robert Meegan)
Date: Tue, 18 Jan 2000 19:18:21 GMT
Subject: Why not use tabs? (was: Python Rocks!)
In-Reply-To: <388466D4.CC797CFC@earth.ox.ac.uk>
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk>
Message-ID: <20000118.19182100@robert.polytopic.com>

Nick -

Absolutely nothing is wrong with tabs, as long as you only use tabs. 
The problem comes when people mix tabs and spaces, because it would 
seem that every programmer on Earth uses a different default tab width 
(I'm lucky, I got in when three spaces per tab was still available).

Some people use eight-space tabs and type four spaces for an 
indentation level so that their code has indentation:

<-->
<- \t ->
<- \t -><-->
<- \t -><- \t ->

etc.

When you import this into your editor (set to three-space tabs, like 
God intended) the effect is horrific.

- Robert

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 1/18/00, 7:12:52 AM, Nick Belshaw <nickb at earth.ox.ac.uk> wrote 
regarding Re: Python Rocks!:


> >
> > Get used to it, and don't use tabs.
> >

> ! - whats wrong with tabs?

> :-)))))))))))))


> --
> http://www.python.org/mailman/listinfo/python-list






From aa8vb at yahoo.com  Fri Jan 21 11:07:15 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Fri, 21 Jan 2000 11:07:15 -0500
Subject: anydbm - a simple question
Message-ID: <20000121110715.A184793@vislab.epa.gov>

   Straight from the Effbot guide, but it fails when you try to add an element.
Any idea what it doesn't like?

  >>> import anydbm
  >>> db = anydbm.open( "database", "c" )
  >>> db[ "1" ] = "one"

  Traceback (innermost last):
    File "<stdin>", line 1, in ?
    File "/home/rhh/software/python-1.5.2/lib/python1.5/dumbdbm.py", line 111,\
      in __setitem__ 
      (pos, siz) = self._addval(val)
    File "/home/rhh/software/python-1.5.2/lib/python1.5/dumbdbm.py", line 87, \
      in _addval
      f.write('\0'*(npos-pos))
  TypeError: can't multiply sequence with non-int


-- 
Randall Hopper
aa8vb at yahoo.com



From bowman at montana.com  Sun Jan 23 22:45:55 2000
From: bowman at montana.com (bowman)
Date: Mon, 24 Jan 2000 03:45:55 GMT
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com> <86fl0m$gmj$1@inputplus.demon.co.uk>
Message-ID: <TVPi4.2485$K5.22193@newsfeed.slurp.net>

Ralph Corderoy <ralph at inputplus.demon.co.uk> wrote:

>eavesdropping on the perl5-porters mailing list.  I like well written,
>terse, books.  Given all of that it sounds to me as if I should remove
>_Programming Python_ from my shopping basket and add _Python Essential
>Reference_.  Your views are sought.

I haven't seen PER, but a local bookstore has Programming Python. After
browsing through it, I elected to leave it on the shelf. I did buy Learning
Python sight unseen; it is not too bad, but tends to mention stuff that 
will be explained later. Sometimes this has to be done, but I really think
there are too many instances in the book, and they seem to indicate poor
layout. The index is lacking also. The desktop reference is literally lacking
an index, though there is one available as errata.

In retrospect, the tutorial and library reference are the documentation for
the language. If you can pick up Perl from man pages, this is all you really
need.



From fdrake at acm.org  Wed Jan  5 10:09:43 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Wed, 5 Jan 2000 10:09:43 -0500 (EST)
Subject: Grail
In-Reply-To: <38727E76.2C7487DA@callware.com>
References: <387237F3.C35B7FA1@callware.com>
	<14450.17021.111722.710187@weyr.cnri.reston.va.us>
	<3872441A.7E86CCED@callware.com>
	<20000104220152.A3894@stopcontact.palga.uucp>
	<38727E76.2C7487DA@callware.com>
Message-ID: <14451.24247.756830.511585@weyr.cnri.reston.va.us>

Ivan Van Laningham writes:
 > Looks pretty good, but it doesn't seem to use tables.  I rely pretty
 > heavily on tables.  I could make a special version of the help without
 > tables, but I'd rather not.  I'll do it if the pysol help is all there

Ivan,
  If you rely on decent table support, don't bother with Grail.  In
all cases, Grail table support is too slow, and it's often broken in
various ways.  Handling of large tables is also bad; the Tk text
widget is built around the "everything is a character no matter how
big" model, which doesn't work when your table (one character) is
large (esp. larger than the window).
  Sorry, fixing *that* isn't in the "low two-figures" ballpark.  ;(
Otherwise, I'd seriously consider looking at the original problem (at
least for half a minute! ;).


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From thantos at chancel.org  Sat Jan  8 18:18:14 2000
From: thantos at chancel.org (Alexander Williams)
Date: Sat, 08 Jan 2000 23:18:14 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com> <38774EBA.40716D1E@maxtal.com.au>  <857jne$kcb$1@nntp8.atl.mindspring.net>
Message-ID: <slrn87fhdl.uiv.thantos@chancel.org>

On 8 Jan 2000 15:05:18 GMT, Aahz Maruch <aahz at netcom.com> wrote:

>>	for i in X: 
>>		if X[i]=e: goto Found
>>	print "Not Found"
>>	goto next
>>	Found: ...
>>	next
>>
>Why is using a flag or testing afterwards not acceptable?

Perhaps equally as important, why not use return to turn it into a
terminating event?

    >>> def searchX(lst, val):
    >>>     for e in X:
    >>>         if e == val: return 1  # Or perform some other block
    >>>                                #   of computation 
    >>>     return 0                   # Likewise

I suppose testing the resultant value of the func for results counts
as testing afterwards ... but I don't see any intuitive way of dealing
with the issue /without/ a test at some point.  In Skaller's original
code, the test occurs implicitly in the structure of the goto ... but
Python doesn't have arbitrary positional jumps like that (save for
Exceptions, which I feel have far better semantics, anyway).

Maybe I'm weird, but this feels like a "made up" issue.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From tye4 at yahoo.com  Mon Jan 17 20:32:02 2000
From: tye4 at yahoo.com (tye4)
Date: Mon, 17 Jan 2000 17:32:02 -0800
Subject: Python Rocks!
Message-ID: <860ftk$bas@news.or.intel.com>

I'm a python newbie.. but I like Python better than Perl (no ugly @$#@.. in
syntax).

Python does have one quirk: no end if or end loops

if x == 10:
    print 'x is 10'
print 'Hello world'

# easy to get confused whether this line is in 'if' block or not
Having an 'end' block would help tremendously.

if x == 10:
    print 'x is 10'
end if
print 'Hello World'

or
for x in range(1, 10):
    print x,
end loop

Indentation is a confusing and lame way to separate what is inside a block
and what's outside it.

tye4




From clarence at beach.silcom.com  Mon Jan 31 22:13:28 2000
From: clarence at beach.silcom.com (Clarence Gardner)
Date: Tue, 01 Feb 2000 03:13:28 GMT
Subject: This thread was almost about Python once
References: <z8kl4.783$S3.31411@newsfeed.avtel.net> <001401bf6c16$d5657660$3acbd9c2@peridot.optichrome.com> <20000131201258.G19725@xs4all.nl>
Message-ID: <sbsl4.794$S3.31761@newsfeed.avtel.net>

Thomas Wouters <thomas at xs4all.net> wrote:
: *) on Linux systems, passing an open fd is actually fairly easy and fairly
: compatible; old (or perhaps just all) interpreters could be started with
: '/proc/fd/3' as the script. There was talk about this on the linux kernel
: list a year or so ago (and probably more often before and after that) but
: the problem was, and probably still is, that it would break interpreters
: that rely on opening a filedescriptor right after starting and assuming it
: will be fd 3.

What interpreter is so braindead as to assume that particular file
descriptors are available when it is invoked?  And even if there are
any, catering to them is worse than all of the Microsoft backward-
compatibility-to-1985 hacks put together, IMNSHO.

-- 
Clarence Gardner
Software Engineer
NetLojix Communications    NASDAQ:NETX
clarence at netlojix.com


From news at dorb.com  Thu Jan 20 22:13:31 2000
From: news at dorb.com (Darrell)
Date: Thu, 20 Jan 2000 22:13:31 -0500
Subject: regex question
References: <roy-835F11.21172620012000@netnews.nyu.edu>
Message-ID: <K6Qh4.1395$VD4.37444@typhoon.nyroc.rr.com>

import re

pat = re.compile (r'(?:\D|^)\d(?:\D|$)')
x = pat.search ("1")
print 'X:', `x`
y = pat.search ("12")
print 'Y:', `y`
z = pat.search ("the number 1, don't you know?")
print 'Z:', `z`

######### OUTPUT
X: <re.MatchObject instance at 7f75b0>
Y: None
Z: <re.MatchObject instance at 7f8500>

--
--Darrell




From donn at oz.net  Sun Jan 16 15:11:55 2000
From: donn at oz.net (Donn Cave)
Date: 16 Jan 2000 20:11:55 GMT
Subject: IMAP - fetch using imaplib
References: <85ssc4$8jk$1@snipp.uninett.no>
Message-ID: <85t8mb$adt$0@216.39.151.169>

Quoth "Asle Pedersen" <apederse at siving.hia.no>:
| I have adressed this question to the group previoussly without any success
| so I try once more :
| I need help with the following IMAP command:
| DF56 FETCH 1:3 (RFC822.SIZE BODY[HEADER.FIELDS (FROM SUBJECT)]).
| I have tried using the imaplib for other fetch commands without trouble, but
| this one I can't seem to get working utilising the imaplib.

I have tried to help you previously without any success, so I try
once more:

DF56 and FETCH will be sent by the fetch() function.  '1:3' should be
its first parameter, and the '(RFC822...)' expression its second.  Done
that way, it works for me.

	Donn Cave, donn at oz.net


From malcolmt at smart.net.au  Thu Jan 20 22:46:39 2000
From: malcolmt at smart.net.au (Malcolm Tredinnick)
Date: Fri, 21 Jan 2000 14:46:39 +1100
Subject: regex question
In-Reply-To: <roy-835F11.21172620012000@netnews.nyu.edu>; from Roy Smith on Thu, Jan 20, 2000 at 09:17:26PM -0500
References: <roy-835F11.21172620012000@netnews.nyu.edu>
Message-ID: <20000121144639.A1454@Ridcully.home>

On Thu, Jan 20, 2000 at 09:17:26PM -0500, Roy Smith wrote:
> This isn't really a python question per-se, but it came up while using 
> the re module, so there is a connection :-)

Yes it is, don't deny it. It's ok ... you can admit you use Python here. :-)

> 
> Let's say I've got a pattern "[0-9]", i.e. a 1-digit number.  I want to 
> search a string for that pattern, and only return a match if it's found 
> preceeded by either a non-digit or the beginning of the string, and 
> likewise followed by either a non-digit or the end of the string.  Is 
> there a single compact regex I can write to cover that?

One method would be to use the (?!...) construct, which matches the preceeding
reg-exp if it's not followed by what's in the brackets. So you could try:

	[0-9](?![0-9])|[0-9]$

(The second alternative is necessary because the (?!...) needs to match
something. At least, I *think* this is true and I don't have Python on my
current machine to test it. Hmm .. have to remedy that problem.)

> The best I can come up with would be something like:
> 
> (^[0-9][^0-9])|([^0-9][0-9][^0-9])|([^0-9][0-9]$)|(^[0-9]$)

Ugh!

> This uglyness is exaccerbated by the fact that 
> I'm not really looking for 1-digit strings, I'm looking for 12-digit hex 
> numbers (i.e. ethernet MAC addresses), where the basic pattern is:
> 
> [0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-f
> A-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f][0-9a-fA-f].

I'm not 100% up to speed on MAC addresses, but if they really are just 12
digit hex strings, why not use:
	[0-9a-fA-F]{12, 12}
as your pattern?

If you want to save each digit in a group, put parentheses around the [...]
bit. It can also probably be improved by just using case-insensitive matching
(if you are just matching this expression). When you compile the pattern, add
the re.I flag to it.

Hope this helps, or at least, doesn't hinder too much.

Cheers,
Malcolm Tredinnick

--
Eagles may soar, but weasels don't get sucked into jet engines.




From sparhawk at aon.at  Fri Jan 21 12:31:23 2000
From: sparhawk at aon.at (Gerhard Gruber)
Date: Fri, 21 Jan 2000 18:31:23 +0100
Subject: How to submit a module
Message-ID: <aj5h8so5eljvnnqjqgekeq4f5pl8de5u11@4ax.com>

When I had to do a project using CGI and HTML we used perl with a module named
FastTemplate. This module was very handy because you can define templates with
variables and substitute these variables with text. Thus you can easily create
dynamic HMTL pages or also any other kind of text files where you can use
templates (like printer masks, predefined email sendings, etc.). Since I
prefer python I have been looking for a similar module but didn't find one, so
I wrote a class that does exactly this. Now I thought that I would want to
contribute this module to the python community because I think that it might
be usefull to other programmers as well, but how do I go about it?

-- 
Bye,
   Gerhard


From achatham at hotmail.com  Sun Jan  2 00:55:02 2000
From: achatham at hotmail.com (achatham at hotmail.com)
Date: Sun, 02 Jan 2000 05:55:02 GMT
Subject: C Extension Questions (Binary Trees)
Message-ID: <84mobo$4d9$1@nnrp1.deja.com>

I have been trying to work on an extension module in C, binary
(search) trees. I know that you can do most of that stuff with lists or
dictionaries, and I'm sure that someone has already created a tree data
type, but I'm doing it mostly for practice (I have trouble coming up
with my own examples). It has certainly given me a greater appreciation
of how things work in Python: I now know a whole lot more about
reference counting, attribute lookup, etc. And I have much greater
respect for the built-in dictionary type (no matter how contrived my
test cases are, a dictionary is still faster than my search trees by
10-20%).

But I have some questions that I haven't been able to clear up:

Question 1. I keep getting warnings from GCC (Redhat 6.1) such as:
"tree.h: 34: warning: initialization from incompatible pointer type"
I am not very fond of this warning, though things seem to work fine. The
relevant parts of tree.h are:

--------------------
typedef struct {
  PyObject_HEAD
  PyObject *left;
  PyObject *right;
  PyObject *parent;
  PyObject *key;
  PyObject *data;
  PyObject *current;
  PyObject *func;
  int nodes_in_tree;
} treeobject;

staticforward PyTypeObject TreeType;

static PyObject *tree_insert(treeobject *self, PyObject *args);
static PyObject *tree_search(treeobject *self, PyObject *args);
static PyObject *tree_issibling(treeobject *self, PyObject *args);
static PyObject *tree_successor(treeobject *self, PyObject *args);
static PyObject *tree_predecessor(treeobject *self, PyObject *args);
static PyObject *tree_remove(treeobject *self, PyObject *args);
static PyObject *tree_map(treeobject *self, PyObject *args);
static PyObject *tree_as_tuple(treeobject *self, PyObject *args);

static struct PyMethodDef tree_methods[] =
{
  {"insert", tree_insert, METH_VARARGS},
  {"issibling", tree_issibling, METH_VARARGS},
  {"search", tree_search, METH_VARARGS},
  {"succ", tree_successor, METH_VARARGS},
  {"pred", tree_predecessor, METH_VARARGS},
  {"remove", tree_remove, METH_VARARGS},
  {"map", tree_map, METH_VARARGS},
  {"datatuple", tree_as_tuple, METH_VARARGS},
  {NULL, NULL}
};
-------------------

The warnings are from the assignments in tree_methods. Every one (except
{NULL, NULL}) generates that warning. How can I get rid of it?

Question 2: Is there a way to override "map"? I created a nice little
__getitem__ function for the tree class which works by getting the
leftmost node in the tree if the index is 0, or the successor to the
last item if the index > 0, so I can do something like "for x in T:
print x.data" to print out all the data elements in the tree using an
inorder traversal. The only problem is that repeatedly calling successor
is not a very efficient way to do an inorder traversal of a tree. For
__getitem__ I really don't have a choice about doing it that way, since
it's hard to store the call stack necessary to do the recursive
traversal between calls of __getitem__. But for "map", it would make
more sense to do that, since it's a single call. But from what I can
tell, map just called __getitem__ on its own until it gets an
IndexError. This isn't such a big deal, since I just made a member
function "map", but it'd be nice to override the real one.

Question 3: A minor question. How do I get member variables to show up
when I do "dir(T)"? All I get is a list of methods, when I know that
T.left is a valid reference (it works fine). Or maybe I'm just too tired
and "dir()" doesn't show variables normally.

Again, it's not like I really _need_ this thing working, but I'd like to
at least get these things straight. And so far the files for this data
structure have turned out larger than any single Python program I've
ever written! Yeah, I'm a newbie :)

Thanks,
Andrew


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


From aahz at netcom.com  Tue Jan 11 11:56:20 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 11 Jan 2000 16:56:20 GMT
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov>
Message-ID: <85fnbk$j6h$1@nntp9.atl.mindspring.net>

In article <20000111113656.A42270 at vislab.epa.gov>,
Randall Hopper  <aa8vb at yahoo.com> wrote:
>Doug Hellmann:
> |
> |Could one of you post a brief description of what a multimethod is, or
> |maybe just a pointer to a definition somewhere?
>
>I had the same question yesterday and did a little surfing.  Basically,
>instead of switching off just the type of 'self' to determine which method
>to call, you switch off the type of other arguments as well.

IOW, similar to C++-style multiple dispatch?
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From mikael at isy.liu.se  Fri Jan 21 12:58:02 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 21 Jan 2000 18:58:02 +0100 (MET)
Subject: Congratulations, Tye!
In-Reply-To: <20000121165652.A843@stopcontact.palga.uucp>
Message-ID: <XFMail.000121185802.mikael@isy.liu.se>

On 21-Jan-00 Gerrit Holl wrote:
 >  Congratulations on letting Ivan say RTFM!

>From what I've seen so far in this newsgroup, I guess that should be 
interpreted as "Read the fine manual".

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    21-Jan-00
Time:    18:55:37

This message was sent by XF-Mail.
-----------------------------------------------------------------------



From dworkin at ccs.neu.edu  Wed Jan 19 23:24:38 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 19 Jan 2000 23:24:38 -0500
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: Evguenii Smogailov's message of "Wed, 19 Jan 2000 23:37:45 +0300 (MSK)"
References: <Pine.SOL.3.91.1000119233305.25615A-100000@sunse>
Message-ID: <vndr9fd1ibd.fsf@betelgeuse.ccs.neu.edu>

Evguenii Smogailov <sei at sunse.jinr.ru> writes:

> Generally for improving readability,
> (and Perl compatibility).

That's not an and, that's an xor.

-Justin

 



From vivek_kapadia at REMOVE_writeme.com  Tue Jan 25 14:45:11 2000
From: vivek_kapadia at REMOVE_writeme.com (Vivek)
Date: Tue, 25 Jan 2000 11:45:11 -0800
Subject: Python on VxWorks..
Message-ID: <388DFD47.8328A9DF@REMOVE_writeme.com>

Hi,

I am just getting started with Python.. Can somebody point me to info on
working with Python and C++ together on VxWorks environment.

Thanks in advance.

Vivek.


From emile at fenx.com  Wed Jan 19 22:57:53 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 19 Jan 2000 19:57:53 -0800
Subject: Mapping Column Names to results using DBI
References: <38863BEB.BCB9E15C@exceptionalminds.com>
Message-ID: <035a01bf62fa$87ad70a0$01ffffc0@worldnet.att.net>

I normally end up cut and pasting the column names ala:

    select namefield, addressfields, emailfield from contacts

then:

    namefield, addressfields, emailfield = returned_record

This gets the values into the named fields, and is a lot easier than
dealing with index numbers, but you've got to remember to change both
places when you change things around (or get the unpack error ;-).

I've been meaning to ask this as well, or play with other structures,
but haven't had the time yet.

Emile van Sebille
emile at fenx.com
-------------------


----- Original Message -----
From: Timothy Grant <tjg at avalongroup.net>
To: Python People <python-list at python.org>
Sent: Wednesday, January 19, 2000 2:34 PM
Subject: Mapping Column Names to results using DBI


> Hi again,
>
> Right now I have an app that works ok, but I forsee a problem in the
> future.
>
> As you know, the results from a database query are returned in a
tuple.
> I can access those results by position just fine, is it possible to
map
> the column names onto the results, so I can access columns by name
> rather than by number?
>
> While indexing by number works, it is incredibly fragile and will
start
> failing the minute the underlying DB changes.
>
> Once again, it appears to me from the DBI spec that there should be a
> way to do this, I just can't figure it out from the spec.
>
> I'm using mxODBC if that helps.
>
> Thanks.
> --
> Stand Fast,
>     tjg.
>
> Chief Technology Officer              tjg at exceptionalminds.com
> Red Hat Certified Engineer            www.exceptionalminds.com
> Avalon Technology Group, Inc.                   (503) 246-3630
> >>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<
>
> --
> http://www.python.org/mailman/listinfo/python-list
>





From alet at unice.fr  Mon Jan 24 04:41:14 2000
From: alet at unice.fr (Jerome ALET)
Date: Mon, 24 Jan 2000 10:41:14 +0100
Subject: PIL Problem (Python Imaging Library)
Message-ID: <388C1E3A.2CF9F1FF@unice.fr>

Hi,

Does someone know if the standard PIL library can allow me to 
extract many graphic files (JPEG in my case) from a special undocumented
format 
archive ?

the only thing I know is the offset at which begins each jpeg file, but
I don't know
their sizes, and I don't want to analyze the jpeg format to finally
discover the size (in fact I want PIL to do it for me).

I've made a very small patch to PIL which works and allows me to do the
following (say the jpeg files are all on 256 bytes boundaries):
--------------------
        file = open(filename, "rb")	# the 'b' is only for portability to
winXX
	# ...
        szbuf = 256     # all included documents seem to be on a 256
bytes boundary
        count = 0
        eof = 0
        while not eof :
                jpegpos = file.tell()
                read = file.read(szbuf)
                if len(read) :
                        if isjpeg(read) :	# test if the read chunk
ressembles a jpeg file
                        	count = count + 1
                                outfilename = ("%02i" % count) + '.jpeg'

                                newpos = file.tell()
                                file.seek(jpegpos, 0)

				# I know I'm at the beginning of a graphic file
				# but I don't know the graphic file's size
                                image = Image.open(file)

				# fortunately the save function loads the image in
				# memory before saving, so I don't have to know its size
                                image.save(outfilename, quality = 100)

                                file.seek(newpos, 0)
				print "\t" + outfilename, "\tSize:", image.size
                else :
                        eof = 1
        file.close()
--------------------

but I wanted to know if there's a standard way with PIL to do exactly
this without any patch
(and without knowing the included jpeg filesizes) in order for me to be
able to redistribute my program without worrying people about having the
patched PIL. I suppose if another way exists then it should be without
calling Image.open because this function assumes the graphic file begins
at offset 0 which is false in my case. (the same for the save function).

Please could you CC any answer to my email address sine I don't read
newsgroups very often ?

-- 
Jerome ALET - alet at unice.fr - http://cortex.unice.fr/~jerome


From nhv at cape.com  Thu Jan 13 13:15:46 2000
From: nhv at cape.com (Norman Vine)
Date: Thu, 13 Jan 2000 13:15:46 -0500
Subject: Python and MapObjects (Was: Python and ArcView GIS)
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>   <Pine.SGI.3.95.1000112172541.19328B-100000@clavey.des.ucdavis.edu>   <85kci6$i9l$1@newsie2.cent.net> <85kq6t$4a4$1@newsserver.rrzn.uni-hannover.de>
Message-ID: <85l4p2$s4h$1@newsie2.cent.net>

Bernhard Reiter wrote

> "Norman Vine" <nhv at cape.com> writes:
>
> > but there is the OpenSource mapserver project
> > that works with shapefiles and has a SWIGed interface
>
> > http://mapserver.gis.umn.edu/
>
> Mapserver still has to clarify its license (though they are aiming
> towards BSD style, AFAIK.)
>

Here is the license from the current mapserver distribution

seems to be Python compatable to me

:)

Norman



LICENSE

Copyright (c) 1996-1999 Regents of the University of Minnesota.

Permission is hereby granted, free of charge, to any person obtaining a copy
of
this software and associated documentation files (the "Software"), to deal
in
the Software without restriction, including without limitation the rights to
use,
copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
the
Software, and to permit persons to whom the Software is furnished to do so,
subject
to the following conditions:

The above copyright notice and this permission notice shall be included in
all
copies of this Software or works derived from this Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS
FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS
OR
COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
WHETHER
IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.




From loewis at informatik.hu-berlin.de  Thu Jan 20 07:57:02 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Jan 2000 13:57:02 +0100
Subject: [Q] What is needed to use other TclTk Widget's within Python ...
References: <863m5q$3g$1@bw107zhb.bluewin.ch>
Message-ID: <p6qzou0c34x.fsf@informatik.hu-berlin.de>

"Markus Meng" <meng.engineering at bluewin.ch> writes:

> Do I need for each of these package an additional wrapper like Tkinter?

Not necessarily. You can emit Tcl commands as plain text if you wish,
using the .call method of the tkapp object; see Tkinter.py for examples.

You probably need to load the extensions before using them, using
Tcl's 'package require' command.

> Is it possible to write the GUI-Client completly in TclTk and then
> connect to the Python server through sockets?

You can create an interpreter, and pass the entire script to the Tcl
interpreter, using the tkapp's .eval method.

> Do we have other possibilities for bringing the 'best' of both
> worlds together?

I don't know, what else do you want?

Regards,
Martin


From davidm at softdesigns.com  Thu Jan 27 08:25:46 2000
From: davidm at softdesigns.com (David Metzler)
Date: Thu, 27 Jan 2000 08:25:46 -0500
Subject: database connectivity
Message-ID: <s90ihpl47q511@corp.supernews.com>

I love what I have seen so far of Python. I am in serious need of guidance
however. I spend a couple of nights till the wee hours and have run into a
block. What I want to do is create some python database programs that will
work on Windows 98/NT and on Linux. The idea is I can slip the programs in
at work on Windows and no one will notice. Then I can swap out Windows for
Linux in some places and no one will notice.
This is what I found:
At ftp://starship.python.net/pub/crew/jbauer I got MySQL.pyd and
libmySQL.dll and copied them to my Python directory and was able to make a
connection on Win98. Thanks Jeff. It took a message from Jeff to find the
files because the precompiled version was not mentioned on the web site that
I could find. The first thing I tried however failed. I tried c=db.cursor()
and it did not work. I figure that cursor is not supported. I'm do not know
enough to know if this will be a problem in the future. Also I do not know
where to find what the features that are supported are.

I then tried Andy Dustman's stuff. I downloaded his package but found that
there was no precompiled version for Windows and I am not a "C" programmer
and do not have VC 5.0.

I took a look at ODBC but it seems to be for Windows only and I a looking
for Windows/Linux.
MxODBC looks like an interesting product but the author does not support
windows directly and the license would require me to talk to the lawyers. I
cannot do that because this project not being officially approved would mean
that they would ignore me.

I also am interested on comparing MySql with Postgres. It looks like
Postgres will only work on NT. Now I'm thinking that I would be better off
using Linux as the server from the start and make the client program cross
platform.
The only problem is that now I have to find out how to make the connection
with both windows and Linux to the Linux server. And of course I cannot take
the project with me on a single Win 98 notebook to work on, when on the
road.

I can spend some money on this and am willing to buy stuff if it will help.
A tutor from Connecticut would be nice, MS VC, a book. I can pay for tools
or training but the license cannot limit deployment as I have limited
control over that. Of course if I get the database question answered I'll
have to work on the graphical cross-platform next.

Is what I want to do possible or is Python just for "C" programmers or text
parsing?
For example is Python the right tool to create an inventory program?










From aahz at netcom.com  Tue Jan 18 01:43:26 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jan 2000 06:43:26 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
Message-ID: <86122e$1g0$1@nntp9.atl.mindspring.net>

In article <860ftk$bas at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>
>I'm a python newbie.. but I like Python better than Perl (no ugly @$#@.. in
>syntax).
>
>Python does have one quirk: no end if or end loops

Yup.  Have you ever tried reading code from three different programmers,
each of whom had a different idea of the proper location for end of
block?
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From robin at jessikat.demon.co.uk  Wed Jan  5 08:49:24 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 5 Jan 2000 13:49:24 +0000
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
In-Reply-To: <20000105082703.A9441@vislab.epa.gov>
References: <841d0c$j0n$1@violet.singnet.com.sg>
 <387f6d37.25522789@news.demon.co.uk> <19991230065516.A778039@vislab.epa.gov>
 <W8TkoEATH4a4Ewpu@jessikat.demon.co.uk> <387378c0.4549271@news.demon.co.uk>
 <xc1BqAAkwpb4EwWW@jessikat.demon.co.uk> <20000105082703.A9441@vislab.epa.gov>
Message-ID: <uBCWdBAkv0c4Ewdy@jessikat.demon.co.uk>

In message <20000105082703.A9441 at vislab.epa.gov>, Randall Hopper
<aa8vb at yahoo.com> writes
>Robin Becker:
> |Is there an obvious way to allow an extension module to write messages to
> |various fd type objects?
>
>Do you mean, change a module that's written to dump to stdout or stderr so
>that it writes to objects?  You can always replace sys.stdout/stderr
>before/after calling the extension module.  Not pretty, but it works.
>
> |Also under certain extreme circumstances my code would like to report a
> |total failure eg attempt to divide by error. I would like to ask more
> |experienced pythoneers how arithmetic errors are handled using built in
> |IEEE exceptions and how does one gracefully crash a python script? (eg
> |an abend/abort).
>
>raise ArithmeticError, "Oops"
>raise ZeroDivisionError, "Ouch"
>
>?
>
yes that's python, but what's the best way to do it in an existing C
library?
-- 
Robin Becker



From parkw at better.net  Sun Jan 16 18:25:02 2000
From: parkw at better.net (William Park)
Date: Sun, 16 Jan 2000 18:25:02 -0500
Subject: If you can only buy *one* Python book...
In-Reply-To: <N%rg4.4360$3b6.20735@ozemail.com.au>
References: <N%rg4.4360$3b6.20735@ozemail.com.au>
Message-ID: <20000116182502.A1106@better.net>

On Sun, Jan 16, 2000 at 10:55:41PM +0000, Jason Stokes wrote:
> Which one should you get?  I quite like the look of the New Riders
> Python Reference, but is it comprehensive enough?  I already know Perl,
> and I actually prefer learning from reference works.  Suggestions?
> 
> -- 
> Jason Stokes: jstok at bluedog.apana.org.au
> -- 
> http://www.python.org/mailman/listinfo/python-list

Read the online 'tutorial'.  After that you have many choices:
    - Learning Python
    - Quick Python Book
    - etc.
I have printout of all online docs, except 'library'.  But, I still
prefer and recommend a bound book.  I refer to 'language' and 'library'
references only when I have syntax or library problems.

William Park



From mal at lemburg.com  Thu Jan 20 13:44:22 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 20 Jan 2000 19:44:22 +0100
Subject: DBI spec
References: <3884AFF2.9BE7773D@exceptionalminds.com>
Message-ID: <38875786.9D28382A@lemburg.com>

Timothy Grant wrote:
> 
> More questions regarding the DBI spec:

These should go to the DB DIG, not the main list...
 
> First, is it possible to walk backward through a cursor? All the
> fetchXXX() functions appear to only move forwards, however, one of my
> SQL books mentions the ability to move backwards as well as forwards.

Cursors are not widely supported by the different DB interfaces.
mxODBC is one that does allow naming cursors. Still, there
are no specific methods to walk in different directions. You
DB may have support for this through SQL statements though.
 
> Second, and very much related to the first, how are the results of a
> cursor query stored, are they stored in temporary storage on disk, are
> they loaded in memory, etc.

Depends on the driver. The Python interfaces only fetch the
data from the driver (and then create Python datatypes out of
it).
 
> Third, at the moment, given the current size of the results that my
> queries retrieve, I can implement the backward cursor movement by using
> fetchall() and traversing the resultant list, however, I can forsee
> that, given a large result set, that method will cause grief, how have
> others resolved the bi-directional cursor traversal issue?

I'd suggest playing some tricks with VIEWS and auto-increment
columns... depends on the database how exactly that would
have to be done. Stored procedures are also a nice way to
implement these things, but again very DB dependent.

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




From robin at jessikat.demon.co.uk  Tue Jan 18 05:31:12 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 18 Jan 2000 10:31:12 +0000
Subject: WIN32 : how to list the current processes by their executable na me?
References: <326536345498D311B3BC00105A39802A07457E@newsintern.dspace.de>
Message-ID: <CxKdRDAwDEh4Ewzj@jessikat.demon.co.uk>

In article <326536345498D311B3BC00105A39802A07457E at newsintern.dspace.de>, Stefan Migowsky
<smigowsky at dspace.de> writes
>Hi,
>
>see the example given by Mark Hammond in the win32 package
>
>...\Python\win32\scripts\killProcName.py
>
>Stefan
>
...
this fails for me with win95.

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-2000 Mark Hammond (MHammond at skippinet.com.au)
>>> import win32api, win32pdhutil, win32con, sys
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "C:\Python\win32\lib\win32pdhutil.py", line 21, in ?
    import win32pdh, string, win32api
ImportError: No module named win32pdh
>>> 
-- 
Robin Becker


From mfletch at tpresence.com  Fri Jan 21 21:31:49 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 21 Jan 2000 21:31:49 -0500
Subject: Minor inconsistency between cStringIO and (StringIO or normal fil
	es)...
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3092@VRTMAIL>

Following demonstrates the inconsistency, essentially, a seek to maxint
raises a memory error on a cStringIO object, whereas the StringIO and file
objects just move to the end of the file for further actions (which is very
handy).  Is there a better way to spell "move to end of file" (keeping in
mind that this is for a class doing effectively random access in the
buffer)?

Cheers,
Mike

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at simporkippinet.com.au)
>>> import cStringIO, sys
>>> d = cStringIO.StringIO()
>>> d.seek(sys.maxint)
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
MemoryError: 
>>> d = open( 'z:\\temp\\test.wrl','r')
>>> d.seek(sys.maxint)
>>> d.tell()
2147483647
>>> import StringIO
>>> d = StringIO.StringIO()
>>> d.seek(sys.maxint)
>>> d.tell()
2147483647
>>> 

__________________________________
 Mike C. Fletcher
 Designer, VR Plumber
 http://members.home.com/mcfletch



From tye4 at yahoo.com  Thu Jan 20 17:04:42 2000
From: tye4 at yahoo.com (tye4)
Date: Thu, 20 Jan 2000 14:04:42 -0800
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com> <867t9d$aoi$1@mordred.cc.jyu.fi>
Message-ID: <8680qp$e9t@news.or.intel.com>

Brilliant idea. Thanks. That's a lot less work than I had planned.

-tye4

Sami Hangaslammi <shang.remove_edu at st.jyu.fi.edu> wrote in message
news:867t9d$aoi$1 at mordred.cc.jyu.fi...
> > I intend to modify the language's syntax to add
> > 'end' delimiters for the blocks.
>
> Wouldn't it be better (and easier) to write a script that converts
> between .py and .pye (python with end delimeters) scripts. This would
> allow you to both use existing Python libraries and distribute modules
> you have made to others.
>
> --
> Sami Hangaslammi
> shang (at) st (dot) jyu (dot) fi
>
>




From aahz at netcom.com  Fri Jan 21 10:36:53 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 15:36:53 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de>
Message-ID: <869uel$suj$1@nntp1.atl.mindspring.net>

In article <m3n1q0fzg7.fsf at greebo.nodomain.de>,
Bernhard Herzog  <herzog at online.de> wrote:
>
>curried functions

All right, I've seen it enough times this week, would somebody mind
giving the Stupid Person's definition of currying?
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From michel at digicool.com  Wed Jan 19 18:06:57 2000
From: michel at digicool.com (Michel Pelletier)
Date: Wed, 19 Jan 2000 18:06:57 -0500
Subject: def args idiom and signatures
References: <02b001bf629c$d7ce3960$01ffffc0@worldnet.att.net>
Message-ID: <38864391.97AB0AE6@digicool.com>

Emile van Sebille wrote:
> 
> In /F's Standard Python Library I spotted:
> 
>     def _d(y, m, d, days=(0, 31, 59, ..., 334, 365)):
> 
> where the balance of the function relies on days having this value.
> 
> Trying to understand why it may be an argument vs defined within
> the function, I suspect that as the def is only processed to bytecode
> once, whereas an in-line assignment would be re-processed with
> each execution, that the decision is performance based.

You are correct.  There are caveats, however, for instance, it is highly
recommened that these objects be used in a read-only fasion in a
multi-threaded program (or use locks).
 
> Now, in light of the various discussions pertaining to function
> signatures, it seems that this idiom presents a particular problem.

What problem is that?
 
> Various munging of the name or introduction of a static type keyword
> may help address this, but I'm curious if this common usage and
> what people's thoughts are.

Zope uses this all time time, I find it quite useful.

-Michel


From phd at phd.russ.ru  Wed Jan 26 11:51:04 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 26 Jan 2000 16:51:04 +0000 (GMT)
Subject: Keywordargument in string.split
In-Reply-To: <1D9EA098C@nds-3.azg.nl>
Message-ID: <Pine.LNX.4.21.0001261650310.763-100000@fep132.fep.ru>

On Wed, 26 Jan 2000, P.J.W.S. Vrijlandt wrote:
> It seems that the string module I import, is not the string.py in the 
> standard distribution but a not-quite-equivalent c-implementation of 
> it.

   But of course! Read the very string.py and find strop there...

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From philh at vision25.demon.co.uk  Thu Jan 13 10:57:46 2000
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 13 Jan 00 15:57:46 GMT
Subject: Python Pre-processor (like php/asp)
References: <387D7349.DCF6567A@sprint.ca>
Message-ID: <947779066snz@vision25.demon.co.uk>

In article <387D7349.DCF6567A at sprint.ca>
           ovazquez at sprint.ca "Orlando Vazquez" writes:
> Hi I'm looking for a Python equivalent of php or asp. I like their basic
> ideas but I dislike the languages. Is there any module or set of modules
> for this purpose? (I've seen one on the snippets site but it's kind of
> bare)

I read an article in a recent Dr Dobbs Journal about a Python Server
Pages. I've not tried it, but it looked cool.

-- 
Phil Hunt - - - phil at comuno.com

 See Comuno's Linux coverage at <http://www.comuno.com/linux/> for:
* Links to useful sites * Linux-related news * Open source software *



From JamesL at Lugoj.Com  Wed Jan 19 00:24:25 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Tue, 18 Jan 2000 21:24:25 -0800
Subject: Using tabs effectively. Was: Why not use tabs?
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <20000118.19182100@robert.polytopic.com>
Message-ID: <38854A89.B3FBFA12@Lugoj.Com>

Robert Meegan wrote:
> Absolutely nothing is wrong with tabs, as long as you only use tabs.
> The problem comes when people mix tabs and spaces, because it would
> seem that every programmer on Earth uses a different default tab width
> (I'm lucky, I got in when three spaces per tab was still available).
> 
> Some people use eight-space tabs and type four spaces for an
> indentation level so that their code has indentation:
> 
> <-->
> <- \t ->
> <- \t -><-->
> <- \t -><- \t ->
> 
> etc.
> 
> When you import this into your editor (set to three-space tabs, like
> God intended) the effect is horrific.

Under Solaris and Linux (and probably most other Unices) there exist the
commands:

expand
unexpand

Read their man pages. Note that they take an option about how wide tabs are
to be interpreted. Also note that some versions (BSD?) of the "pr" command
also have the "-e" option for setting tab width.

The issue of "how many spaces to a tab" is an old enough problem *in all
languages* that I'm surprised that it has never become a convention to put a
comment at the top of all source files indicating what the tab settings for
decent viewing should be. I've come across a lot of source code where tabs
were intended to be every 4 characters to conclude this is a sizable number
of authors. (I used to use 3 spaces/tab early in my career.) The fact that
some languages like Python are sensitive to the issue detracts from them not
a wit in my opinion.

I humbly offer the following rules to a long and healthy editing experience:

1.0) If using tabs (regardless of tab setting):
   1.1) Use nothing but tabs for indentation!
        Do not mix tab and space indentation!
   1.2) Indent at whole multiples of your tab settings lest you make it
        impossible to conform to rule 1.1.
   1.3) Place a comment at the top of your file indicating the tab
        spacing. Even if using tab settings at 8.
   1.4) Run the unexpand utility (if on Unix) on the file to minimize
        the chance of accidentally violating rule 1.1.
2.0) If using spaces:
   2.1) Use nothing but spaces for indentation!
        Do not mix space and tab indentation!
   2.2) Place a comment at the top of your file indicating that tabs
        do not exist as indentation and should not exist.
   2.3) Run the expand utility (if on Unix) on the file to insure
        that you have not accidentally violated rule 2.1 and 2.2.
3.0) If trying to impose a particular tab setting as standard:
   3.1) Try using a thimble to turn back the tide, as you will have better
        luck at success at that than in imposing a specific tab spacing
        as "correct".

Rules to follow for extra credit:
4.0) For lines longer than 80 characters:
   4.1) Keep all lines under 80 characters and manually insert line
        breaks at reasonable points. Do not rely on automatic editor
        line wraps.
   4.2) If reasonably possible, (and it normally is) indent line
        continuations to at least the same indentation level as the
        continued line. (Much like the sentences in these paragraphs
        have been "continued" at the same indentation level.)

5.) Don't expect any of these rules to be followed except this one.


P.S. I sure hope Netscape Communicator didn't convert any of my spaces to
tabs! ;)


From bruce_dodson.nospam at bigfoot.com  Tue Jan 11 19:02:26 2000
From: bruce_dodson.nospam at bigfoot.com (Bruce Dodson)
Date: Tue, 11 Jan 2000 20:02:26 -0400
Subject: Python and ArcView GIS
Message-ID: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>

If anyone is interested in this combination, I have an ArcView 3.x extension
which embeds the Python interpreter within ArcView.  The benefits are
obvious: access to COM objects and Python modules, plus a full-featured
language, more suitable for large applications than ArcView's native
scripting language.  In case anyone is wondering if this is spam, the
extension is not for sale.  I'll be putting it on some public web page soon,
but first I am hoping to find some Python hackers who are also ArcView gurus
to put it through its paces.  Are there of those out there, or am I the only
one? :-)

Sincerely,
Bruce



From landauer at apple.com  Mon Jan 24 19:29:00 2000
From: landauer at apple.com (Doug Landauer)
Date: Mon, 24 Jan 2000 16:29:00 -0800
Subject: OFFTOPIC: *-ly y'rs,
References: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com> <3dvh4jz0m9.fsf@amarok.cnri.reston.va.us>
Message-ID: <landauer-FCBE43.16290024012000@news.apple.com>

In article <3dvh4jz0m9.fsf at amarok.cnri.reston.va.us>, 
akuchlin at mems-exchange.org (Andrew M. Kuchling) wrote:
>
> The first occurrence, at least in the python-list archives, was [...]
> 	-- Tim Peters, 29 Dec 91 

I recall seeing it during the first half or so of my
ten years at Sun.  That woulda been somewhere between
1986 and 1990.  The guy there who used it most often
said it was a common signoff at Xerox PARC in the early
80's.  That version didn't include the "y'rs" part, though,
making it just that wee bit more crpytic.  

Archeo-logic-ly,
-- 
    Doug Landauer        landauer at apple.com (work)
                         landauer at scruznet.com (not-work)


From gmcm at hypernet.com  Thu Jan 20 22:51:00 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 20 Jan 2000 22:51:00 -0500
Subject: DBI Commentary
In-Reply-To: <3887BB31.A9B8E90F@exceptionalminds.com>
Message-ID: <1263695833-12432931@hypernet.com>

Timothy Grant writes:
[snip]
> First, embedding SQL in Python, even using the """SQL""" method is
> really ugly especially while keeping <tye4>indentation</tye4> consistent
> and it get's phenomenally ugly if the queries get at all complex. I
> don't think there is a solution to this problem, and I would hazzard a
> guess that no matter what language is in use it would be equally ugly.

You're right. Actually, in most languages, it's significantly 
uglier.

> Aahz made the very valid point that specifying columns in your query
> *guarantees* tuple return order. I had not fully appreciated that, and I
> have put that into use in a number of places. However, I have to agree
> with Lance when he says that "SELECT * FROM table" is not as often a bad
> thing as Aahz claimed. 

That was me, not Aahz. I don't know exactly what Lance was 
dealing with, but I'd guess that I wouldn't have done it that 
way. I will also grant you that the rules are quite different for a 
5 user app and a thousand user app. But "SELECT *" is 
inefficient on both ends, and it doesn't tell the DB or the DBA 
what you are interested in. It doesn't give the drivers any 
chance to prebind for efficient data transfer. And a getting the 
descriptor of the result set is another query.

If you want the descriptor for a table (and can't use 
DESCRIBE), get the descriptor for this :
"SELECT * from <table> where 1 = 0". Now store that 
somewhere.

> ... Using "SELECT field1, field2, field3 FROM table"
> simply becomes *far* to cumbersome and difficult to read/debug if you
> get beyond five or so fields. 

I should show you some raw ODBC code. Queries have up to 
50 columns, they are parameterized and the output columns 
are array bound. It takes 100s of lines of (very boring and very 
repetitive) C for each query. But it's about 5x faster than the 
(very popular) DB library it replaced.

> If you are dealing with a large number of
> columns (my current project has one table with 33). Debugging an error
> becomes an exercise in frustration (especially when using the """SQL"""
> method as the interpreter points the error message at the first line of
> the multi-line execute statement.

If you're using ODBC, turn ODBC tracing on. The log file will 
be huge, but it will tell you exactly what went wrong. If it's 
direct connect, the DB should have an equivalent facility.

If it's the Python you're worried about, define the SQL string 
above the execute, and execute the named string.

custbyzip = """\
 SELECT c_name, c_addr1, c_addr2, city, state, zip 
   FROM customer
   WHERE state = ? """

rslt = execute(custbyzip, ("MA",))
 
> Lance posted some very nice code to build dictionaries out of the
> results from a query... 

I've used the trick myself - but it was for building HTML tables 
to display the results from an ad-hoc query built from an HTML 
form. I personally would never use "*" if I new what columns I 
needed.


- Gordon



From backov at nospam.csolve.net  Thu Jan 13 18:02:12 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Thu, 13 Jan 2000 23:02:12 GMT
Subject: PyErr_SetString() = Access violation in python15.dll!
Message-ID: <UPsf4.57298$Ke.53289@news21.bellglobal.com>

I'm not entirely sure what's going wrong here, as I think they were working
before, but now any time a wrapper routine calls PyErr_SetString(), I get an
access violation in python15.dll.. Typical uses are:

        PyErr_SetString(PyExc_TypeError,"incorrect number or type of args");
        PyErr_SetString(PyExc_IOError,"unable to load form file");

etc. Am I missing something? This is supposed to work, is it not?

Also, a python function does
.loadform(("blah","blahaswell",{'test':12345})) - shouldn't the third arg be
considered a dictionary?

   if (PyArg_ParseTuple(args,"(ssO)",&formname,&elename,&members))
    {
        element=inter->GetElement(elename,formname);
        if (element!=NULL)
        {
            // Make sure the members list is a dict
            if (PyDict_Check(members))

members is not a dictionary, according to this little bit of code. Other
than these strange errors, I'm really liking Python. I'm hoping someone can
help me out here.

Cheers,
Jason





From ScrewMe at mypad.com  Tue Jan 18 19:42:35 2000
From: ScrewMe at mypad.com (TechSpood)
Date: Tue, 18 Jan 2000 18:42:35 -0600
Subject: More problems....
Message-ID: <2E7h4.230$Pe6.2431@newsfeed.slurp.net>

I have been reading tutorials for about 2 weeks now, and just now decided to
put that knowledge to work, but I am having a huge problem...I will write a
simple program (in the IDLE enviroment) then when I run it, it will run it
til it gets to the first input statement, at which time it will stop reading
the program.  I don't get a syntax error  or any other kind of error.  After
I went back and read and re-read the tutorials i got very frusterated after
not being able to find absolutly any errors in my code... I even went so far
as to copy the code excactly as it was in a tutorial and put it in the
compiler.... same damn problem.  Is there some option I need to turn on or
some option that I might of accidently turned on that I wasn't supposed to?

Thanks




From quinn at mono.ugcs.caltech.edu  Fri Jan 28 17:26:53 2000
From: quinn at mono.ugcs.caltech.edu (Quinn Dunkan)
Date: 28 Jan 2000 22:26:53 GMT
Subject: tabs do WHAT?
References: <XFMail.000128190301.mikael@isy.liu.se>
Message-ID: <slrn894606.7f4.quinn@mono.ugcs.caltech.edu>

>On 28-Jan-00 Dirk-Ulrich Heise wrote:
> >  What bothers me even more, they don't mind writing functions without the
> >  function comment headers that shall be used in the project.
> >  You end up with functions containing not a single char of comment.
> >  It would be another improvement to Python if it would force you to
> >  write a doc string for each function of method.

But doc strings are not appropriate for all functions.  Most functions I find
myself writing are 5 - 25 lines long, and have a descriptive name and
one to three descriptive args.  Python is so readable I find documentation
would be clutter for these.

This is different for functions someone might use interactively, I try to have
a signature to refresh the memory.

In theory  :)

My rule is that if I can imagine myself looking at the code one year later,
and not being sure how exactly the function is used after examining it, it
either needs to be documented or fixed.

so-i-find-myself-documenting-a-lot-more-c-functions-than-python-ones'ly y'rs


From jblazi at netsurf.de  Fri Jan 21 10:24:41 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Fri, 21 Jan 2000 16:24:41 +0100
Subject: function "overloading"
Message-ID: <3888787d_2@goliath.newsfeeds.com>

I have 3 classes called Tline, Tsegment and Tpoint, repectively. I'd like to
write a function "subset" that cheks if one of the instances of the above
classes  is a subset of another instance (as long as this is sensible).

I am afraid I cannot overload the name "subset" and additionally when subset
has control it cannot decide the type of its arguments: Does this mean I
have to write several *different* subset functions having different names?

Janos Blazi




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From gerrit.holl at pobox.com  Sat Jan 29 14:24:38 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 29 Jan 2000 20:24:38 +0100
Subject: Some questions about history...
Message-ID: <20000129202438.A5089@stopcontact.palga.uucp>

Hello,

I'm creating a Dutch Python page, and want to tell something about
the history. But I can't find out the following things:

	- When did the PSA begin?
	- When was the domain www.python.org registered?
	- When was c.l.py founded? When was c.l.py.announce founded?

Even better than answers would be a link to a story telling the
answers to those questions.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From thamelry at vub.ac.be  Thu Jan 27 07:09:54 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 27 Jan 2000 12:09:54 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com>
Message-ID: <86pcii$600$1@mach.vub.ac.be>

[Thomas Hamelryck  <thamelry at vub.ac.be>]
: But many languages have evolved exactly this way: users criticize,
: complain, praize and whine, thereby influencing the shape of the
: languge in question.

[Doug Landauer <landauer at apple.com>]
: Yes, this works well in many cases, including Python.  Notice that
: it is, for the most part, *non*-users of python who complain about
: the indentation.  _Users_, for the most part, like it.  Languages
: that change by catering to the tastes of non-users tend not to do
: so well.

I agree that most users of python seem to like the strict indentation rule.
I seem to be one of the few actual python users who do not like it. The
complaints about the indentation do not really come from *non*-users of python.
Rather, they come from people who have gotten interested in python for some
reason or another and are completely turned down by the idiosyncratic use of
syntactically significant whitespace and/or tabs. So python is missing a lot
of potential momentum beacuse of the indentation stuff. I find this a pity.

Still a great language though. 

Cheers,

---
Thomas Hamelryck    Institute of Molecular and Structural Biology
Aarhus University   Gustav Wieds Vej 10C
DK-8000 Aarhus C    Denmark                                                                        


From neelk at brick.cswv.com  Wed Jan 12 02:17:13 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: Wed, 12 Jan 00 07:17:13 GMT
Subject: Multimethod.py -- multimethods in Python
Message-ID: <mt2.0-27553-947661433@news.informatik.uni-bonn.de>

I'd like to announce the existence of Multimethod.py, a package that
implements real (eg, CLOS/Cecil/Dylan) multimethods for Python. These
are nifty for oodles of reasons, but some of the more significant are:

  o They make the Visitor and Strategy patterns trivial to implement.
  
  o The whole "binary method/double dispatch" problem goes away, thus 
    promoting loose coupling and other software-engineering buzzwords.
  
  o They resolve the whole contravariance/covariance debate so totally
    it's hard to remember what the problem was.

You can find it on my webpage at:

  <URI:http://www.sff.net/people/neelk/open-source>

It should work out of the box on CPython 1.5.2, and Jpython if Jpython
has lists that support the insert() method. I don't know about Viper,
since Viper apparently has a rather different type system from vanilla
Python, but it should be easy to port, though.

The current version is quite rough, with lame algorithms used in a
couple of places and very substandard documentation. But it all works,
and is quite usable for experimentation and learning.

(Incidentally, is this the sort of thing that should go on the
types-sig?)


Neel

<P><A HREF="http://www.sff.net/people/neelk/open-source">Multimethod.py</A> - 
support for real (i.e. CLOS/Cecil/Dylan-style) multimethods for
Python.  (07-Jan-2000)

-- 
----------- 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 bsb at winnegan.de  Mon Jan 17 15:16:20 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 17 Jan 2000 21:16:20 +0100
Subject: putenv() and os.environ
In-Reply-To: Gerrit Holl's message of "Mon, 17 Jan 2000 20:35:59 +0100"
References: <20000117203559.A1246@stopcontact.palga.uucp>
Message-ID: <87puv0a1yj.fsf@baal.winnegan.de>

Gerrit Holl <gerrit.holl at pobox.com> writes:

> Hello,
> 
> >>> os.putenv('aaa', 'bbb')
> >>> os.environ.get('aaa')
> 
> I can't get any environment variables not set with the os.environ
> dict. Shouldn't os.environ be rewritten in C then, so it can get
> the REAL environment variable?

That's documented behavior, cf. Lib Reference lib/os_procinfo.html:

putenv (varname, value) 
  Set the environment variable named varname to the string value. Such
  changes to the environment affect subprocesses started with
  os.system(), popen() or fork() and execv(). Availability: most
  flavors of Unix, Windows.

  When putenv() is supported, assignments to items in os.environ are
  automatically translated into corresponding calls to putenv();
  however, calls to putenv() don't update os.environ, so it is
  actually preferable to assign to items of os.environ.

HIH
  Siggy




From duncan at rcp.co.uk  Wed Jan 12 10:37:14 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 12 Jan 2000 15:37:14 GMT
Subject: class methods?
References: <20000112145448.29699.qmail@web804.mail.yahoo.com>
Message-ID: <8EB99DF4Aduncanrcpcouk@194.238.50.13>

sjz18 at yahoo.com (Stuart Zakon) wrote in
<20000112145448.29699.qmail at web804.mail.yahoo.com>: 

>However, Python seems to want a class instance to call a method. Does
>every method on a class need self as a parameter?
Yes. Any unbound method in a class needs a class instance as its first 
parameter; any function which you store in a class magically becomes an 
unbound method. You can however store a bound method in a class and call it 
as though it were a class method.
 
>
>Certainly I could make a separate statistics class and instantiate it
>once; however, the nature of the problem is such that it is more natural
>for the statistics collection to be private to the original class. 
>
You could make a separate statistics class which is itself private to the 
original class. For example:

class C:
    class __Private:
        ncalls = 0
        def method(self):
            self.ncalls = self.ncalls + 1
            print "Called %d times" % self.ncalls
    __classmethods = __Private()
    classmethod = __classmethods.method

C.classmethod()
C.classmethod()

'method' isn't restricted to variables in __Private, it can also refer to 
class variables within C as 'C.whatever' so long as C is not changed at the 
module level.

The only other option I can think of is really horrible:
class C:
   def f(): print "This is f"
C.f.im_func()
will call f as a function rather than as a method.


From pbible at home.com  Sun Jan 16 19:23:32 2000
From: pbible at home.com (Paul E. Bible)
Date: Mon, 17 Jan 2000 00:23:32 GMT
Subject: problem with makepy output
References: <4vg48skfrigntmuslpfm40ngea5pdl4r2m@4ax.com>
Message-ID: <8itg4.308$iL4.4699@news.rdc1.tx.home.com>

It looks like makepy is grabbing an older version of Excel for some reason.
Try using makepy from the Excel9.olb (this is Excel's Type Library) from the
..\Microsoft\Office\ subdirectory.  Then copy the generated Python file to
your own  import file (eg. Excel9.py) somewhere on your PYTHONPATH.  Once
you've done that, just import Excel9 and that's it.  I'm running Windows/NT
(SP4) with Office 2000 and this works fine.

Good Luck,
Paul



"Calishar" <calishar@*NOSPAM*home.com> wrote in message
news:4vg48skfrigntmuslpfm40ngea5pdl4r2m at 4ax.com...
> Hello again folks,
>
>   Just tried using makepy (from win32all-127) to create a python
> module wrapping MS-Excel 2000, it worked fine, I moved the module to a
> directory in the python path, and imported it. This was the result.
>
> ------------
> PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
> win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
> >>> import excel5
> >>> app=excel5.Application()
> Traceback (innermost last):
>   File "<interactive input>", line 1, in ?
>   File "c:\Program Files\Python\kadabra\excel5.py", line 804, in
> __init__
>     oobj = pythoncom.new(self.CLSID)
> com_error: (-2147221164, 'Class not registered', None, None)
> ------------
>
> However, the same process worked on MS-Word 2000. Just wondering how I
> would start debugging this.
>
> Calishar




From blessing at fc.hp.com  Thu Jan  6 13:27:03 2000
From: blessing at fc.hp.com (Michael Blessing)
Date: Thu, 6 Jan 2000 11:27:03 -0700
Subject: Python Article on HP Developer's Resource
Message-ID: <852mi8$95b$1@fcnews.fc.hp.com>

Hi,

I am part of the team that publishes the HP Developer's Resource, a web site
for software developers using HP systems (primarily HP-UX). I recently
wrote a column about Python.  I invite you to give us a visit.

The DevResource home page is at http://devresource.hp.com/ ,
or you can go directly to the Python column at
http://devresource.hp.com/devresource/Features/Columns/MixedBlessings/mb003.html .

Michael Blessing
HP Developer's Resource





From grant at nowhere.  Wed Jan 19 17:20:38 2000
From: grant at nowhere. (Grant Edwards)
Date: Wed, 19 Jan 2000 22:20:38 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com> <slrn88c42t.53m.grant@grante.comtrol.com> <388624F4.8B2EB257@netscape.com>
Message-ID: <slrn88ce52.5ga.grant@grante.comtrol.com>

>Well, here is an extract of the file produced by makepy (under
>gen_py): - BTW, I selected the "OLE/Messaging 1.0 Object
>Library" not sure what is the msoutlk85.olb.-

It's got someting to do with MS Outlook, and it's not the right
file (as you've pointed out).  A few of the attribute names
matched and did what I expected Subject and Size, for example and
that led me down the wrong path for a while.

Once I finally realized that I had the wrong object
definitions, I then found that the definitions for the correct
objects are the ones in the cdo.dll library.

Things are working much better now.  :)

Thanks for the pointers.

-- 
Grant Edwards                   grante             Yow!  I am having FUN... I
                                  at               wonder if it's NET FUN or
                               visi.com            GROSS FUN?


From Christian.Schaller at mchp.siemens.de  Thu Jan 13 12:25:02 2000
From: Christian.Schaller at mchp.siemens.de (Christian Schaller)
Date: Thu, 13 Jan 2000 18:25:02 +0100
Subject: Win32: using PySECURITY_DESCRIPTOR Object
Message-ID: <387E0A6E.850D0A2A@mchp.siemens.de>

Hello...

  Does anyone know how to use a PySECURITY_DESCRIPTOR object in Python? 
I was reading file security settings via:
>>> import win32security
>>> res = win32security.GetFileSecurity('c:\', \
    win32security.DACL_SECURITY_INFORMATION)

  How can I print all values contained in res?  Please help me!

  Thanks a lot in advance!

bye
  Chris...


From neelk at brick.cswv.com  Tue Jan  4 21:49:14 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 5 Jan 2000 02:49:14 GMT
Subject: how to copy a dictionary
References: <roy-5C6597.21015404012000@netnews.nyu.edu>
Message-ID: <slrn875c9g.gvl.neelk@brick.cswv.com>

Roy Smith <roy at popmail.med.nyu.edu> wrote:
> If I do:
> 
> d = {'hedgehog': 'spiney norman'}
> temp = d
> temp['scotsman'] = 'earnest potgorney'
> 
> I end up changing the original dictionary, d.  It's obvious that what's 
> going on is when I do temp = d I get a pointer to the same object 
> instead of a new object.  My question is, how do I force a new object to 
> be created, so when  modify it, I don't also modify the original?

Yup. Variable assignment in Python is always a matter of rebinding. 
Thinking of it as pasting a label on an object, and then peeling it
off and pasting it on a new object is the mental model to use. :)

Your specific problem is solved with the copy module, eg:

  >>> import copy
  >>> x = {'foo': 'bar', 'baz': 'quux'}
  >>> x
  {'foo': 'bar', 'baz': 'quux'}
  >>> y = copy.copy(x)
  >>> y
  {'foo': 'bar', 'baz': 'quux'}
  >>> y['thud'] = 'garp'
  >>> y
  {'foo': 'bar', 'thud': 'garp', 'baz': 'quux'}
  >>> x
  {'foo': 'bar', 'baz': 'quux'}

This creates a shallow copy of the object. If you want a deep 
copy, use copy.deepcopy(). 


Neel


From mwh21 at cam.ac.uk  Thu Jan 13 12:29:29 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jan 2000 17:29:29 +0000
Subject: Doing things in one line?
References: <78D9D2737D484D028525685B005D29CA.005D2ACF8525685B@ttm.com> <3870DB47.51B4C5AA@ttm.com>
Message-ID: <m3wvpdj2x2.fsf@atrus.jesus.cam.ac.uk>

Scott Anderson <sanderson at ttm.com> writes:

> Actually, just search the mailing list archives back just before the
> holidays, because I asked that very question. There were about 30
> responses IIRC so I'm not going to summarize them here. :-)
> 
> Here's a one-liner I did the other day for the heck of it:
> 
> reduce(lambda x, y: x + y, range(1, int(raw_input('Enter a number: ')) +
> 1))
> 
> It accepts a number from the user and adds 1 to it. :-)
> 

No it doesn't. It computes the nth "triangular number", where n is the
number the user enters.

Or it dies if the user enters a number less than 1.

Cheers,
Michael


From david.maslen at iname.com  Fri Jan 28 10:00:03 2000
From: david.maslen at iname.com (David Maslen)
Date: Fri, 28 Jan 2000 15:00:03 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com>
Message-ID: <874sby5m6l.fsf@iname.com>

I didn't really want to get involved in this thread, but I have one
comment to make.

Using the python-mode in emacs, getting the code to line up with tabs
is easy. I press tab, and it puts it in the right place. If it doesn't
it's because I've forgotten the : after an if statement, or perhaps
it's because of where a #comment is directly above it.

It is possible to get it wrong, but agian when running python from
inside emacs, emacs jumps to the line with the exception and it's
quickly fixed.

However, not everyone uses emacs. As an emacs user I wouldn't wish it
on you for all it's virtues. I think the python form of indenting
might be rather annoying if it weren't for the programing tool I
use. I suspect were I inclined to code with vim, I might get the
identation wrong, more often than I do now, and that might make me
complain that {} are a good thing, like quite a few people here have
been doing :)


From cjc26 at nospam.cornell.edu  Thu Jan 27 18:37:56 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: 27 Jan 2000 23:37:56 GMT
Subject: lib reference in Unix man format
References: <200001271548.KAA04215@trm.bmo.com>
Message-ID: <slrn891ljb.5tk.cjc26@bridget.mindriot.net>

Pada Thu, 27 Jan 2000 10:48:53 -0500 (EST), Tao Lu bilang:
| Where can I find the lib reference in Unix man format? Forgive me if this has 
| been asked before.
| 
| Have been programming in Perl for years, I found those perlfunc/perlop/perlfaq 
| are very convenient. Any similar work done for Python developers/newbies?

There isn't any documentation in man format.  I know this because Tom
Christiansen came over here once and VERY LOUDLY complained about the
lack of this. :)  Actually I find this annoying too, so maybe I'll
work on converting the on-line docs to man pages (when I have some
free time...:).  I'd prefer to have each built-in function have its
own page though, rather than glomming them all together into one
large "pythonfunc" page.


-- 
cliff crawford  -><-  http://www.people.cornell.edu/pages/cjc26/
                     "All I need is a small town
                      with good Internet connectivity" - Nadine


From thantos at chancel.org  Fri Jan 28 17:25:07 2000
From: thantos at chancel.org (Alexander Williams)
Date: Fri, 28 Jan 2000 22:25:07 GMT
Subject: lib reference in Unix man format
References: <200001281339.IAA12329@trm.bmo.com> <20000128090950.A2767@better.net>  <oqsnzirygy.fsf@titan.progiciels-bpi.ca>
Message-ID: <slrn8945q3.fh3.thantos@brimstone.mecha>

On 28 Jan 2000 10:46:05 -0500, Fran?ois Pinard
<pinard at iro.umontreal.ca> wrote:

>I have a big deal of Texinfo documentation in both Info and HTML format,
>here, and Info is much more useful that HTML browsers.  In my opinion, Info

And, of course, if you have written your docs /in/ Texinfo, then you
can from that generate the output in both Info /and/ HTML; I do it all
the time.  Choice of hypertexts, the choice of a Gnu generation.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From mgushee at havenrock.com  Wed Jan 12 08:50:25 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 12 Jan 2000 08:50:25 -0500
Subject: Threading & garbage collection
Message-ID: <m3embnpffi.fsf@havenrock.com>

Hi, folks--

Another dumb threading question here:

I'm trying to write a subclass of threading.Thread where each
thread:

  1) is invoked by a GUI event handler
  2) collects some data (to be exact, it downloads a web page
	and extracts some information from it), and finally
  3) invokes a GUI method to display the data

The run() method of my subclass would look something like this
(unless someone has a better idea):

    def run(self):
	## various thready actions
	.....
        self.win.thread_done(data)

... where self.win is, of course, a GUI object. Do the standard
rules about circular references apply to threads? I.e., should I
add something like:

	self.win = None

to the above?

Thanks in advance for your tips!

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From nielsenjf at my-deja.com  Wed Jan  5 23:23:37 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Thu, 06 Jan 2000 04:23:37 GMT
Subject: how to include a file -- not import??
Message-ID: <8515bs$2vn$1@nnrp1.deja.com>

Hopefully, there's a simple answer to a problem I am having.

I am using import to include a file, however, it looks like the module
then has another namespace, which means it can't access some variables.

Is there a way to simply include the raw text and insert it into your
own code, as if it was not in a another file?

Thanks for any help,

john



--
nielsenjf at my-Deja.com


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


From paul at prescod.net  Thu Jan 20 16:43:36 2000
From: paul at prescod.net (Paul Prescod)
Date: Thu, 20 Jan 2000 13:43:36 -0800
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com> <388774C7.1784B662@callware.com>
Message-ID: <38878188.B6ECB0B7@prescod.net>

Ivan Van Laningham wrote:
> 
> ...
> If you had bothered to use Python, you would already know that it is not
> a compiler but an interpreter.

Well, the thing that makes .pyc's can be called a compiler.

> However, I should remind you again.  Shit is not Python.

I think we'll find his experiment quite harmless so there is no need to
get excited. Might as well be amused rather than hostile.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Earth will soon support only survivor species -- dandelions, roaches, 
lizards, thistles, crows, rats. Not to mention 10 billion humans.
	- Planet of the Weeds, Harper's Magazine, October 1998



From paul at prescod.net  Mon Jan 10 15:40:42 2000
From: paul at prescod.net (Paul Prescod)
Date: Mon, 10 Jan 2000 12:40:42 -0800
Subject: Python is much faster than Java?
References: <387A492A.CBB58B86@iqsoft.hu>
Message-ID: <387A43CA.250B52F4@prescod.net>

Java and Python implementations vary widely so it is probably not useful
to say that one is always faster than the other. Presuming you used
CPython, the big variable is probably the JVM you used for Java. I would
expect some to be much faster than CPython and some to be much slower.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
When I'm gone, boxing will be nothing again.  The fans with the cigars
and the hats turned down'll be there, but no more housewives and
little men in the street and foreign presidents.  .... I was the
onliest boxer in history people asked questions like a senator.  
            -- Muhammad Ali



From gerrit.holl at pobox.com  Tue Jan 18 01:41:15 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 18 Jan 2000 07:41:15 +0100
Subject: Any Newbies?
In-Reply-To: <s2Wf4.974$sv3.3886@newsfeed.slurp.net>; from ScrewMe@mypad.com on Sat, Jan 15, 2000 at 02:24:38AM -0600
References: <s2Wf4.974$sv3.3886@newsfeed.slurp.net>
Message-ID: <20000118074115.A664@stopcontact.palga.uucp>

TechSpood wrote on 947899478:
> Anyone besides me out there almost a complete newbie to all programming and
> is learning python for a first language?

I did almost a year ago. I'm 14 years old.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From gerrit.holl at pobox.com  Sun Jan  2 17:41:41 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 2 Jan 2000 23:41:41 +0100
Subject: Need help with canvas widget in Tkinter please
In-Reply-To: <386EC3E6.FAC5BC0@telebot.com>; from flounder_pounder@telebot.com on Sun, Jan 02, 2000 at 03:20:07AM +0000
References: <386EC3E6.FAC5BC0@telebot.com>
Message-ID: <20000102234141.A6316@stopcontact.palga.uucp>

flounder wrote on 946779607:
> display = Canvas(mw, width=640, height=480).pack()
> 
> item = display.create_image(0, 0, anchor=NW, image=photo)

That's not a TKinter related problem. You're not creating a class instance,
as you might think. .pakc() doesn't return an instance. Try:
display = Canvas(mw, width=640, height=480)
item = display.create_image(0, 0, anchor=NW, image=photo)
...
display.pack()

regards,
Gerrit.

-- 
"The move was on to 'Free the Lizard'"

  -- Jim Hamerly and Tom Paquin (Open Sources, 1999 O'Reilly and Associates)
 11:27pm  up 12:26, 13 users,  load average: 0.01, 0.05, 0.00



From aahz at netcom.com  Tue Jan 18 16:12:27 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jan 2000 21:12:27 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com>
Message-ID: <862kvr$15i$1@nntp9.atl.mindspring.net>

In article <862jqd$l9s at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>
>Are you insinuating that this is not a problem since I'm not as experienced
>as Guido? I don't need to design a language to detect flaws in it.

So, what do you think would be better than indentation, keeping in mind
my earlier comment about the difficulty of reading multiple block coding
styles?
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From zippy at cs.brandeis.edu  Thu Jan 20 23:59:40 2000
From: zippy at cs.brandeis.edu (Patrick Tufts)
Date: Thu, 20 Jan 2000 20:59:40 -0800
Subject: regex question
References: <roy-835F11.21172620012000@netnews.nyu.edu>
Message-ID: <200120002059400984%zippy@cs.brandeis.edu>

[[ This message was both posted and mailed: see
   the "To," "Cc," and "Newsgroups" headers for details. ]]

In article <roy-835F11.21172620012000 at netnews.nyu.edu>, Roy Smith
<roy at popmail.med.nyu.edu> wrote:

You can simplify the 12-digit hex pattern to

[0-9A-F]\{12\}  

use case-insensitive matching when compiling the regexp to take care of
the "a-f" case.

--Pat


From sjz18 at yahoo.com  Wed Jan 12 11:44:17 2000
From: sjz18 at yahoo.com (Stuart Zakon)
Date: Wed, 12 Jan 2000 08:44:17 -0800 (PST)
Subject: bdb: which class?
Message-ID: <20000112164417.4994.qmail@web805.mail.yahoo.com>

"stephen j. turner" <sjturne- at ix.netcom.com> wrote: 

> How about the following:
> 
> def where_am_i(self, frame):
>     from types import InstanceType, MethodType
>     code = frame.f_code
>     name = code.co_name
>     if code.co_argcount:
>         arg0 = frame.f_locals[code.co_varnames[0]]
>         if type(arg0) is InstanceType:
>             attr = getattr(arg0, name, None)
>             if attr and \
>                type(attr) is MethodType and \
>                attr.im_func.func_code is code:
>                 classname = arg0.__class__.__name__
>     ....
> 
> If the current frame is for a method invocation, then its corresponding
> code object must have at least one argument, and the first named
> argument (probably `self', but we can't assume) identifies the class
> instance.  This name is then used as an index into the frame's `locals'
> dictionary to get the first argument.
> 
> Since, however, the frame could also be for a function invocation, we
> make sure that (1) the type of the first argument is a class instance,
> (2) the instance has an attribute whose name matches the code object's,
> (3) the instance attribute is in fact a method, and (4) the method's
> underlying code object is the same as the frame's.  If all these are
> true, then we get the class name from the instance.

My gosh! I was afraid it was going to be that complicated!
If I had any spare time I would look into why there couldn't just be the
following assignment to get the class:

class_name = frame.f_code.co_class_name

Then you could check to see if it was 'None' to indicate that the method is not
on a class but on a module.

Regards,
Stuart
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com



From mskott at image.dk  Sat Jan 29 15:34:15 2000
From: mskott at image.dk (Martin Skøtt)
Date: 29 Jan 2000 21:34:15 +0100
Subject: Writing multiple lines to a file
Message-ID: <emb0obw8.fsf@jumbo.skott.dk>

Hi
I am trying to create a program that writes several lines to a file
but I have encountered some minor problems.
I write to the file using the following code:
f.write(S)  # S contains a string I wish to write to the file
f.close()

The line gets written to the file but if I run the program again the
line generated in the previous run gets overwritten? I dont want that,
I want the program to append my string to the file without touching
other lines. Can you help me.
I am sorry if this is a FAQ but i'm all new to Python.

-- 
Martin Sk?tt
mskott at image.dk

"Every time I learn something new it pushes some old suff out of my brain"
Homer Simpson


From skip at mojam.com  Mon Jan 10 09:46:07 2000
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 10 Jan 2000 08:46:07 -0600 (CST)
Subject: fetch a html page via proxy?
In-Reply-To: <387978ff.3083491@news.jps.net>
References: <387978ff.3083491@news.jps.net>
Message-ID: <14457.61615.944464.4935@beluga.mojam.com>

    Ricardo> I need to fetch a page from the web but the linux machine
    Ricardo> doesn't have direct connection to the net.  It needs to go thru
    Ricardo> a proxy to connect to the web.  How can I do this with Python?

The urllib module handles proxies.  If you set the http_proxy environment
variable appropriately on Unix or Windows or do the Internet Config thing on 
Mac, it will use the proxy instead of attempting to make a direct connection 
to the server.

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 tarude at whitestar.localdomain  Sat Jan 15 02:43:41 2000
From: tarude at whitestar.localdomain (Trebor A. Rude)
Date: 15 Jan 2000 00:43:41 -0700
Subject: Help me improve my Python Skills, please
Message-ID: <m2g0vzsrte.fsf@excalibur.peernet>

I just started picking up Python a few days ago, and am rather
enjoying it.  To help me learn it, I converted one of my old TCL
scripts into Python, which I've attached below. However, this being my 
first serious python script, I'm pretty sure I'm doing some things
less efficiently than necessary, through lack of familiarity with the
library and language in general. So I'd like to enlist the group's
help in improving the code. Logic, style, better ways of doing things, 
whatever.  Thanks in advance.

A little explanation before the code.  The script recursively descends 
a directory structure, building a .jdb (Jukebox database) file for
each directory based on the mp3 tag info of the mp3s in that
directory and all its subdirectories. It also repairs damaged comment
fields caused by a slightly buggy encoder program.

#!/usr/bin/env python

# Does this look strange? My basic philosophy on
# using "import" vs. "from <foo> import <bar>" is
# based on my conception of how easy it might be
# to misinterpret the meaning of the function. Also,
# if I need a bunch of names from a package, I'll
# tend toward "import".

import os
import re
import select

from os.path import join, basename, isdir, splitext
from glob import glob
from fnmatch import fnmatch
from operator import concat
from commands import getoutput

# Only reason this is a function is because it gets called recursively.

def Make_Playlist ():
    pwd = os.getcwd()
    filename = join(pwd, basename(pwd) + '.jbd')

    try:
        listfile = open(filename, 'w')
    except IOError:
        sys.exit("Cannot open " + filename + " for write: " + IOError)

    # Since sort() returns None, need to go through these extra
    # steps instead of just putting all on the for line.
    
    files = glob(join(pwd, '*'))
    files.sort()
    
    for file in files:
        if isdir(file):   # Recursively descend directory.
            os.chdir(file)
            Make_Playlist()
            os.chdir(pwd)

            # Concatente the just-created .jbd onto the current one.
            
            infile = open(join(file, basename(file) + '.jbd'))
            listfile.writelines(infile.readlines())
            infile.close()
        elif fnmatch(file, '*.mp3'):
            newfilename = re.sub(' ', '_', basename(file))
            if (newfilename != basename(file)):
                os.rename(file, newfile)
                file = join(pwd, newfile)

            # Don't really need this since getting a better encoder,
            # but still might want to make it optional.
            
            #artist = re.sub('_', ' ', basename(pwd))
            #title = re.sub('_', ' ', splitext(basename(file))[0])
            #os.system("mp3info -t " + title + " -a " + artist \
            #+ " -g 1 " + file)

            # Yes, I really do have filenames which use all the
            # "bad for the shell" characters below.
            
            escaped_filename = re.sub('(["\'()& ])', r'\\\1', file)
            
            # Deleted the actual format because it was too long.
            # If anyone's interested, just ask.

            mp3info = getoutput('mp3info -f "..." + escaped_filename)

            # The encoder program I use is slightly buggy and tends to 
            # leave spaces followed by one random control character at 
            # the end of the comment field of the mp3 tag. Generally,
            # this isn't a problem, but it IS if that control
            # character is ^J, so I fix it here.

            # Notice the (\B|\b) construct.  What I really wanted to
            # do was make the whole of \1 optional, but then it causes 
            # an error if \1 doesn't match, which is bad because the
            # point is to delete the stuff after it.  Is there a
            # better way to do this?

            fixedinfo = re.sub(\
            r'\|((\B|\b)([A-Za-z \'/.]*[A-Za-z])?) +[\1-\26]\|', \
            r'|\1|', \
            reduce(concat, mp3info))

            # Hmm, come to think of it, I might want to repair the
            # actual mp3 file here.  I'll have to look into the
            # "Matcher" objects. (Would that be the right way to save
            # \1 from above?)

            listfile.write(fixedinfo)
            listfile.write('\n')
            
    listfile.close()

# os.path.walk() might simplfy things mightily, but would require some
# information to be saved outside the function to keep track of
# just-constructed .jbd files for inclusion in the current one.

Make_Playlist()



-- 
Trebor A. Rude
trebor at bwn.net
Registered Linux User #89308
http://counter.li.org/


From tiddlerdeja at my-deja.com  Sun Jan 30 14:33:08 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Sun, 30 Jan 2000 19:33:08 GMT
Subject: COM register ok. Unregister??????
References: <871ohi$tfl$1@nnrp1.deja.com>
Message-ID: <8723lm$59r$1@nnrp1.deja.com>

Ok, I have the answer to this one:

python com.py --unregister

In article <871ohi$tfl$1 at nnrp1.deja.com>,
  tiddlerdeja at my-deja.com wrote:
> Could someone tell me how I unregister my COM stuff in Python?
>
> I used:
> import win32com.server.register
> win32com.server.register.UseCommandLine(COMBid)
>
> To register it.
>
> Any help appreciated.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From smb at bby.com.au  Sun Jan  9 22:49:18 2000
From: smb at bby.com.au (Sarah Burke)
Date: Mon, 10 Jan 2000 13:49:18 +1000
Subject: Zope and NT databases
Message-ID: <387956BD.CFDF17DD@bby.com.au>

Hello...
I want to connnect up to a NT Access 97 Database through Zope running on
Unix. Is the mxODBC module in an External Method the way to go?
Thanks in advance



From matthewm at zebrasoft.co.nz  Tue Jan 25 02:49:02 2000
From: matthewm at zebrasoft.co.nz (Matthew Miller)
Date: Tue, 25 Jan 2000 20:49:02 +1300
Subject: using threads with python and winnt
References: <1264075593-4560755@hypernet.com> <38835f4c.1380925@news.iol.ie> <388C1CE3.3391A5CF@zebrasoft.co.nz> <86ismn$50a$1@athena.netset.com>
Message-ID: <388D556E.1DC8C084@zebrasoft.co.nz>

Hi Guy,

Maybe in the heat of the moment I was overly glib about Win9x and python
threading. To start with, I almost never use Win9x which is certainly not the
ideal starting point for assessing its threading capability/coupling with
python. The project was helping our local high school implement a teaching aid
to help students understand the inner workings of tcp/ip. The idea was that
students create a text-file like:-
# Connect to an SMTP server and send a basic message
/r smtp.foobar.com:25
/r helo matthew
/r mail from: matthew at matthew.com
/r rcpt to: matthew at matthew.com
/r data
.
/r quit

and submit it to a 'Protocol SERVER' (listeneing on a tcp port - eg: 12345)
which would then dispatch the contents to either a real or simulated tcp service
(ftp:21, pop:110, etc) for processing. the /r tag meant to wait for a server
response. The Protocol Server was fully threaded, and the client apps started a
new thread for each file (or stdin keystrokes) sent. Win95 was used on all
machines. Ultimately a response file was sent back to each students home
directory to see how they did, and sometimes (depending on protocol, etc) would
get a real world result like their actual e-mail. Anyway, to cut a long story
short, I gave up and implemented the project in Java (SORRY!). As I said, I'm no
expert in Win95 or Python, but I found the Win95 server 'behaviour' of
locking-up (requiring a hard reboot) just too hard on the nerves. I don't know
whether it's Win95's tcp/ip architecture (which I'm told is not so robust), ME!
or the threading/loading factor. A no-frills server and client was developed
under NT using Python and seemed to work fine, but when the same program was run
under Win95, the protocol server coped for an arbitrary amount of time and then
it didn't. Hopefully you get the gist of what the project was about. The only
comment I can make is that Win95, lots of threads and lots of socket activity
didn't work for me. Can't say that Java would have been different under Win95.
In the meantime, the PC running the Protocol Server was upped to NT (and I'm
hoping will finally make the final evolutionary step to Linux sometime soon),
and the Java stuff works great. Maybe I'll revisit this as a Python project in
the future. To conclude:- NT/Python gets the thumbs up, Python/Win95gets the
thumbs down.

HTH, Matthew.

"Guy L. Oliver" wrote:

> Any more details?  I've been using threading package under 98, 95, nt,
> linux, and solaris with no problems.  Also, the semaphores, conditions,
> queue, etc.  ive never had any problems (thread pool based web server
> experiment, with python generated dynamic content)
>
> Id be very interested to hear of any problems you had so I can watch out for
> them.
>
> glo
>
> "Matthew Miller" <matthewm at zebrasoft.co.nz> wrote in message
> news:388C1CE3.3391A5CF at zebrasoft.co.nz...
> > Be afraid, be very afraid! Seriously though, I head-banged with threading
> > under Win98 and eventually gave up. The app though works perfectly under
> > WinNT. Unless it is REALLY neccessary to support Win9x, spare yourself
> from
> > the pain. matthew.
> >
> > Sean Mc Grath wrote:
> >
> > > >Kreiner Karl or Charly or somebody wrote:
> > > >
> > > >>             does anyone have expiriences with using threads and winnt
> ?
> > > [...]
> > > <gmcm at hypernet.com> wrote:
> > > >Threads work dandy on NT
> > > [...]
> > >
> > > Does anybody know how good/bad/indifferent the threading stuff works
> > > on Windows 95/98? I am considering using threads for an app
> > > under development but I'm a bit worried (read "terrified") about
> > > the issues that might pop up in '95 and '98.
> > >
> > > regards,
> > > Sean McGrath
> > >
> >
> > --
> > Matthew Miller
> > Zebra Software Ltd.
> >
> > E-mail: matthewm at zebrasoft.co.nz
> > Ph: 025-2920625 / FAX: 03-3842397
> >
> >

--
Matthew Miller
Zebra Software Ltd.

E-mail: matthewm at zebrasoft.co.nz
Ph: 025-2920625 / FAX: 03-3842397




From bowman at montana.com  Sat Jan  1 18:05:43 2000
From: bowman at montana.com (bowman)
Date: Sat, 01 Jan 2000 23:05:43 GMT
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com>
Message-ID: <bLvb4.2732$S%2.10694@newsfeed.slurp.net>

randy_shaffer at my-deja.com <randy_shaffer at my-deja.com> wrote:

>I have experimented with tkinter and the PMW library.  Things seem to
>operate a bit slow.  The other options/considerations I?ve pondered are:

Is the speed a real showstopper? Assuming your NT clients already have 
Tcl/Tk and Tkinter running this may well still be the best solution.

Most other GUI packages may require additional libraries to be installed
on NT, and this may be a problem. Also, they may not show much of a speed
increase.

Java 1.x may work if you are familiar with Java already. If not, there is an
additional learning curve. I haven't tested the latest 2.0 implementations,
but when 2.0 first came out, it was noticeably slow on a 233 mhz system 
where Tkinter is perfectly acceptable. You pay dearly for every level of 
abstraction.




From helge.hess at mdlink.de  Fri Jan  7 23:05:51 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Sat, 08 Jan 2000 05:05:51 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <852jbk$9tt$1@nntp9.atl.mindspring.net> <3875099A.2CF9333C@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net>
Message-ID: <3876B79F.30439283@mdlink.de>

Aahz Maruch wrote:
> 
> Not directly, no.  I think it breaks compatibility with Python's
> extremely simple object model and I would be surprised if Guido
> disagrees with me.

Hm, I wonder why it is simpler to have two separate kinds of methods
instead of a consistent behaviour for all methods ?

Helge
-- 
SKYRIX-OS  Web Application Environment - http://www.skyrix.com
SKYRIX-IAS Groupware Application Suite - http://www.skyrix.com


From nolan_d at bigfoot.com  Mon Jan  3 12:46:45 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Mon, 03 Jan 2000 11:46:45 -0600
Subject: Considering using python for something, and need a bit of info :)
In-Reply-To: <0.3903b60a.25a2245f@aol.com> (PartyG2816@aol.com)
References: <0.3903b60a.25a2245f@aol.com>
Message-ID: <E125BZ7-0007Uv-00@ethereal>

>>>>> "PartyG2816" == PartyG2816  <PartyG2816 at aol.com> writes:

    PartyG2816> Allo list, I'm involved in writing a MUD server from
    PartyG2816> scratch with a small group of other people.  For you
    PartyG2816> who do no know, a MUD is a multi-user text game.  I've
. . . 
    PartyG2816> Could the Python interpreter be embedded and used in a
    PartyG2816> way that would allow me to use it like this within my
    PartyG2816> application?

I suppose it could be embedded, yes. A more practical approach would
  probably be to write your server in either Python or JPython
  completely. I was using such an approach until I discovered Muq
  (http://muq.org/~cynbe/muq/), and now I'm trying to switch to that
  as my platform, and write a Python-to-Muq assembly compiler. Granted,
  you don't have the freedom which a full Python-based product would
  allow, but if I ever finish this, the language will be extremely
  similar to Python, minus some of the built-in modules like sys which
  aren't useful in a MUD environment.

    PartyG2816> Another concern of mine is: how secure can it be? IE
    PartyG2816> can i restrict certain aspects of it based upon a
    PartyG2816> condition? Say I have a a normal builder, ie someone
    PartyG2816> who simply makes areas and may perform a few scripts
    PartyG2816> for them, I wouldn't want to let him do certain
    PartyG2816> things... Whereas me as head programmer, I would have
    PartyG2816> access to many more options within my scripts :) Or
    PartyG2816> take a normal "player".  I'm going to be allowing
    PartyG2816> small scripts and macros and such for my players to
    PartyG2816> automate a few 'boring' things, I would want to give
    PartyG2816> them access to a limited set of use, and not let them
    PartyG2816> say re-arrange everything in the room. *chuckles* Ah
    PartyG2816> well, maybe you peoples understand what I'm asking and
    PartyG2816> can help, I'd appreciate any response :) If you have

This seems more like an application issue rather than a Python one. If
you make the items property in a room hidden, you can probably detur
many casual crackers from touching it. However, it would still be
possible to modify the hidden/private variable, see section 9.6? in
the tutorial. It may be possible to code Property objects which store
MUD information, and have some sort of security mechanism on them to
determine who can modify them, and preventing folks from simply typing
commands directly into a Python prompt. Then there's the issue of
preventing people from doing:

import sys
sys.exit()

and bringing down your game. :)

I originally started by writing my own server in Python, but I then
realized that I didn't know enough about security to patch every
single (or even most) point of attack. I think that Muq will really
shine here. I'm not saying that successful servers can't be written;
see below for a list of examples.

    PartyG2816> p.s. - If it turns out that yall think Python could be
    PartyG2816> suitable for osmething like this, could you include a
    PartyG2816> few linkst o places that discuss and give some working
    PartyG2816> examples of embedding python in another application
    PartyG2816> and running scripts through it using whatever API is
    PartyG2816> inherent in Python for that use? :)

Sure. There's POO, a Python implementation of MOO at
http://www.strout.net/python/intro.html. POO is actually listed there,
but there are some other cool programs you may find useful.

Then there's Momoko (http://uts.cc.utexas.edu/~blanu/mud/) which uses
JPython instead of plain Python, thus allowing you to script in both
Java and Python.

There is a Python patch for TinyMUSH 3.0B5 which lets you write code
on MUSH objects in Python. See
http://www.helsinki.fi/~mstenber/tm30b5pv001.diff for this.

Moebius (http://www.bofh.net.au/mud/) is another project which I
stumbled on recently. It stores object code and such in an SQL
database, and seems interesting. It also supports server distribution,
allowing multiple servers to access the same MySQL database.

Varium (http://www.varium.com) is a project to develop a distributed
MUD server in Python. I haven't looked much at this one.

If you're more interested in embedding/extending, and not in
Python-based MUD servers, check out SWIG (http://www.swig.org). It
reads SWIG interface files or even C/C++ headers, and generates
wrappers in Perl, Python, TCL, Java, and a few other languages. I've
had lots of success using this with other projects.

And I should probably shamelessly plug my own project along with all
the others. :) Ethereal (http://etherealmud.sourceforge.net) is an
attempt to develop a MUD platform with a long list of what I feel are
fairly interesting ideas. The MUD concept of this project is fairly
solid and currently (albeit slowly :) being written down in a design
document, while the server-side aspects are still being
decided. Lately I've been considering using Muq because many of its
concepts seem to mesh with mine, but I've gone from writing my own
server to planning to use Momoko to wanting to use Muq, so who knows
what I'll be doing next month. :)



From garry at sage.att.com  Mon Jan 10 11:07:07 2000
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Mon, 10 Jan 2000 16:07:07 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net> <slrn87cvv9.gcu.kc5tja@garnet.armored.net>
Message-ID: <387A03AB.3373AF8E@sage.att.com>

"Samuel A. Falvo II" wrote:
> 
> >>(Personally, I think the COM solution is a superior one anyway, but that's
> >>just my opinion.)
> >
> >Sure it is -- but compatibility is even better.
> 
> Then why aren't we all using CP/M?

we are.  they call it "windows" now.

-- 
Garry Hodgson			Every night 
garry at sage.att.com		a child is born
Software Innovation Services	is a Holy Night.
AT&T Labs			 - Sophia Lyon Fahs


From emuller at painewebber.com  Mon Jan  3 09:43:43 2000
From: emuller at painewebber.com (emuller at painewebber.com)
Date: Mon, 03 Jan 2000 14:43:43 GMT
Subject: Help with Installer beta 3e
References: <3867a311@news.gvsu.edu>
Message-ID: <84qcir$8u9$1@nnrp1.deja.com>

run 'freeze.py <script name>.py' and it should work (it does for me).

In article <3867a311 at news.gvsu.edu>,
  "Andrew Sterian" <steriana at gvsu.edu> wrote:
> I'm trying to use Gordon McMillan's Installer beta 3e software to
distribute
> a set of scripts as a package. This is simple stuff, just a bunch of
scripts
> that need to be installed in a directory somewhere on the Python path
> (i.e., no DLL's, no Tkinter, etc.)   I need this distribution to be a
simple
> EXE that the user can download and run, assuming Python has already
> been installed.
>
> The problem is I can't figure out how to use the Installer package. I
> tried running the Simple script on my top-level script and the
generated
> EXE file fails to run (it says it can't find  .PY file that is part of
my
> package).
> I deleted all .PYC files as the installhelp file suggests, but nothing
helped.
>
> So...is there any documentation or tutorial that explains how to get
started
> with Installer to do basic stuff? I'm afraid Gordon's work is so
robust and
> flexible that the basics have been obscured.
>
> Many thanks,
>
> Andrew.
>
>


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


From aahz at netcom.com  Tue Jan 11 21:22:20 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 12 Jan 2000 02:22:20 GMT
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <1264508924-1480880@hypernet.com> <85g09q$o7e$1@nntp3.atl.mindspring.net> <slrn87nemb.qla.kc5tja@garnet.armored.net>
Message-ID: <85gogs$v35$1@nntp5.atl.mindspring.net>

In article <slrn87nemb.qla.kc5tja at garnet.armored.net>,
Samuel A. Falvo II <kc5tja at garnet.armored.net> wrote:
>In article <85g09q$o7e$1 at nntp3.atl.mindspring.net>, Aahz Maruch wrote:
>>
>>Hmmm...  So it's kind of like sparse overloading?  ;-)
>
>An excellent description!  Consider it stolen.  :)

My pleasure.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From paul at prescod.net  Fri Jan 21 12:04:50 2000
From: paul at prescod.net (Paul Prescod)
Date: Fri, 21 Jan 2000 09:04:50 -0800
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com>
Message-ID: <388891B2.4923A78D@prescod.net>

tye4 wrote:
> 
> I've used some stupid Unix shell languages which fail if I add whitespace:

Right. But that isn't Python. People come to Python's indentation with
an attitude like: "well I heard that a lot of Dutch people use drugs so
you must use drugs." "Make" uses whitespace in a manner that is "on
drugs". Python uses whitespace in a manner which is demonstrably,
testably and provably easy and scalable.

> Why is my question about whitespace raise so many eyebrows? This feature is
> commonplace.

Because the feature of whitespace insensitivity is so commonplace, we
all know about it so there is no need to describe it for us or act as if
we hadn't considered it. After all, Guido is a C programmer.

-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Earth will soon support only survivor species -- dandelions, roaches, 
lizards, thistles, crows, rats. Not to mention 10 billion humans.
	- Planet of the Weeds, Harper's Magazine, October 1998



From w6uv at hotmail.com  Mon Jan 17 15:41:47 2000
From: w6uv at hotmail.com (Jerry Gardner)
Date: Mon, 17 Jan 2000 20:41:47 GMT
Subject: Any Newbies?
References: <s2Wf4.974$sv3.3886@newsfeed.slurp.net>
Message-ID: <slrn886p0h.1ssj.w6uv@voyager.home.com>

On Sat, 15 Jan 2000 02:24:38 -0600, TechSpood wrote:
>Anyone besides me out there almost a complete newbie to all programming and
>is learning python for a first language?

Yes, sort of. I'm just starting to learn Python, but it's not my first
language.

It is a good language to start with--probably easier than C, C++,
Java, and even Perl--for most people.

Recently, I've been doing almost all of my programming in Perl, and
Python is like a breath of fresh air in comparison. I'm looking
forward to being able to read my Python programs 6 months after I
write them--something I find difficult to do with Perl code.

-- 
Jerry Gardner     | "Bill Clinton has all the steely resolve
w6uv at hotmail.com  | of a kamikaze pilot on his 37th mission."


From news at dorb.com  Sat Jan  1 11:15:15 2000
From: news at dorb.com (Darrell)
Date: Sat, 1 Jan 2000 11:15:15 -0500
Subject: pickle problem:cant reload object saved with binary option
References: <386e1de1.998928692@news-server>
Message-ID: <YHpb4.914$Sr1.104202@typhoon.nyroc.rr.com>

Try this:
        fp=open(fn,'r+b')


--
--Darrell
"Marshall" <python at scoobysnacks.com> wrote in message
news:386e1de1.998928692 at news-server...
> I created a class instance (self.info) to hold all my variables and
> I'm trying to save it for use in later sessions.  The binary file is
> much smaller but if I set the binary flag while saving, it blows an
> error when I try to load it.  I'm trying this on Win98 and I'm calling
> it from a Menu widget using command=self.save and I change the default
> bin=0 to bin=1 and it won't load anymore.  The load error is at the
> bottom.  Any direction is greatly appreciated.  Thanks.
>





From wtanksle at hawking.armored.net  Sun Jan  9 22:09:58 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 10 Jan 2000 03:09:58 GMT
Subject: control structures (was "Re: Sins")
References: <1264769486-16171098@hypernet.com> <38788F7E.76302733@maxtal.com.au> <3878B8EE.B4939F3A@connection.com>
Message-ID: <slrn87ijtu.thc.wtanksle@hawking.armored.net>

On Sun, 09 Jan 2000 11:35:58 -0500, Colin J. Williams wrote:
>skaller wrote:
>> Gordon McMillan wrote:

>> > That's the reason that Python has the "for
>> > .. else" construct that William Tanksley
>> > hates so much.

>> >  for i in X:
>> >    if X[i]==e:
>> >      # do the "found" thing
>> >      break
>> >  else:
>> >    print "Not Found"

>> That doesn't work. The 'do the "found" thing'
>> is in the wrong place: it must be outside the loop,
>> since it is usually the 'normal continuation'
>> of the routine.

>Aside from that.  It seems to me that the weakness of the
>   for ...
>   else 
>structure is that the else clause is executed whether the 
>'for' succeeds or not. The Language Reference Manual has:

>	"The expression list is evaluated once; it should yield a     
>sequence. The suite is then executed once for each item in the 
>sequence, in the order of ascending indices. Each item in turn 
>is assigned to the target list using the standard rules for 
>assignments, and then the suite is executed. When the items are 
>exhausted (which is immediately when the sequence is empty), the 
>suite in the else clause, if present, is executed, and the loop 
>terminates."

In case you were wondering, the reason Gordon mentions me above is that I
said the same thing you're about to.  I still agree with you; but neither
one of us is going to live to see the change we've proposed.

A better choice would be to advocate some other way of doing what's needed
-- labelled breaks would work superbly.  Even gotos would be better
(although I hope they're not the method chosen :-).

By the way, I think your reading of the current rules is wrong.  The else
clause will not be executed in case of a break -- it doesn't always get
executed, as you imply.

The 'try' statement has an 'else' clause with a similar mistake in
definition.  Argh.

>More useful would be to execute the else clause only if the sequence is
>empty.  This might be expressed as:

>    for x in L:
>	...
>    otherwise ...

I agree.

>Colin W.

-- 
-William "Billy" Tanksley, in hoc signo hack


From mgushee at havenrock.com  Mon Jan 17 18:31:45 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 17 Jan 2000 18:31:45 -0500
Subject: Tkinter: wm_protocol() error on RedHat Linux 6.0
References: <m3snzwtq5u.fsf@havenrock.com> <005101bf6136$a0670ea0$f29b12c2@secret.pythonware.com>
Message-ID: <m3embgtgv2.fsf@havenrock.com>

"Fredrik Lundh" <fredrik at pythonware.com> writes:

> for backwards compatibility, use the "protocol"
> alias instead.

Thank you!

... and looking at Tkinter.py, I gather the same must
be true for all the 'wm_' names.

Are there any other backwards-compatibility 'gotchas' I
should be aware of? Is there a list of this stuff anywhere?

Thanks for any info.
-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From lvirden at cas.org  Fri Jan 21 16:43:15 2000
From: lvirden at cas.org (lvirden at cas.org)
Date: 21 Jan 2000 21:43:15 GMT
Subject: Looking for Solaris 2.6/Python 1.5.2 make test fix
Message-ID: <86ajtj$8f8$1@srv38.cas.org>

When I configure then do a make all and then make test, Python core dumps:


rm -f /ldatae/Misc/python/Python-1.5.2/Lib/test/*.py[co]
PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py 
Bus Error - core dumped
make: [test] Error 138 (ignored)
PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py 
Bus Error - core dumped
make: *** [test] Error 138

$ dbx python core
anachronisms'.
Reading python
core file header read successfully
Reading ld.so.1
Reading libreadline.so
Reading libtix4.1.8.0.so
Reading libBLT24.so
Reading libjpeg.so.62
Reading libtk8.0.so
Reading libtcl8.0.so
Reading libX11.so.4
Reading libncurses.so.5
Reading libsunmath.so.1
Reading libm.so.1
Reading libz.so
Reading libsocket.so.1
Reading libnsl.so.1
Reading libdl.so.1
Reading libthread.so.1
Reading libc.so.1
Reading libjpeg.so
Reading libXext.so.0
Reading libmp.so.2
Reading libc_psr.so.1
Reading nss_files.so.1
Reading nss_nis.so.1
detected a multithreaded program
t at 1 (l at 15) terminated by signal BUS (invalid address alignment)
(dbx 1) where
current thread: t at 1
=>[1] t_delete(0x9c8da0, 0xeefa3100, 0x20202046, 0x9b8d98, 0x80f4b8, 0x10000), at 0xeef464dc
  [2] realfree(0x9b8da0, 0x9c8d98, 0xeefa9710, 0xeefa3100, 0x9b8d98, 0x10003), at 0xeef46198
  [3] cleanfree(0xa51110, 0xeefa3100, 0xeefa9684, 0xeefa9704, 0xeefa9684, 0xa51124), at 0xeef46994
  [4] realloc(0xa51110, 0xf22, 0xeefa3100, 0x4, 0xab, 0xa51110), at 0xeef45de4
  [5] _PyString_Resize(0xefffc838, 0xf0a, 0x0, 0x661378, 0x0, 0xeefa3100), at 0x4d628
  [6] PyZlib_flush(0x64f8a8, 0xe23c8, 0xd1b38, 0x144ec4, 0x6622c, 0xad9196), at 0x66020
  [7] call_builtin(0x125478, 0xe23c8, 0x0, 0x0, 0x0, 0x0), at 0x24a1c
  [8] PyEval_CallObjectWithKeywords(0x125478, 0xe23c8, 0x0, 0x2, 0xc779c, 0xff0000), at 0x248e8
  [9] eval_code2(0x1a2e54, 0x0, 0x1a2e6c, 0xe23c8, 0x0, 0x125478), at 0x23340
  [10] PyEval_EvalCode(0x65d4e8, 0x660d68, 0x660d68, 0x63000000, 0x6300, 0xff00), at 0x20c00
  [11] PyImport_ExecCodeModuleEx(0x126a70, 0x65d4e8, 0xefffcb28, 0x0, 0x0, 0xeefa3100), at 0x2c9ac
  [12] load_source_module(0xefffd508, 0xefffcb28, 0x65d4e8, 0xefffd014, 0x81010100, 0xeef8824c), at 0x2ced8
  [13] load_module(0xefffd508, 0xe0810, 0xefffd018, 0x1, 0xc78b8, 0x401), at 0x2d658
  [14] import_submodule(0xc779c, 0xefffd508, 0xefffd508, 0x62, 0x81010100, 0xefffd508), at 0x2e4c8
  [15] load_next(0xc779c, 0xc779c, 0xefffd508, 0xefffd508, 0xefffd504, 0x60), at 0x2dfd8
  [16] import_module_ex(0xefffd508, 0xe81b8, 0x64ec68, 0x1261c8, 0x64f970, 0x61), at 0x2dbdc
  [17] PyImport_ImportModuleEx(0x1423ec, 0xe81b8, 0x64ec68, 0x1261c8, 0xefffd9f4, 0xefffd9f0), at 0x2dd80
  [18] builtin___import__(0x0, 0x22a770, 0x4, 0x0, 0x0, 0xeefa3100), at 0x4f378
  [19] call_builtin(0xe1e28, 0x22a770, 0x0, 0x0, 0x0, 0xfdcc8), at 0x24a1c
  [20] PyEval_CallObjectWithKeywords(0xe1e28, 0x22a770, 0x0, 0x0, 0x2, 0x155cf8), at 0x248e8
  [21] eval_code2(0x155cbc, 0x1423d8, 0x155cf0, 0x22a770, 0x0, 0xe1e28), at 0x23340
  [22] eval_code2(0x142cb4, 0x0, 0x142d38, 0x4, 0x0, 0xf6aa0), at 0x231e4
  [23] eval_code2(0xede34, 0x0, 0xede3c, 0x0, 0x0, 0x152018), at 0x231e4
  [24] PyEval_EvalCode(0x1107e0, 0xe81b8, 0xe81b8, 0xefffde20, 0x81010100, 0xff0000), at 0x20c00
  [25] run_node(0xf1f48, 0x1107e0, 0xe81b8, 0xe81b8, 0x0, 0x0), at 0x33fc0
  [26] run_err_node(0xf1f48, 0xefffe461, 0xe81b8, 0xe81b8, 0xe0c40, 0xeef8824c), at 0x33f70
  [27] PyRun_File(0xe07e0, 0xefffe461, 0x101, 0xe81b8, 0xe81b8, 0x2e), at 0x33f44
  [28] PyRun_SimpleFile(0xe07e0, 0xefffe461, 0xec038, 0xe3e48, 0x0, 0x2f), at 0x3344c
  [29] PyRun_AnyFile(0xe07e0, 0xefffe461, 0xe07e0, 0x0, 0x2f, 0x2), at 0x33020
  [30] Py_Main(0xe0400, 0x0, 0xefffe1dc, 0xe0800, 0x0, 0xe07e0), at 0x1f094
  [31] main(0x2, 0xefffe1dc, 0xefffe1e8, 0xe0548, 0x0, 0x0), at 0x1ea98
-- 
<URL: mailto:lvirden at cas.org> <URL: http://www.usenix.org/events/tcl2k/>
<*> O- <URL: http://www.purl.org/NET/lvirden/>  Tcl2K - Austin, Texas, US
Unless explicitly stated to the contrary, nothing in this posting
should be construed as representing my employer's opinions.
-><-


From jblaine at shell2.shore.net  Thu Jan 13 13:16:31 2000
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Thu, 13 Jan 2000 18:16:31 GMT
Subject: Imaging library woes
References: <85i4gc$2es$1@nnrp1.deja.com>
Message-ID: <slrn87s5jv.i3.jblaine@shell2.shore.net>

Tell your compiler -I/usr/local/include/tk8.0

Simply using -I/usr/local/include does not state 'search for include files
recursively under that directory'.

>--------------------begin my pain-----------------------
>doug:~/Imaging-1.0# make
>gcc -fpic  -IlibImaging -I/usr/local/include -I/usr/local/include
>-DWITH_TKINTER -I/usr/local/include -g -O2
>-I/usr/local/include/python1.5
>-I/usr/local/include/python1.5 -DHAVE_CONFIG_H -c ./_imaging.c
>./_imaging.c:63: warning: `WITH_TKINTER' redefined
>*Initialization*:1: warning: this is the location of the previous
>definition
>./_imaging.c:2303: tk.h: No such file or directory
>*** Error code 1
>
>--------------------end my pain--------------------------
>tk.h does indeed exist:
>
>/usr/local/include/tk8.0/tk.h


From grant at nowhere.  Tue Jan 11 10:53:28 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 11 Jan 2000 15:53:28 GMT
Subject: Newbie simple question - can't find answer
References: <85fhha$6js$1@nnrp1.deja.com>
Message-ID: <slrn87mkfo.1d1.grant@grante.comtrol.com>

In article <85fhha$6js$1 at nnrp1.deja.com>, eleisonme at juno.com wrote:
>Hello all,
>I am using Python 1.5.2-7 on intel Red Hat Linux 6.1.
>I do not seem to be able to use any of the math or cmath functions
>either in a script or on the python command line.
>
>For example if I type:
>>>>import math
>>>>exp (4)

You need to tell it which module contains "exp":

>>> math.exp(4)
54.5981500331
>>> 


Your other option is to explicitly import "exp" from the math
module into the global namespace.  This is discouraged (esp. in
larger applications), since it could result in namespace
collisions and confusing in a reader's mind about where "exp"
comes from:

>>> from math import exp
>>> exp(4)
54.5981500331
>>> 


-- 
Grant Edwards                   grante             Yow!  -- I have seen the
                                  at               FUN --
                               visi.com            


From emile at fenx.com  Wed Jan 19 11:17:45 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 19 Jan 2000 08:17:45 -0800
Subject: Python Rocks!
Message-ID: <025401bf6298$b9a347e0$01ffffc0@worldnet.att.net>

I use an editor with visible tabs which helps to map
the logic.  When I tried spaces, I found it too easy
to my eye to jump columns in larger blocks and lose
the construct.

It seems the best advice for newbies is don't mix it up.

Now-I-make-more-smaller-pieces-anyway-ly y'rs

-- 

Emile van Sebille
emile at fenx.com
-------------------





From Lance.Ellinghaus at avnet.com  Thu Jan 20 01:00:04 2000
From: Lance.Ellinghaus at avnet.com (Lance Ellinghaus)
Date: Wed, 19 Jan 2000 22:00:04 -0800
Subject: Mapping Column Names to results using DBI
Message-ID: <8825686C.0020F844.00@avnet.com>

There are other reasons to do a "SELECT * FROM ..." and not "SELECT x,y,z FROM
..."

For example, if you have columns that are only valid depending on other values
in the returned row you would want to pull them all over and use only the ones
that are valid. When you are moving around NULL columns, there is very little
overhead.

I have a table that has over 75 columns and the columns that are valid are based
on a number of columns in the row. The other columns are NULL and thus take
minimal overhead to maintain and query against. That is why I wrote a that
little bit of code I posted earlier today to convert the returned row into a
dictionary. Makes it easier to use.

Lance

-------From: "Gordon McMillan"  on 01/19/2000 10:31:47 PM-------

From: "Gordon McMillan" <gmcm at hypernet.com>
To: Timothy Grant <tjg at avalongroup.net>, Python People <python-list at python.org>
cc:
Subject: Re: Mapping Column Names to results using DBI






Timothy Grant wrote:

> Aahz Maruch wrote:
>
> > Try "SELECT x, y, z" instead of "SELECT *".
>
> I knew it couldn't be that simple...
>
> I must be missing something

DBs return the columns in the order requested. So you will
always get tuples of (x,y,z).

"Select *" is like "from module import *" - something you
should really only do in interactive mode.

I'll agree that this creates a distasteful coupling between the
SQL code and the Python, but performance makes it more
palatable.



- Gordon

--
http://www.python.org/mailman/listinfo/python-list





From dnagata at creo.com  Sun Jan  9 00:44:02 2000
From: dnagata at creo.com (Dale Nagata)
Date: Sat, 08 Jan 2000 21:44:02 -0800
Subject: Compiling on Alpha-Win2K/WinNT (Python 1.5.2)
References: <850p0u$q6a$1@nnrp1.deja.com>
Message-ID: <38782022.1214@creo.com>

tmick at mail.com wrote:
> 
> I am trying to compile Python 1.5.2 (as extracted from py152.tgz on
> www.python.org) for Win2000 on an Alpha. Has anyone done this? I
> presume that it should be similar to compiling for Alpha/WinNT.
> 
[snip]
> 
> I do not know of a file called "fpu.h" and cannot find one on my system.
> 
> Any help would be appreciated. Thanks in advance.
> 

There are some bugs compiling Python on NT Alpha.
It correctly detects the processor type from a predefined macro,
but assumes it's Digital UNIX, not NT, so it tries to pull in the
wrong header.

I was able to build all of Python 1.5.2 plus (most of) the Win32
Extensions (Build 125) plus Sam Rushing's windll module on NT 4.0 Alpha
with Visual C++ 5.0. I haven't tried running it on W2K.

I submitted the Python 1.5.2 fixes for NT Alpha awhile ago, but never
heard that they got officially rolled into the standard codebase.

Email me if you want the fixes.


--
Dale Nagata         |  tel : +1 604.451.2700 ext. 2254 (UTC-0800)
Software Developer  |  fax : +1 604.437.9891 
Creo Products Inc.  |  pgr : +1 604.691.8279
Burnaby BC Canada   |  http://www.creo.com/


From volumeregeling at hotmail.com  Mon Jan 24 07:49:04 2000
From: volumeregeling at hotmail.com (volumeregeling)
Date: Mon, 24 Jan 2000 12:49:04 GMT
Subject: global variable problem
Message-ID: <388c496a.9657176@122.0.0.250>

Hi,


I am writing a program without classes.

I have the following :

Debugging = 0

def remote_debug():
      Debugging = 1



For some reasong the Debuggine in the function is not the same as the
global one , can you tell me how i can update the global variable from
within a function ????  
so i would like to update the global one when i am in the function and
not create a second Debugging which disapears when the functions dies.


thakns,

Volumeregeling.


From wlfraed at ix.netcom.com  Mon Jan 24 23:22:10 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Mon, 24 Jan 2000 20:22:10 -0800
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <388670D2.2FA842F6@starmedia.net> <Pine.SOL.3.91.1000120001532.29235B-100000@sunse> <tIeGOHk6cMDxz41yCwn96Iy45Dgx@4ax.com> <388CF7C4.D6247285@ndh.net>
Message-ID: <8RuNOAk+ivimZ5yVND181Kou==16@4ax.com>

On Tue, 25 Jan 2000 02:09:24 +0100, Stefan Schwarzer
<s.schwarzer at ndh.net> declaimed the following in comp.lang.python:

> Hello Dennis,
> 
> Dennis Lee Bieber schrieb:
> >         The only languages I've ever seen that use a text string for
> > those operators are FORTRAN and VMS DCL.
> 
> The only programming languages I (personally) know to pay attention
> to amount of whitespace are Fortran, Assembler and Python ;-) .
>
	So far as I know standard Fortran 77 is still white space
indifferent -- ignoring the old card punch parts: columns 1-5 for
statement labels, and column 6 for continuation, and the last 8 (of 80)
for sequence/comment. Most compilers these days will accept a single tab
as getting you past the continuation column, and the VMS compiler will
interpret "<tab><digit>" to mean "continuation". (As far as my coding
style -- start everything with a <tab> except labels and continuations,
in which I still use column 6 for continuation){disclaimer: Fortran 90
now if fully free format, using "&" at the end of the continued line,
rather than a marker on the continuation line}

	Within a statement line, however, spaces only matter in string
constants.

	i f ( 1 00    .0 .   l      t.   a bs (x    y     z ) ) t he n

is the same as

	if (100.0 .lt. abs(xyz)) then

	Fortran is not an easy language to parse -- it has to reach the
decimal point or comma to determine which of the following is an
assignment and which is a do-loop:

	do	10	i	=	1 	. 	35
	do	10	i	= 	1	,	35

(at which point it has to back track to either reference a floating
point variable "do10i" or a loop index "i" with statement label "10").

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From robin at jessikat.demon.co.uk  Sat Jan  1 01:08:55 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 1 Jan 2000 06:08:55 +0000
Subject: fun with stock quotes
References: <Fnn296.A3w@world.std.com>
Message-ID: <ncREqFA3nZb4Ew1G@jessikat.demon.co.uk>

In article <Fnn296.A3w at world.std.com>, Will Ware <wware at world.std.com> writes
>I recently discovered that Yahoo's stock-quote-serving web
>page is now offering historical stock data in machine-friendly
>form. The following scripts will pull in a bunch of stock data
>and format it as a C module. Then you can back-test any silly
>investment algorithms you come up with. Nerdy greedy fun for
>the whole family.
>
here's my take on that which just returns a class with all the values
######################################################
import urllib, string, normalDate, math, regex, sys, os

class _YahooData:
    def __init__(self,ticker,start,finish,Open,Low,High,Close,Volume):
        self.ticker=ticker
        self.start=start
        self.finish=finish
        self.Open=Open
        self.Low=Low
        self.High=High
        self.Close=Close
        self.Volume=Volume

def fetchRangeAsCSV(ticker, first, last):
    '''go to Yahoo and fetch a range of historical stock
    data, where 'first' and 'last' are normalDates'''
    key = (first, last)
    url = 'http://chart.yahoo.com/table.csv?s=%s' \
          '&a=%d&b=%d&c=%d&d=%d&e=%d&f=%d&g=d&q=q&y=0&z=%s' \
          '&x=.csv' % (ticker,
                       first.month(), first.day(), first.year(),
                       last.month(), last.day(), last.year(),
                       ticker)
    return urllib.URLopener().open(url).readlines()

def _yahoo_to_normalDate(date):
    date = string.split(date, '-')
    date[0] = eval(date[0])
    date[2] = eval(date[2])
    if date[2] > 90: year = date[2] + 1900
    else: year = date[2] + 2000
    month = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep',
            'Oct','Nov','Dec'].index(date[1]) + 1
    return normalDate.ND(year * 10000 + month * 100 + date[0])

def fetchRange(ticker, first, last):
    Open = []
    Low = []
    High = []
    Close = []
    Volume = []
    finish=None
    for x in fetchRangeAsCSV(ticker,first,last)[1:]:
        F = string.split(x, ',')
        Open.append(eval(F[1]))
        Low.append(eval(F[2]))
        High.append(eval(F[3]))
        Close.append(eval(F[4]))
        Volume.append(eval(F[5]))
        if finish is None:
            finish=_yahoo_to_normalDate(F[0])

    start = _yahoo_to_normalDate(F[0])

    # Yahoo gives reverse time order
    Open.reverse()
    Low.reverse()
    High.reverse()
    Close.reverse()
    Volume.reverse()
    if len(data) == 0:
        raise 'ouch'
    return _YahooData(ticker,start,finish,Open,Low,High,Close,Volume)

if __name__=='__main__':
    # just do a quick test for now
    masterList = [
        # tech stocks
        'LNUX', 'MSFT', 'AMZN', 'INTC', 'RHAT', 'DELL', 'GTW',
        'CSCO', 'AAPL', 'AOL', 'COMS', 'CPQ', 'LCOS', 'LU',
        'ORCL', 'SUNW', 'TXN', 'YHOO',
        # single-letters and blue chips
        'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
        'J', 'K', 'L', 'N', 'O', 'P', 'R',
        'U', 'W', 'X', 'Y', 'Z',
        # Dow Jones stocks
        'AA', 'AXP', 'T', 'BA', 'CAT', 'CHV', 'KO',
        'DIS', 'DD', 'EK', 'GE', 'GM', 'GT', 'HWP',
        'IBM', 'IP', 'JNJ', 'MCD', 'MRK', 'MMM', 'JPM',
        'MO', 'PG', 'S', 'TRV', 'UK', 'WMT',
        # biotech
        'AFFX', 'BIOM', 'CRA', 'ENMD', 'HGSI',
        'MATX', 'MLNM', 'RGEN', 'SAFS', 'DNA', 'GENE',
        'INCY', 'GLGC', 'PFE',
        'AVXT', 'CTIC', 'CRXA', 'GZMO', 'IDPH',
        'ILXO', 'ISIP', 'RZYM', 'SUPG', 'TGEN', 'VICL',
        # retailers, general interest
        'BBY', 'COST', 'UPS', 'HD',
        ]
    masterList = masterList[:12]
    start = normalDate.ND(19950101)
    finish = normalDate.ND()   # today
    d = {}

    for t in masterList:
        d[t]=fetchRange(t, start, finish)
        print   "data for %s obtained start=%d(%f) finish=%d(%f)" % \
                (t,d[t].start,d[t].Close[0],d[t].finish,d[t].Close[-1])
-- 
Robin Becker


From sjturner at ix.netcom.com  Wed Jan 12 10:01:03 2000
From: sjturner at ix.netcom.com (Stephen J. Turner)
Date: Wed, 12 Jan 2000 10:01:03 -0500
Subject: bdb: which class?
References: <B9AA1D5DA95ED211900100805F6FFF9C02F3D8F6@exbhbhmb01.us.dbisna.com>
Message-ID: <387C972F.E4AC9F32@ix.netcom.com>

"Zakon, Stuart" wrote:
> So far I am using the following within a subclass of Bdb:
> 
>         def where_am_i (self, frame):
>                 name = frame.f_code.co_name
>                 filename = frame.f_code.co_filename
>                 lineno = frame.f_code.co_firstlineno
>                 self.here_i_am = '%s, %s, %d' % (name, filename, lineno)
> 
> I would like to have the class name for methods of classes which are being
> called but I didn't see it in the documentation for code objects.
> Anybody know if this can be found?

How about the following:

def where_am_i(self, frame):
    from types import InstanceType, MethodType
    code = frame.f_code
    name = code.co_name
    if code.co_argcount:
        arg0 = frame.f_locals[code.co_varnames[0]]
        if type(arg0) is InstanceType:
            attr = getattr(arg0, name, None)
            if attr and \
               type(attr) is MethodType and \
               attr.im_func.func_code is code:
                classname = arg0.__class__.__name__
    ....

If the current frame is for a method invocation, then its corresponding
code object must have at least one argument, and the first named
argument (probably `self', but we can't assume) identifies the class
instance.  This name is then used as an index into the frame's `locals'
dictionary to get the first argument.

Since, however, the frame could also be for a function invocation, we
make sure that (1) the type of the first argument is a class instance,
(2) the instance has an attribute whose name matches the code object's,
(3) the instance attribute is in fact a method, and (4) the method's
underlying code object is the same as the frame's.  If all these are
true, then we get the class name from the instance.

Regards,
Stephen

--
Stephen J. Turner <sjturner at ix.netcom.com>


From skip at mojam.com  Tue Jan 25 18:44:47 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 25 Jan 2000 17:44:47 -0600 (CST)
Subject: anydbm - a simple question
In-Reply-To: <388E2BF3.8CCF5849@inka.de>
References: <20000121110715.A184793@vislab.epa.gov>
	<3diu0lbz8e.fsf@amarok.cnri.reston.va.us>
	<20000124112320.A440571@vislab.epa.gov>
	<388CE28F.D3A100AB@inka.de>
	<14477.47754.65118.177514@beluga.mojam.com>
	<388E2BF3.8CCF5849@inka.de>
Message-ID: <14478.13679.746001.81905@beluga.mojam.com>

    Michael> I did not find a "Lib" directory or similar under

    Michael>   http://www.musi-cal.com/~skip/python/

    Michael> Can you post the exact URL? Thank you.

Ack!  Should have been

    http://www.musi-cal.com/~skip/python/Python/

Apologies...

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From skip at mojam.com  Thu Jan 20 15:04:34 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 20 Jan 2000 14:04:34 -0600 (CST)
Subject: How to store a function pointer in class?
In-Reply-To: <867nbb$1ho$1@mordred.cc.jyu.fi>
References: <867j5p$qvq$1@mordred.cc.jyu.fi>
	<vndaem01s1r.fsf@betelgeuse.ccs.neu.edu>
	<867nbb$1ho$1@mordred.cc.jyu.fi>
Message-ID: <14471.27218.299267.375864@beluga.mojam.com>

    >> A class or instance variable may be a reference to a function object,
    >> though.

    Sami> If wish they could! The problem is that a class variable can NOT
    Sami> refer a function object, only an unbound method object. 

Check out the optional new module (distributed with Python, but not enabled
by default, at least not in the past), especially new.instancemethod.  It
takes a function object, an instance and a class and returns the appropriate
goody.  I've never had occasion to use it in a real situation, but the
following trivial code works:

    import new

    class Foo:
	def addmethod(self, name, func):
	    setattr(foo, name, new.instancemethod(func, self, Foo))

    foo = Foo()

    def sqr(self, a):
	return a*a

    foo.addmethod("sqr", sqr)

    print foo.sqr(4)

Cheers,

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From gerrit.holl at pobox.com  Thu Jan 27 09:55:51 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 27 Jan 2000 15:55:51 +0100
Subject: nesting class definitions?
Message-ID: <20000127155551.A2069@stopcontact.palga.uucp>

Hello,

I do see the point in nesting function definitions, namely as an alternative
on places where one could use lambda. But what's the point in nesting
class definitions? I know it's possible, but I've never seen it used.

There aren't functions in the default library that take a class as
argument, are there?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From jblaine at shell2.shore.net  Mon Jan  3 10:40:42 2000
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Mon, 03 Jan 2000 15:40:42 GMT
Subject: Python Books & improving PyAwareness - How to ?
References: <01bf5597$3dfee1c0$ddc854a6@brancusi.pro-net.co.uk>
Message-ID: <slrn871gnq.h2o.jblaine@shell2.shore.net>

>I noted that ther was no Python books in Linux sections - which I think is
>rather surprisin and dissappointing, especially given the enthusiasm now
>for Linux sections in these stores.

Why would you expect to find Python books in the Linux sections?  It's
a programming language available for a pile of operating systems.  Python
books belong next to Perl books (ideally with a THICK steel divider between
the two).


From m.faassen at vet.uu.nl  Wed Jan 19 11:20:37 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Jan 2000 16:20:37 GMT
Subject: a small problem
References: <20000112015923.A8450@xs4all.nl> <000701bf5cb4$c72320a0$da2d153f@tim> <20000112145107.B8450@xs4all.nl> <20000112165932.A2758@stopcontact.palga.uucp> <20000113130514.A21652@xs4all.nl> <20000113153612.A1808@stopcontact.palga.uucp>
Message-ID: <864o8l$dv6$1@newshost.accu.uu.nl>

Gerrit Holl <gerrit.holl at pobox.com> wrote:
> I don't know any Perl, but I can't understand how to
> write a function that takes any kind of argument if you need to type
> such a prefix...

I have scant knowledge of Perl (being Saved by Python just in time), but
as far as I recall the $ @ stuff can be used to somehow make a particular
variable behave like a number, list, or whatever. So I imagine it allows
arbitrary arguments which you then may analyze and then treat in the way
you desire. Or something. My mind never really wrapped around the concept.

Regards,

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


From neelk at brick.cswv.com  Sun Jan  2 21:50:17 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 3 Jan 2000 02:50:17 GMT
Subject: Python language spec
References: <E124x1i-00053N-00@ethereal>
Message-ID: <slrn8703jp.g72.neelk@brick.cswv.com>

Nolan Darilek <nolan_d at bigfoot.com> wrote:
> 
> Here's where my confusion lies. Maybe I'm studying the language
> reference and taking it too literally. I'm working slowly through,
> section-by-section, writing rules for each construct and testing
> them.

If what you are really interested in is the grammar, you should really
start with the Grammar file in the Python source distribution. It
defines the grammar of the language in a short, succint file. Save the
language reference for the second phase -- once you have a vague idea
of what all the pieces of the grammar are you might be less distracted
by the minutiae of the grammar definition.

> I'm confused by the expression_list definition, however. It reads:
> 
> expression_list:      expression ("," expression)* [","]
>
> I understand the syntax of the definition, but I don't know what the
> syntax for an expression should be. Earlier rules define u_expr,
> a_expr, etc. but not a regular expression (I don't mean regexp, just a
> no-frills expression :). The only rule which defines 'expression' is:

Here's the definition of expression from the grammar file, plus some
more productions that it references.

  expr: xor_expr ('|' xor_expr)*
  xor_expr: and_expr ('^' and_expr)*
  and_expr: shift_expr ('&' shift_expr)*
  shift_expr: arith_expr (('<<'|'>>') arith_expr)*
  arith_expr: term (('+'|'-') term)*
  term: factor (('*'|'/'|'%') factor)*
  factor: ('+'|'-'|'~') factor | power
  power: atom trailer* ('**' factor)*
  atom: '(' [testlist] ')' | '[' [testlist] ']' | '{' [dictmaker] '}' |
  	'`' testlist '`' | NAME | NUMBER | STRING+

Now, a reasonable question to ask is: why are there so many different
kinds of expression when they all seem to reduce to the same thing? 
(This seemed to me to be the thing that was confusing you. Let me know
if I misunderstood.)

The answer is that different operators have different precedences, and
having separate rules is a way of ensuring that the parser will emit a
syntax tree that properly observes the operator precedences that are
expected.  A simple way of getting an intuition about this is to write
down some arithmetic expressions and then manually reduce them using
the rules of the grammar.

For example, try parsing the expressions "0 & 0 | 1" and "3 + 5 * 7".


Neel


From ivanlan at callware.com  Wed Jan 26 18:57:11 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 26 Jan 2000 16:57:11 -0700
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com>
Message-ID: <388F89D7.96DD8DCC@callware.com>

Hi All--

Doug Landauer wrote:
> 

[snip]

>  Notice that
> it is, for the most part, *non*-users of python who complain about
> the indentation.  _Users_, for the most part, like it.  
> ***Languages
> that change by catering to the tastes of non-users tend not to do
> so well.***
> 

You'd think, wouldn't you, that this last little rule (emphasis added)
would not need to be pointed out to people?  Luckily, Guido seems to
have a firm grasp on it.

<no-doctor-they-didn't-warn-me-not-to-put-my-fingers-in-the-lawnmower>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From scherbi at bam.com  Mon Jan  3 14:58:00 2000
From: scherbi at bam.com (Bill Scherer)
Date: Mon, 03 Jan 2000 14:58:00 -0500
Subject: class assignment copy
References: <84qtpn$lr1$1@nnrp1.deja.com>
Message-ID: <3870FF48.B7F0011D@bam.com>

You need the copy module. see

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


tiddlerdeja at my-deja.com wrote:

> Sorry for the lame question.
>
> If I have a class Foo:
>
> class Foo:
>     def __init__(self):
>         self.bar = 1
>
> If I create one instance of the class:
>
> x = Foo()
>
> How do I make a copy of x?
>
> y = x
>
> Just makes y refer to x, so if I change bar on y the value is also
> changed on x.
>
> I realise this is a lame question, I just couldn't find any python
> material on this.
>
> Do I need to create some copy constructor on the class?
>
> Any help appreciated.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Bell Atlantic Mobile
914 365 7265


-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000103/e2632bd1/attachment.vcf>

From quinn at auto.ugcs.caltech.edu  Fri Jan 21 17:33:56 2000
From: quinn at auto.ugcs.caltech.edu (Quinn Dunkan)
Date: 21 Jan 2000 22:33:56 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de> <869uel$suj$1@nntp1.atl.mindspring.net>
Message-ID: <slrn88hnpb.brp.quinn@auto.ugcs.caltech.edu>

On 21 Jan 2000 15:36:53 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>In article <m3n1q0fzg7.fsf at greebo.nodomain.de>,
>Bernhard Herzog  <herzog at online.de> wrote:
>>
>>curried functions
>
>All right, I've seen it enough times this week, would somebody mind
>giving the Stupid Person's definition of currying?

A curried function always takes one argument.  If it needs more values to
'finish', it returns a closure which will accept the next arg.  In python:

>>> def add(a):
...    return lambda b, a=a: a + b
...
>>> add(4)
<function <lambda> at 100b41f8>
>>> add(4)(5)
9

A simple example of how this can be useful:

>>> map(add(5), range(5))
[5, 6, 7, 8, 9]

Many functional languages such as the ml family and haskell have all functions
curried by default, including +, *, etc. (there are no two-argument
functions... of course, you could pass a _single_ tuple of two elements :))

Hugs> map (5+) [0..4]
[5,6,7,8,9]

Whenever you know some args for a function now, and the rest later, you can
partially apply it and toss the closure around.  It's very useful if your
brain is working in the appropriate mode.


From goatesque at my-deja.com  Wed Jan 12 09:53:07 2000
From: goatesque at my-deja.com (goatesque at my-deja.com)
Date: Wed, 12 Jan 2000 14:53:07 GMT
Subject: Imaging library woes
Message-ID: <85i4gc$2es$1@nnrp1.deja.com>

I am having a problem getting the Imaging libs to compile properly under
OpenBSD 2.6:

--------------------begin my pain-----------------------
doug:~/Imaging-1.0# make
gcc -fpic  -IlibImaging -I/usr/local/include -I/usr/local/include
-DWITH_TKINTER -I/usr/local/include -g -O2
-I/usr/local/include/python1.5
-I/usr/local/include/python1.5 -DHAVE_CONFIG_H -c ./_imaging.c
./_imaging.c:63: warning: `WITH_TKINTER' redefined
*Initialization*:1: warning: this is the location of the previous
definition
./_imaging.c:2303: tk.h: No such file or directory
*** Error code 1

--------------------end my pain--------------------------
tk.h does indeed exist:

/usr/local/include/tk8.0/tk.h

Any clues?

Thanks.

d



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


From m.faassen at vet.uu.nl  Wed Jan 19 16:54:51 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Jan 2000 21:54:51 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com>
Message-ID: <865brb$dpm$1@newshost.accu.uu.nl>

Okay, just _one_ serious post then. Sorry all to contribute to this. :)

tye4 <tye4 at yahoo.com> wrote:
[someone gives an example of C's delimiter problem]
> Perfect example.

While yours is bogus:

>    if (a > 3)
>      if (b == 4)
>         c = 2;
>     else
>      c = 3;

In actual semi-Python this example would look like this:

a, b, c = 1, 2, 3 # added to avoid NameErrors 

if a > 3:
  if b == 4:
     c = 2
 else:
  c = 3

Execute it:

inconsistent dedent
  File "test.py", line 6
    else:
        ^
SyntaxError: invalid token

> If you add a space, the 'else' will belong to the inner 'if'.

Ah, I get:

  File "test.py", line 7
    c = 3
    ^
SyntaxError: invalid syntax

> If you delete one, the 'else' will belong to the outer one.

Wow, this variety actually doesn't give a syntax errors. This is because
the indentation lines up.

> Highly error prone.

Yes, the Python interpreter doesn't like your example and complains.

Have you actually _tried_ Python? Indenting takes some getting used to,
and perhaps you're one of those people that will still dislike it after
having tried, but give it a shot first, at least?

Regards,

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


From MSteed at altiris.com  Wed Jan 19 11:04:33 2000
From: MSteed at altiris.com (Mike Steed)
Date: Wed, 19 Jan 2000 09:04:33 -0700
Subject: Python Rocks!
Message-ID: <65118AEEFF5AD3118E8300508B124877073D21@webmail.altiris.com>

> From: Sposhua [mailto:sposhua at my.pc]
> Sent: Wednesday, January 19, 2000 4:37 AM
> To: python-list at python.org
> Subject: Re: Python Rocks!
> 
> 
> On 18 Jan 2000, Justin Sheehy wrote:
> 
> > > ! - whats wrong with tabs?
> > 
> > They don't look the same in every editor, or with each individual's
> > tab-width settings.
> 
> !? If you use ONLY tabs and not mix it with spaces (who the 
> heck wants to jit
> the space bar 4 times when U can hit tab once?) surely all 
> will be well, or have
> I missed something...?

I think what you have missed is the separation between indent width and tab
stop.  For example, I can configure Vim to insert 4 spaces each time I hit
Tab, and "outdent" 4 spaces when I hit backspace.

Even the MS Visual Studio editor supports this idea, so I would be surprised
if *any* programmers editor did not.

--
M.



From pinard at iro.umontreal.ca  Thu Jan 20 19:24:19 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 20 Jan 2000 19:24:19 -0500
Subject: Python Rocks!
In-Reply-To: "Tim Peters"'s message of "Wed, 19 Jan 2000 23:54:06 -0500"
References: <000901bf6302$6134d500$31a2143f@tim>
Message-ID: <oqpuuwb7bg.fsf@titan.progiciels-bpi.ca>

"Tim Peters" <tim_one at email.msn.com> writes:

> Programs-writing-programs is a twist of mind conspicuously absent in
> the Python world.

It has probably been debated to death already, and I have a folder full
of suggestions about scanners and parsers for Python, waiting for me to
read and study, which I gleaned from this very mailing list.

Isn't there a need for compiling grammars into fast scanners and parsers,
a bit like Flex and Bison do for C people?  To be happy, I presume one would
have to write small amounts of Python code for each "action".  Or maybe just
provide the name of a function to call.  Yet, the result has to be Python.

Or are you suggesting that such tools should directly generate byte-code?

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





From sigue at thecia.net  Mon Jan 17 15:25:49 2000
From: sigue at thecia.net (Carl L. Gay)
Date: Mon, 17 Jan 2000 20:25:49 +0000
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim>  	 <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov>  	 <20000111114923.B42270@vislab.epa.gov> <387B6733.2C49FB8A@home.com> <p6qd7r7hycx.fsf@informatik.hu-berlin.de>
Message-ID: <38837ACD.CA7F0CEF@thecia.net>

Martin von Loewis wrote:
> 
> Doug Hellmann <doughellmann at home.com> writes:
> 
> > How is this different from function overloading, as implemented in C++?
> > I'm not a C++  user, but my impression is the effect would be the same.
> > Is that the point?
> 
> No. Function overloading in C++ is a static thing. You look at the
> static types, and at all statically visible methods, and then chose
> the "best" one (in C++, this considers conversions from argument types
> to parameter types as well).
> 
> In multi-methods, lookup is dynamic. You don't do anything statically,
> but defer all decisions to run-time. Then, you care only about the
> dynamic types of the arguments.

I may be taking this post out of context because I didn't see
the previous posts in this thread, but taken on its own the
last paragraph is a bit of an overgeneralization, at least if
intended to refer to more than just Python.  In Dylan, for 
example, the compiler is frequently able to resolve multi-method
dispatch at compile time if there is enough type information
provided by the programmer.  But everything _is_ defined in
terms of run-time semantics, so the effect is the same, modulo
speed.

My apologies if I've taken you out of context.

-Carl (who doesn't normally read this newsgroup)


From rgacote at AppropriateSolutions.com  Wed Jan 26 07:58:38 2000
From: rgacote at AppropriateSolutions.com (Raymond Cote)
Date: Wed, 26 Jan 2000 07:58:38 -0500
Subject: Python Rocks!
In-Reply-To: <388BB4E1.974E63B7@prescod.net>
References: <000d01bf6312$77edd660$31a2143f@tim>
 <388BB4E1.974E63B7@prescod.net>
Message-ID: <v0422080ab4b49ea2e7a6@[192.168.24.43]>

>Wouldn't it be easier if we just added an optional, (but
>compiler-recognized if present) block delimiter to Python 2?

Reminds me of when I first moved from Pascal to C (oh so many years ago).
My first programs had constructs such as:
   #define begin {
   #define end }

so I could make my C code look 'sensible' like a language should. :}
Obviously, that didn't last long. I learned to use the language the way it was meant to be used and in such a way that other programmers could quickly understand what it was I was doing.

All languages have their weird eccentricities. Indenting happens to be one of Python's. I'll admit I was leery of it when I started, but after just a few weeks, I find my C++ and Pascal code looking wordy and redundant.

Ray
-------------------------------------------------------------------
Raymond Cote, President                 Appropriate Solutions, Inc.
www.AppropriateSolutions.com       rgacote at AppropriateSolutions.com
603.924.6079(v)  POB 458, Peterborough, NH 03458    603.924.8668(f)



From jjlucsy at concentric.net  Thu Jan 27 16:07:10 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Thu, 27 Jan 2000 16:07:10 -0500
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com> <86q5vv$mci$1@nnrp1.deja.com>
Message-ID: <Qq2k4.61$RU4.2585@news4.usenetserver.com>

It's new in MSVC6.0. If you use MSVC5, you won't be able to find it.

- Joel Lucsy (jjlucsy at concentric.net)







From katz at glue.umd.edu  Fri Jan 21 21:28:55 2000
From: katz at glue.umd.edu (Roey Katz)
Date: Sat, 22 Jan 2000 02:28:55 GMT
Subject: Python dumps core with this.. how come?
Message-ID: <38891316.393423@news.erols.com>

I'm running Red Hat Linux 5.2, kernel 2.0.36 with 128 MB of memory. 
"Python 1.5.2 (#2, Jun 13 1999, 17:00:15)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam"

When I import this module, the interpreter segfaults. 
Thiis module provides a little guard mechanism to keep track of
changes to registered variables. The error occurs when you ty to
instantiate class 'basic_monitor'. Also, I have another question (very
old, and I can't find this in the FAQ):  I want to explicitly take a
reference to an integer.  Take  __setattr__() for example:  is origval
a reference or merely a copy of self.res[ val ]? that is, do I change
self.res[ val ] by changing origval? Will Python2 implement some 
sort of epxlicit referencing operator so that these situations won't
seem as confusing? 
Whatever happened to 
   var :=  val   
for assignment, and
  var2 = &var1
for referencing? The addition of these explicit facilities would
greatly relieve a large portion of my Python problems (is there some
sort of rule-list to follow to figure out when exactly Python will
take a reference (like assigning from an object) or a literal value
(int's)? ). 

Thanks :)
Roey
===================================================


"""module baseMonitor"


class resource_entry:
    def __init__( self, minval, maxval, min_callback, max_callback ):
        self.minval = minval
        self.maxval = maxval
        self.min_callback = min_callback
        self.max_callback = max_callback


class basic_monitor:

    def __init__( self ):
        self.xres = []     # list of resource entries

    def track( self, name, minval, maxval, 
	min_callback, max_callback ):

        self.res[ name ] = resource_entry( minval, maxval,
		 min_callback, max_callback ) 


    def __getattr__( self, name ):
        return self.res[ name ]

    def __setattr__( self, name, val ):
        origval = self.res[ name ]
        if minval < val < maxval:
            origval = val
        elif val  <  minval:
            origval.min_callback()
        elif val  >  maxval:
            origval.max_callback()


if __name__=="__main__":
    m = basic_monitor()




From tylere at greenvillenc.com  Sat Jan  1 19:41:07 2000
From: tylere at greenvillenc.com (Tyler Eaves)
Date: Sat, 1 Jan 2000 19:41:07 -0500
Subject: Finding where Python is on a Unix server through Telnet?
Message-ID: <V8xb4.85295$Ym1.1550463@tw11.nn.bcandid.com>

What is the command?

Tyler




From tim_one at email.msn.com  Fri Jan 14 17:53:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 14 Jan 2000 17:53:17 -0500
Subject: Tcl/Tk Path problems in Python 1.5.2 installation
In-Reply-To: <387F80E5.A0AF82A5@sourcesignal.com>
Message-ID: <000901bf5ee2$25e44540$a0a2143f@tim>

[David Nichols]
> I just installed python 1.5.2 (running py152.exe and
> answering 'yes - I want to install Tcl/Tk')

Would help to know which OS; Win95/98 are generally more trouble than NT.
I'm guessing you're not using NT.

> and have been unable to get the path set up correctly
> for graphical applications.  For instance, if I go to the
> idle directory (//E/Programs/Py152/Tools/idle) and try to
> launch it I get the following result.

That's one clue:  you installed Python into a "non-standard" location.  So
you probably put Tcl/Tk somewhere unexpected too.

> ============================================================
> BASH.EXE-2.02$ python idle.py

OK, you're not running one of the Windows shells.

There's also an implication that you've fiddled your system path in some way
(else python would not have been found from a prompt in the IDLE directory).

To make life easier on yourself, don't try to debug this by running IDLE;
instead get into an interactive cmdline shell Python and type

>>> import Tkinter
>>> Tkinter._test()

You'll know that this is working (and then IDLE will too) when a little
window pops up in response.

In fact, there's no evidence of a problem with Python, so you could probably
cut Python out of the loop entirely and just try to get Tcl's Wish shell
working instead (Start -> Programs -> Tcl -> Wish).

> Traceback (innermost last):
> ...
> TclError: Can't find a usable init.tcl in the following
> directories:
>     {} ./lib/tcl8.0
>     .
>     E:/Programs/Py152/tcl8.0/library
>     E:/Programs/Py152/Tools/library

This msg comes from Tcl, so Python found Tcl, and Tcl *started* to execute.
Tcl is telling you which directories it looked in for one of its own startup
files.   So you have to change Tcl's view of which directories to look in --
or change your directories to match Tcl's view.

> The directories in sys.path are as follows ...
> E:\PROGRAMS\PY152
> [etc]

These don't matter (at least not at this point):  Tcl is a separate
language, and doesn't know anything about Python's sys.path.  As the msg
said, "This probably means that Tcl wasn't installed properly" -- which, of
course, doesn't mean that it wasn't installed properly <wink>.

> Tcl in installed in E:\PROGRAMS\Py152\\Tcl, and init.tcl
> exists in
> E:\PROGRAMS\Py152\Tcl\lib\tcl8.0 ...
> but this directory does not seem to be searched.

Right, the Tcl msg above did not list this directory.

> Do I need to set additional environmental variables or
> make changes to the registry beyond those made by the
> installer?

It's hard to say without rummaging around your machine.  The "init.tcl"
flavor of failure is rare, and *may* be due to having installed some other
version of Tcl in the past.  Scriptics has changed their mind about where
Tcl "should be" installed, and older installations leave crucial Tcl DLLs in
the system directories where they may confuse later installations.

A description of Tcl envars you can try setting:

    http://www.pythonware.com/people/fredrik/fyi/fyi02.htm

If that doesn't fix it, I would uninstall Python, Tcl and Tk (using
Add/Remove Programs); ruthlessly track down and destroy (at least rename to
harmlessness) every file matching tk*.dll and tcl*.dll on your machine; and
try again.

it-will-work-in-the-end-no-matter-how-long-that-takes-ly y'rs
    - tim





From tim_one at email.msn.com  Sun Jan  9 19:51:55 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 9 Jan 2000 19:51:55 -0500
Subject: fastest way to merge lists.
In-Reply-To: <85ab3d$h4q$1@nntp3.atl.mindspring.net>
Message-ID: <000001bf5b04$e4e3e2c0$842d153f@tim>

[Tim, replying to RoyS]
> You can save a little typing via e.g.
>
> for s in list1, list2, list3, list4, list5:
>     for x in s:
>         d[x] = 1

[Aahz Maruch]
> In practice, I know that the integer "1" is a cached,
> unique object, but that's an implementation detail that I
> prefer not to rely on.

You could force this issue by doing e.g. "one = 1" before the loop nest and
referring to "one" within it, but, as you know, that doesn't make any
difference today (& very unlikely it ever will -- not only is "1" (along
with some other small integers) globally unique today, but within a code
block all numeric literals of any kind (strings too, BTW) are optimized to
uniqueness within the block).  So Guido would have to lose his mind twice
over for this to change.

For a surprising reason given below, you may actually want to use this
assignment trick if you use None.

> I therefore almost always would use "d[x] = None"; that's
> particularly true if I *might* want to attach data to it
> later, in addition to making it clear that I'm interested
> only in the dictionary keys.

Fine by me!  Guido usually uses None too.  I used 1 partly because Roy did,
but I confess it's a habit I carried in from other languages.  For example,
before Icon grew a distinct "set" type, the idiomatic way to do sets was to
use Icon's flavor of dict with a default value of 0, and map the items in
the set to 1.

Brief digression:  when you create an Icon dict, you can specify a default
value D, which is any old object.  Then an Icon dict lookup works like the
Python d.get(lookitup, D).  You get D on key absence even if you would
rather have gotten an error, so it's a dubious feature, but it can be
convenient (indeed, that's why Python's dicts grew a .get() method!).

Anyway, the idea of representing set membership by a "true" value and
absence by a "false" one got so deeply ingrained, that even now when I see
"d[key] = None", I mechanically think "OK, so they're taking key *out* of
the set now"(!).  This is why Python should be the only language people are
allowed to learn <wink>.

Curiously, Roy has an actual reason to use 1 (or something else other than
None), because he's specifically looking for speed.  "1" is fetched up by
index from the local constant vector, but "None" is a builtin (not a
keyword) so first suffers a failing lookup in the global dict before a
successful lookup in the builtin dict.  So each dynamic occurrence of

    d[whatever] = None

actually triggers 3(!) dict lookups (the obvious one, plus two hidden ones
to find the None object).

The effect varies across platforms, but can be significant.  Run the
attached to see what it's like on yours (on my P5-166, using 1 is a bit over
25% faster *overall* than using None, via the usual speedup = (slower -
faster) / faster * 100).  Using the "assignment trick" to create a local
name for None eliminates the speed difference.

just-three-of-the-millions-of-fascinating-None-stories<wink>-ly
    y'rs  - tim


def timeit(f, n=10):
    from time import clock
    start = clock()
    for i in range(n): f()
    finish = clock()
    print f.__name__ + ":", round(finish - start, 4)

indices = range(50000)

def timeNone():
    d = {}
    for i in indices:
        d[i] = None

def timeOne():
    d = {}
    for i in indices:
        d[i] = 1

def timeLocalNone(None=None):
    d = {}
    for i in indices:
        d[i] = None

timeit(timeNone)
timeit(timeOne)
timeit(timeLocalNone)





From glen at electricorb.com  Fri Jan 28 07:27:51 2000
From: glen at electricorb.com (Glen Starchman)
Date: 28 Jan 2000 06:27:51 -0600
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be>
Message-ID: <38918A29.8AB61742@electricorb.com>

The forced indentation in python does something that no amount of cajoling,
lambasting, etc... can do for a development staff: it forces them to group their
code in logical structures, making it MUCH easier to debug, etc...

what would you rather debug:

for (i=1; i < 100; i++){
do something hideous here;
and then something else here;}

or

for i in range(100):
    do something pleasant here
    and then something else

Personally, there is nothing I hate more in the curly-brace languages than the fact
that developers don't have to line up their braces to show groupings of code. I
guess if every C/C++/Java, etc... developer wrote code like:

for (i=0; i < 100; i++)
{
    do something;
} /* notice how the braces line up and the body is indented!*/

I would be happy with that approach. However, there is nothing better than forcing
clean code... in my very humble opinion.



Thomas Hamelryck wrote:

> [Thomas Hamelryck  <thamelry at vub.ac.be>]
> : But many languages have evolved exactly this way: users criticize,
> : complain, praize and whine, thereby influencing the shape of the
> : languge in question.
>
> [Doug Landauer <landauer at apple.com>]
> : Yes, this works well in many cases, including Python.  Notice that
> : it is, for the most part, *non*-users of python who complain about
> : the indentation.  _Users_, for the most part, like it.  Languages
> : that change by catering to the tastes of non-users tend not to do
> : so well.
>
> I agree that most users of python seem to like the strict indentation rule.
> I seem to be one of the few actual python users who do not like it. The
> complaints about the indentation do not really come from *non*-users of python.
> Rather, they come from people who have gotten interested in python for some
> reason or another and are completely turned down by the idiosyncratic use of
> syntactically significant whitespace and/or tabs. So python is missing a lot
> of potential momentum beacuse of the indentation stuff. I find this a pity.
>
> Still a great language though.
>
> Cheers,
>
> ---
> Thomas Hamelryck    Institute of Molecular and Structural Biology
> Aarhus University   Gustav Wieds Vej 10C
> DK-8000 Aarhus C    Denmark



From fdrake at acm.org  Mon Jan 10 14:41:46 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 10 Jan 2000 14:41:46 -0500 (EST)
Subject: Python plugin
In-Reply-To: <3877E637.D3C9CC11@callware.com>
References: <38771491.BA08010@iqsoft.hu>
	<3877E637.D3C9CC11@callware.com>
Message-ID: <14458.13818.952396.355846@weyr.cnri.reston.va.us>

Nemeth Miklos wrote:
 > My conclusion after this discussion thread:
 > (1) Grail was not mentioned as an alternative. I am a bit sad that this
 > project was a failure.

Ivan Van Laningham writes:
 > Yah, me too.  But it would take an awful lot of funding to build a real
 > web browser in Python.  Maybe someday.

  Whether or not this project was a failure is something that could
easily be debated.  "The boss" (meaning Guido's boss, not Guido) has
pointed out that the primary intent of Grail was that it be a useful
testbed for new protocols; for this, it's not bad.  There just aren't
a large number of those that really seem to make sense for a
browser-like tool.
  If you really want a browser-like thing in Python, pick your widget
set and get to work!  There are certainly a lot of people who'd like
to see such a thing, whether it be simply an "HTML widget" or a
complete embeddable browser.
  (But it would be nice if Grail were more capable and robust;
patches, anyone? ;)


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From mrecks at my-deja.com  Sat Jan 29 10:46:22 2000
From: mrecks at my-deja.com (mrecks at my-deja.com)
Date: Sat, 29 Jan 2000 15:46:22 GMT
Subject: strange CGI problem
References: <85ga10$p8u$1@nnrp1.deja.com> <14459.47851.670815.455938@beluga.mojam.com> <85hb86$jjq$1@web1.cup.hp.com>
Message-ID: <86v20e$4nr$1@nnrp1.deja.com>

In article <85hb86$jjq$1 at web1.cup.hp.com>,
  "Arad Rostampour" <arad at fc.hp.com> wrote:
> Do you have a:
>
> print "Content-type: text/html\n"
>
> as your first output?
>
> Enjoy,
> Arad
> --
> Arad Rostampour- arad at fc.hp.com
> NT Platform and I/O Team
> Software and System Development Lab
> Hewlett-Packard
>
> Skip Montanaro wrote in message
> <14459.47851.670815.455938 at beluga.mojam.com>...
> >
> >    >> I'm creating a little web-site; for some reason when I view
the page
> >    >> in Internet Explorer, it's fine, but Netscape shows me the
raw html
> >    >> as if it's a text file.  What would cause this?
> >
> >Sounds like IE is looking at the contents and saying, "hmmm... looks
like
> >HTML", and displaying it.  Make sure your CGI script is spitting out
an
> >appropriate Content-type header.
> >
> >
>
>
Being new to CGI and Perl we have experienced a similar problem (with
Netscape but not with IE) and it appears you can get round it in at
least a couple of ways.

1) The easy way
This problem can be caused by having a non-standard extension to your
CGI program name, eg 'myprogram.cgi'.  Try changing the extension (the
bit after the dot) to a 'standard' one eg 'myprogram.html' Don't forget
to alter the relevant entry in the page that calls the CGI program.

2) A slightly more difficult method.
With a non-standard extension to your CGI program name the (Netscape)
browser 'sees' your cgi program  and attempts to use an external
application, as defined by the extension to read it, if you have not
defined a 'cgi' application it may default to displaying the source (or
plain text)
    Get around this problem by defining an application to cater for
pages/programs with 'cgi' (or whatever you use) extensions and setting
them to use the browser to interpret them instead of an external app.


Hope this helps.

X






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


From malcolmt at smart.net.au  Tue Jan 25 19:34:45 2000
From: malcolmt at smart.net.au (Malcolm Tredinnick)
Date: Wed, 26 Jan 2000 11:34:45 +1100
Subject: vi and python?
In-Reply-To: <86l7iq024mg@news2.newsguy.com>; from Ben Beuchler on Tue, Jan 25, 2000 at 04:22:14PM -0600
References: <85uc1m$5i8$1@pegasus.tiscalinet.it> <86l7iq024mg@news2.newsguy.com>
Message-ID: <20000126113445.A856@Ridcully.home>

On Tue, Jan 25, 2000 at 04:22:14PM -0600, Ben Beuchler wrote:
> There seems to be a general preference for emacs in the python community.  I
> am personally much more comfortable working with vim.  Does anyone have any
> particular tips for python development within vim? Particular shortcuts to
> test code, autoindentation settings, etc...  I would particularly like to be
> able to use some of the idle features with, perhaps, gvim...

Leaving aside the "general preference for emacs" bit (they are possibly just a
bit more vocal than the others :-) ), there was a thread about vim settings
back in late November. The article that best summarised everything is at
http://www.deja.com/getdoc.xp?AN=555098950

Note that this only deals with indenting issues. There was another thread
about jumping over blocks of code in Python (some tricky issues handling
continuation lines), but when I went to look for it just now, Deja slowed down
to a crawl. Search for recent articles with 'vim' in the subject line -- this
thread will be one of the few (about 20 hits) that are returned.

Just in case Deja does not love you either, the short version of the above
thread was that people liked something of this order:
	set expandtabs
	set tabstop=8
	set shiftwidth=4
	set softtabstop=4
	set smarttab
	set autoindent

or, on a single line:
	se et ts=8 sw=4 sts=4 sta ai

Quinn Dunkan also auggested the mapping:
	im :<cr>    :<cr><tab>			(four spaces in the middle)

for automatically indenting after a colon at the end of a line.

Cheers,
Malcolm Tredinnick

--
42.7 percent of all statistics are made up on the spot.




From bjorn at roguewave.com  Mon Jan 10 11:12:30 2000
From: bjorn at roguewave.com (bjorn)
Date: Mon, 10 Jan 2000 09:12:30 -0700
Subject: class destructor
References: <Pine.LNX.4.21.0001101459290.24821-100000@fep132.fep.ru>
Message-ID: <387A04ED.CFE1C1C7@roguewave.com>


Oleg Broytmann wrote:

> On Mon, 10 Jan 2000, [iso-8859-2] Przemys?aw G. [iso-8859-2] Gawro?ski wrote:
> > Is there 'automatic' destructor in python, like AClass::~AClass() in C++
> > ?
>
>    Method __del__(self) will be called when python is about to destroy the
> object (ie, reference count dropped to zero).

Although it doesn't get called for a number of useful cases, e.g. when part of a
circular datastructure, or when an exception is thrown (because the traceback
object has a reference?).  The only safe way I've found to implement
destructor-like behavior is:

class Foo:
    def __init__(self):
        ...
        self.open = 1

    def close(self):
        if self.open:
            # whatever is neccessary to 'delete' the object.

    def __del__(self):
        self.close()

then in the code where you want destructor-like behavior:

    foo = None
    try:
        foo = Foo() # exception can be called during init
        ...
    finally:
        if foo is not None:
            foo.close()

-- bjorn





From aahz at netcom.com  Sun Jan 23 16:11:08 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jan 2000 21:11:08 GMT
Subject: Simple socket loop
References: <ln5gqz5p.fsf@jumbo.skott.dk>
Message-ID: <86fqpc$5q0$1@nntp4.atl.mindspring.net>

In article <ln5gqz5p.fsf at jumbo.skott.dk>, Martin Sk?tt <mskott at image.dk> wrote:
>
>After having communicated with the client the server stop! How can I
>prevent that and keep it listening (and answering)? I guess I have to
>put it inside some kidn of loop but how. Please if you could give me a
>working example (or a link to one) it would be great because i'm still
>new to Python.

Instead of trying to recreate the wheel yourself, take a look at the
asyncore library.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From gerrit.holl at pobox.com  Wed Jan 19 16:34:05 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 19 Jan 2000 22:34:05 +0100
Subject: Python Rocks!
In-Reply-To: <8631bt$1nu@news.or.intel.com>; from tye4@yahoo.com on Tue, Jan 18, 2000 at 04:41:06PM -0800
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net> <8631bt$1nu@news.or.intel.com>
Message-ID: <20000119223405.B4536@stopcontact.palga.uucp>

tye4 wrote on 948210066:
[indentation blocks]
> So you don't think this is a problem?

No. It's a feature.
Consider:
if a: print 'a'
if b: print 'b'

with end delimeters, it would be:

if a: print 'a'
end if
if b: print 'b'
end if

Horrible nightmare!

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From wmcclain at salamander.com  Fri Jan  7 08:10:24 2000
From: wmcclain at salamander.com (Bill McClain)
Date: Fri, 7 Jan 2000 07:10:24 -0600
Subject: Attn Stuart Zakon (was: NEWS + threads)
References: <0t387365ddi452en79%milek@korweta.task.gda.pl> <1e3xrxz.5xo0stryd4xeN@[192.168.0.2]>
Message-ID: <1e40wqs.1ef3iuygbcftsN@[192.168.0.2]>

Bill McClain <wmcclain at salamander.com> wrote:

> I've been working on this recently and have done reference-header
> threading. Contact me via email for details. It's easy once you know
> how...

Stuart: your email address (zakons at objectsbydesign.com) is bouncing.

-Bill 


From md at i-i-s.co.uk  Mon Jan 24 04:42:10 2000
From: md at i-i-s.co.uk (Matt Dunn)
Date: Mon, 24 Jan 2000 09:42:10 -0000
Subject: python cgi, hosting
In-Reply-To: <38848C41.F358E7D5@netscapeonline.co.uk>
Message-ID: <0933F9C2D055D111BE93400001791A050FFE41@iisweb1.i-i-s.co.uk>

No, but check out http://starship.python.net/crew/lemburg/mxCGIPython.html
for a cgiPython module that you can put in your own webspace (if you know
the platform).

Matt Dunn

-----Original Message-----
From: python-list-admin at python.org
[mailto:python-list-admin at python.org]On Behalf Of nickluckcuck
Sent: 18 January 2000 15:53
To: python-list at python.org
Subject: python cgi, hosting


Does anyone know of one of the free web hosting who offers more than
perl cgi execution??

Thanks, Nick.





From mmiller3 at iupui.edu  Fri Jan 14 09:56:15 2000
From: mmiller3 at iupui.edu (M.Miller)
Date: 14 Jan 2000 09:56:15 -0500
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com> <002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com> <jtu2kh52y6.fsf@wazor.biostat.wisc.edu> <009501bf5e69$f908e530$f29b12c2@secret.pythonware.com> <200001141348.HAA09968@wazor.biostat.wisc.edu>
Message-ID: <873ds03do0.fsf@lumen.med.iupui.edu>

>>>>> "William" == William Annis <annis at biostat.wisc.edu> writes:

    >         Well, that's certainly true.  I suppose I should
    > have mentioned in passing that I've had trouble with some
    > operating systems trying to parse this:

    > #!/usr/bin/env python --  # -*- python -*-

    > Solaris has no problem with this.  Linux bleats bitterly
    > about it, which was something of a nightmare when we
    > started using Linux last year.  I assumed Daryl wanted to
    > write executable scripts and forgot to mention the line
    > above is not exactly portable, and thus my preference for
    > the other trick.

This is much more portable:

#!/usr/bin/env python
# -*- python -*-


Mike


From linli at computer.org  Wed Jan 26 14:39:07 2000
From: linli at computer.org (lin li)
Date: 26 Jan 2000 19:39:07 GMT
Subject: Any graph editors?
Message-ID: <86nigr$724$1@pollux.ip-plus.net>

Hi there,

I am doing a prototype where I need a customized graph editor. 
Eventually the editor is going to end up in Java. For the prototype,
however, I want to use something easier to play with, like Python.
As a lazy person, I'd like to start from somewhere more than
scratch. So here is the question: is there any implementation of
some graph editors in Python, JPython (or maybe Java) that I can
copy and modify?

Please answer to the news group. My ISP is having serious 
problems with the mail server now.

Thank you very much.

I-am-thinking-of-something-cute-for-this-place-ly y'rs Lin


From nielsenjf at my-deja.com  Tue Jan 18 19:01:38 2000
From: nielsenjf at my-deja.com (John Nielsen)
Date: Wed, 19 Jan 2000 00:01:38 GMT
Subject: python and ASP
References: <002501bf60ae$1773bc20$b674fea9@desktop>
Message-ID: <862uso$19a$1@nnrp1.deja.com>

Try something like:

Session.SetValue('userID', userid)

john



In article <002501bf60ae$1773bc20$b674fea9 at desktop>,
  "Sameer Chowdhury" <sameerc1 at yahoo.com> wrote:
> How do Sessions work when using python with ASP.   I, for example,
have
> working Python based ASP code, I add the following line:
Session("userID") =
> userID
> after defining userID.  After I add this line, I get the following
error
> when I test it.
> Python ActiveX Scripting Engine error '80020009'
>
> Unknown
>
> ?
>
> Unknown
>
> Can anyone suggest a remedy.  Thanks in advance.
>
> __________________________________________
> NetZero - Defenders of the Free World
> Get your FREE Internet Access and Email at
> http://www.netzero.net/download/index.html
>
>

--
nielsenjf at my-Deja.com


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


From DOUGS at oceanic.com  Sun Jan 30 13:48:02 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Sun, 30 Jan 2000 08:48:02 -1000
Subject: Choosing a language
Message-ID: <5650A1190E4FD111BC7E0000F8034D26A0F2A3@huina.oceanic.com>

Python is an excellent first language (or second or fourteenth for that
matter). I found putting together some cgi programs for the first time easy
enough.  I might have found doing cgi programming as my first Python
programs very frustrating though.

I'd suggest if you aren't familiar with the programming process start by
working through as many tutorials as you can. Follow the links on the Python
home page <http://www.python.org/>, --> 'Introductions to Python'
<http://www.python.org/doc/Intros.html> 'Introductions to Python programming
for non-programmers' where there are three currently listed.  I consistently
hear good things about Alan Gaulds tutorial for the complete beginner.

I find the greatest strength of Python to be the interactive interpreter.
In essence it is a sandbox to play in to try things out.  There is no other
way to learn to program than to write programs, but writing them can be
frustrating if you can't get into the middle of the execution and find out
what those commands you wrote are really doing.  The interpreter is the tool
for that.  Almost without fail if I can't make a program work I'll try a
sequence of commands in the interpreter that show me the problem.  While
you're learning to program, this instant feedback can make all the
difference in speeding you to comprehension.

Trying to jump right into programming cgi puts a couple of layers between
you and the results.  I think the only effective way to do it as a beginner
would be to have the interpreter available to try out and verify your
programming ideas before you try to call them using a web browser.

When you do get started learning, another great resource is the Python Tutor
mailing list.  Check out <http://www.python.org/psa/MailingLists.html#tutor>
to subscribe.  There you can get good direction on the how to program
questions as well as the Python questions.

I'd suggest you spend two weeks to a month absorbing all you can before you
start trying to do the cgi stuff.  When you do, a good resource is
<http://starbase.neosoft.com/~claird/comp.lang.python/web_python.html> which
among other things contains a link to 'Writing CGI Programs in Python'
<http://www.devshed.com/Server_Side/Python/CGI/> which answers most of your
'since I want to focus on CGI, what would be the best resources to check
out' question.

Good luck,
-Doug-

> -----Original Message-----
> From: Patrick Cooper [mailto:future at dotstar.net]
> Sent: Sunday, January 30, 2000 7:20 AM
> To: python-list at python.org
> Subject: Choosing a language
> 
> 
> I want to start learning computer programming and from what I 
> have found so
> far I have chosen to start with Python.  I am going to be 
> focusing on CGI
> first but eventually want to work with desktop applicates.  
> Python seems to
> be the most versitle and easiest to learn.  Before I go in 
> any more depth
> does anyone think I should start with a different language?  I want to
> eventually learn C, but think Python is the best place to start.
> 
> Also, since I want to focus on CGI, what would be the best 
> resources to
> check out?  Anything besides what's listed in  the CGI Tools 
> section of the
> Python web site?
> 
> 
> 
>   -----------== Posted via Newsfeeds.Com, Uncensored Usenet 
> News ==----------
>    http://www.newsfeeds.com       The Largest Usenet Servers 
> in the World!
> ------== Over 73,000 Newsgroups - Including  Dedicated  
> Binaries Servers ==-----
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From gmcm at hypernet.com  Fri Jan 21 08:49:05 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 21 Jan 2000 08:49:05 -0500
Subject: Talking to embedded python
In-Reply-To: <m3snzsdq7l.fsf@brain.acme.org>
Message-ID: <1263658528-14676811@hypernet.com>

Joe Cotellese asks:

> I'm not sure how to ask this question. I'm interested in embedding
> python into a Win32 based application. One reason for this is to allow
> external control of the application.
> 
> My question is how can I communicate with the python interpreter built
> into my application? The idea being that once the program is running I
> can send commands in the form of python scripts to the running
> program.
> 
> I know this is probably not an easy question to answer (or maybe it
> is!) but if anyone could point me in the right direction I would
> appreciate it.

The problem is that there are too many possible answers. 
Unlike extending (which you'll want to do too, so Python can 
reach back into your app), embedding can be done at any 
level. You can start the interpreter and pass it strings to 
evaluate. At the other extreme, you can write Python in C / 
C++. It all depends on what level of control / communication 
you need.

Start with the embedding / extending sections of the docs. 
Look at 
http://starship.python.net/crew/arcege/extwriting/pyext.html
If your app is MFC, look at Pythonwin (part of the Win32 
extensions).

Better yet, rewrite your app in Python and extend with C / 
C++ where it needs some speed. Either Pythonwin or 
wxWindows will give you a fully Win32 compliant GUI.




- Gordon



From evan at tokenexchange.com  Fri Jan 21 16:23:12 2000
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 21 Jan 2000 15:23:12 -0600
Subject: def args idiom and signatures
References: <02b001bf629c$d7ce3960$01ffffc0@worldnet.att.net> <38864391.97AB0AE6@digicool.com> <034401bf62f7$4fa7bba0$01ffffc0@worldnet.att.net>
Message-ID: <s8hjr778eh619@corp.supernews.com>

Emile van Sebille <emile at fenx.com> wrote in message
news:034401bf62f7$4fa7bba0$01ffffc0 at worldnet.att.net...
> Well, wouldn't this be exposing what probably ought to be a private
> function variable?  If a function signature shows the def's argument
> structure, how could a user tell that the intent is to *not* override
> the value, and that doing so would break the function?  I know in
> this case the value itself ("000031028...") pretty much says that, but
> I can easily imagine other values being pre-set in the def to avoid
> the overhead.

That's why in another thread we're pushing for syntax changes which would
allow (among other things)...

def _d(y, m, d, *, days=(0, 31, 59, ..., 334, 365)):

...to prevent 'days' from being part of the function signature while still
allowing it to be pre-defined this way.

Cheers,

Evan @ 4-am




From nemeth at iqsoft.hu  Thu Jan 20 15:37:54 2000
From: nemeth at iqsoft.hu (Nemeth Miklos)
Date: Thu, 20 Jan 2000 21:37:54 +0100
Subject: no multithreading?
Message-ID: <38877222.9F5BBBA9@iqsoft.hu>

>       From:
>            Erno Kuusela
>
>
>
>
> >>>>> "moonseeker" == moonseeker  writes:
>
>     moonseeker> Hi!  I have understood that the python interpreter
>     moonseeker> switches to another thread if a fixed amount of python
>     moonseeker> commands is completed. Is this true?
>
> after a fixed number of bytecode instructions have been executed, yes.
> you can set it with sys.setcheckinterval().
>
>     moonseeker> That means if a task hangs on a command (i.e. file
>     moonseeker> open), all tasks are locked? What happens if I use
>     moonseeker> sleep() in a task?
>
> no, afaik i/o operations and sleep() and such release the interpreter
> lock, enabling other threads to excute.
>
This can only be implemented using real threads. If a Python thread is
about to execute a long operation implemented in C (for example a
database call), this thread will be blocked untill the called procedure
returns.
What operating system threads are going to execute the other Python
threads while this proc is running?

By the way I bought Mark Lutz's Programming Python, and I am a bit
disappointed that  he does not even mention multi-threading in the
870-page book.
Is there another book that explains more advanced features of Python?

NM

>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000120/bc483deb/attachment.html>

From sjohnson at csnet.net  Wed Jan 26 21:31:13 2000
From: sjohnson at csnet.net (Scott Johnson)
Date: Thu, 27 Jan 2000 02:31:13 GMT
Subject: Newbie question
Message-ID: <R5Oj4.892$yY6.13414@news.goodnet.com>

My background is in MVS Cobol, CICS & Visual Basic.
I have played a little with java & Perl.
I have an understanding of classes & their properties & methods.
Could someone explain to me the concept of .self ?



From morse at harborcom.net  Wed Jan 26 08:34:24 2000
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Wed, 26 Jan 2000 13:34:24 GMT
Subject: vi and python?
References: <85uc1m$5i8$1@pegasus.tiscalinet.it> <86l7iq024mg@news2.newsguy.com>
Message-ID: <388ef786.90332811@news.oh.verio.com>

Ben Beuchler <insyte**NOSPAM**@emt-p.org> wrote:

>There seems to be a general preference for emacs in the python community.  
>I am personally much more comfortable working with vim.  
>Does anyone have any particular tips for python development within vim?

It looks like Geocities has some issues write now, but when it comes back up
you can find some scripts for the embedded Python interpreter in VIM at:

http://www.geocities.com/SiliconValley/Peaks/4673/python.html




From skip at mojam.com  Thu Jan  6 11:42:32 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 6 Jan 2000 10:42:32 -0600 (CST)
Subject: Simple cartesian product
In-Reply-To: <iac9023qi3y.fsf@vier.idi.ntnu.no>
References: <000801bf571c$0cf448a0$1a2d153f@tim>
	<iacembwr7qg.fsf@vier.idi.ntnu.no>
	<Pine.LNX.4.10.10001061110050.6929-100000@gaia.gfy.ku.dk>
	<20000106132958.E7015@xs4all.nl>
	<iac9023qi3y.fsf@vier.idi.ntnu.no>
Message-ID: <14452.50680.706438.39124@beluga.mojam.com>

    Magnus> Thomas Wouters <thomas at xs4all.net> writes:
    >> So is this YYYY-MM-DD or YYYY-DD-MM ?
    Magnus> yyyy-mm-dd

    Magnus> Logical, considering other formats like hh:mm:ss, and decimal
    Magnus> numbers in general.

Yes, even more logical considering the fact that sorting a file using
ISO-8601 dates will actually sort the file into proper ascending or
descending order.

Skip




From jasonic at nomadicsltd.com  Sun Jan  2 22:09:34 2000
From: jasonic at nomadicsltd.com (Jason Cunliffe)
Date: 3 Jan 2000 03:09:34 GMT
Subject: Python Books & improving PyAwareness - How to ?
Message-ID: <01bf5597$3dfee1c0$ddc854a6@brancusi.pro-net.co.uk>

Hello

I am very happy to see the new Python books appearing. Some good titles and
all the new ones I've picked up are excellent [Quick Python, The Essential
Python Reference and Python-Annotated archives]...my sincere thanks to
their respective authors.

I imagine that most people shooping for Pyhton books will tend to follow
the links from www.python.org pages and purchase through amazon.com or
similar.

Meanwhile, I have been making an ad-hoc survey in the New York City
bookstores over the holiday. Sad to say there is a very low profile of
python books there. Only a couple of copies at best in most
stores...compared to shelf-fulls of the other P and J lanaguage books. Even
worse [but perhaps significant] is that in most stores the Python books
were extremely oddly placed - sometimes a volume or two in the multimedia
section, sometimes a tome in Internet, somtimes a copy wher you would
expect it a thte end of the Perl shelf. In no store did I find all the
current published Python books together. 

I noted that ther was no Python books in Linux sections - which I think is
rather surprisin and dissappointing, especially given the enthusiasm now
for Linux sections in these stores.

I doubt that anyone here can directly affect much of this - but I still
think it is worth some consideration:

Q1: How to improve the bookstore/publisher handling of Python book
placement?

Q2: Does it matter? 

Either people wil be discouraged about Python because it appears in this
context to be so marginal and insignificant OR they will simply conlcude
that on-line book shopping is the only way to go.. Non-progammers, people
hiring technicians may be less likely to take Python as seriuosly if they
go to the local store to see what's cooking..

Presumably bookstore buyers respond to 4 main forces: 
1.  Distributors
2.  Publishers
3.  Direct customer request 
4.  Conscientious motivated sales staff

I would guess that 1+2 are the major players, but 3 and 4 can often act as
valuable influences. I always makea point of asking in person to see see
what awareness there is.  

Q3: Do you think a coodinated ad-hoc survey feedback from member so this
list woudl have any effect on the publishers of Python books?

any ideas..?

best wishes for Py2k
- Jason



From hjwidmai at foxboro.com  Wed Jan 19 00:45:23 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Wed, 19 Jan 2000 06:45:23 +0100
Subject: Tkinter hangs again under Windows
In-Reply-To: <388481A2.D233D165@endea.demon.nl>
Message-ID: <20000119064523.A15259@foxboro.com>

Niels Diepeveen wrote:

> This seems to be the problem. If you show any dialogs before entering
> mainloop, you won't receive events anymore.

Apparently this is the case. Is this a (known?) bug or just an undocumented
feature, or is it even documented? The problem seems to only exist under
Windows.

Thanks a lot. I'll find a way around it, probably a self.after() that
shows the dialogue shortly after entering the mainloop.

-- 
Hans-Joachim Widmaier         E-Mail:  hjwidmai at foxboro.com
Foxboro Eckardt GmbH          Telefon: +49 711 502 - 397
Pragstrasse 82                Fax:     +49 711 502 - 300
D-70376 Stuttgart 



From gerrit.holl at pobox.com  Mon Jan 31 11:12:55 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 31 Jan 2000 17:12:55 +0100
Subject: tabs do WHAT?
In-Reply-To: <Teil4.12357$t_.270545@news.rdc1.nj.home.com>; from BgPorter@NOacmSPAM.org on Mon, Jan 31, 2000 at 03:54:27PM +0000
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <874sby5m6l.fsf@iname.com> <20000131153329.B19725@xs4all.nl> <Teil4.12357$t_.270545@news.rdc1.nj.home.com>
Message-ID: <20000131171255.A1124@stopcontact.palga.uucp>

Brett g Porter wrote on 949330467:
> 
> "Thomas Wouters" <thomas at xs4all.net> wrote in message
> news:20000131153329.B19725 at xs4all.nl...
> > """
> > Tabs are 8 characters, and thus indentations are also 8 characters.
> > There are heretic movements that try to make indentations 4 (or even 2!)
> > characters deep, and that is akin to trying to define the value of PI to
> > be 3.
> >
> > Rationale: The whole idea behind indentation is to clearly define where
> > a block of control starts and ends.  Especially when you've been looking
> > at your screen for 20 straight hours, you'll find it a lot easier to see
> > how the indentation works if you have large indentations.
> >
> ...except that studies of actual human programmers say otherwise. From Steve
> McConnell's "Code Complete" (pp 409-410):
> "The same study ['Program Indention and Comprehensibility', Miaria, et al,
> 1983] found...[t]he lowest comprehension scores were achieved on programs
> that were not indented at all. The second lowest were achieved on programs
> that used six-space indentation. The study concluded that two to four space
> indentation was optimal. Interestingly, many subjects in the experiment felt
> that the six-space indentation was easier to use than the smaller
> indentations, even though their scores were lower."
> 
> The original study was printed in the Nov 1983 Communications of the ACM.

Tell Linus!

-- 
Please correct any bad Swahili you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From jepler at inetnebr.com  Wed Jan 26 21:34:58 2000
From: jepler at inetnebr.com (Jeff Epler)
Date: Thu, 27 Jan 2000 02:34:58 GMT
Subject: urllib.py patch: proxy authentication
Message-ID: <slrn88vbmb.pl8.jepler@craie.inetnebr.com>

The following patch adds support for proxies like
        http_proxy=http://USER:PASS at host:port/
which would formerly confuse the urllib into giving a traceback about a
non-numeric port number. (the PASS at host:port portion)

This patch is probably not ready for inclusion as-is, because the CVS
version of python has a modified urllib that is too far distant for this
patch to apply.  (This is relative to 1.5.2)  If I get around to to
a patch that will apply to the CVS tree as well, I will submit this
patch again.

Other than that, I believe I have tried to live up to the standards asked
for in Python patches.

I confirm that, to the best of my knowledge and belief, this
contribution is free of any claims of third parties under
copyright, patent or other rights or interests ("claims").  To
the extent that I have any such claims, I hereby grant to CNRI a
nonexclusive, irrevocable, royalty-free, worldwide license to
reproduce, distribute, perform and/or display publicly, prepare
derivative versions, and otherwise use this contribution as part
of the Python software and its related documentation, or any
derivative versions thereof, at no cost to CNRI or its licensed
users, and to authorize others to do so.

I acknowledge that CNRI may, at its sole discretion, decide
whether or not to incorporate this contribution in the Python
software and its related documentation.  I further grant CNRI
permission to use my name and other identifying information
provided to CNRI by me for use in connection with the Python
software and its related documentation.


--- urllib.py.orig	Wed Jan 26 20:22:36 2000
+++ urllib.py	Wed Jan 26 20:25:06 2000
@@ -230,11 +230,13 @@
                 user_passwd, host = splituser(host)
                 host = unquote(host)
             realhost = host
+            proxy_passwd = None
         else:
             host, selector = url
             urltype, rest = splittype(selector)
             url = rest
             user_passwd = None
+            proxy_passwd, host = splituser(host)
             if string.lower(urltype) != 'http':
                 realhost = None
             else:
@@ -245,6 +247,11 @@
                     selector = "%s://%s%s" % (urltype, realhost, rest)
             #print "proxy via http:", host, selector
         if not host: raise IOError, ('http error', 'no host given')
+        if proxy_passwd:
+            import base64
+            pauth = string.strip(base64.encodestring(proxy_passwd))
+        else:
+            pauth = None
         if user_passwd:
             import base64
             auth = string.strip(base64.encodestring(user_passwd))
@@ -258,6 +265,7 @@
         else:
             h.putrequest('GET', selector)
         if auth: h.putheader('Authorization', 'Basic %s' % auth)
+        if auth: h.putheader('Proxy-Authorization', 'Basic %s' % pauth)
         if realhost: h.putheader('Host', realhost)
         for args in self.addheaders: apply(h.putheader, args)
         h.endheaders()


From david.maslen at iname.com  Thu Jan 27 20:00:05 2000
From: david.maslen at iname.com (David Maslen)
Date: Fri, 28 Jan 2000 01:00:05 GMT
Subject: Stackless on the Palm Pilot Re: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com> <slrn88pg1h.scl.wtanksle@hawking.armored.net> <m3r9f3veja.fsf@chinon.cnrs-orleans.fr> <UIYj4.75$q64.2465@weber.videotron.net> <389059A0.B37CEBBF@tismer.com> <38908832.48062DA1@pacificnet.net>
Message-ID: <87snzj6n8d.fsf_-_@iname.com>

"Jeffery D. Collins" <jcollins at pacificnet.net> writes:
> The release of python-1.5.1 for the PalmIII (or greater) will
> be announced within the next couple of weeks.  I need to clean
> up the distribution and documentation first. 

I'd be interested to know what sort of expectations I should have for
this. In particular speed. As I'm sure all palm owners are aware,
speed is critical in a 'good' application.

I recently installed KVM, the java virtual machine for the palm. It's
impressive, but the delay in starting the virtual machine and then the
actually program running a little slower that similar C-Coded
applications is disappointing.

Disappointing is probably the wrong word. It's what I expect from java
and the virtual machine, but is this what I should also expect from
stackless python on the pilot?


From mfletch at tpresence.com  Fri Jan 21 04:54:34 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Fri, 21 Jan 2000 04:54:34 -0500
Subject: Python Rocks!
Message-ID: <CC59868A0A76D311B1F50090277B321C1F3089@VRTMAIL>

Curse your Xenophobia!  What's wrong with embracing a little alien
technology once in a while?  Hath not an alien grammars, productions, rules?
If you prick us, do we not ooze a clear ectoplasm?

Seriously, I would guess wrapping Antlr might be a better approach?  That
would (hopefully) allow for compatible modules on both the C and Java
implementations.  From what I've heard, the Antlr codebase is well designed
and fairly fast.

And, when you've gained the wisdom to venture out among the stars and
appreciate the differences in the universe, my people will be waiting to
share our technology with you.

Nanu,
Mike

-----Original Message-----
From: Tim Peters [mailto:tim_one at email.msn.com]
Sent: Friday, January 21, 2000 4:38 AM
To: Fran?ois Pinard
Cc: python-list at python.org
Subject: RE: Python Rocks!
...
[Fran?ois Pinard]
> ...
> Isn't there a need for compiling grammars into fast scanners
> and parsers, a bit like Flex and Bison do for C people?

Yes, but that's really a different topic.  The Python world has no good
approach to that now, paying attention to the "fast" part, and where "good"
means "enough like Flex and Bison that you don't feel you've been stranded
on some strange alien planet" <wink>.
...



From thomas at xs4all.net  Wed Jan 12 08:51:07 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Wed, 12 Jan 2000 14:51:07 +0100
Subject: a small problem
In-Reply-To: <000701bf5cb4$c72320a0$da2d153f@tim>; from tim_one@email.msn.com on Tue, Jan 11, 2000 at 11:23:29PM -0500
References: <20000112015923.A8450@xs4all.nl> <000701bf5cb4$c72320a0$da2d153f@tim>
Message-ID: <20000112145107.B8450@xs4all.nl>

On Tue, Jan 11, 2000 at 11:23:29PM -0500, Tim Peters wrote:

> Whereas people like Guido <wink> would not have considered writing anything
> other than
> 
> >>> sum = 0
> >>> for n, s in q:
>         sum = sum + n

Aahh, yes. I dont know what I was thinking when I wrote it using map and
reduce instead of a much more obvious for loop. My only excuse is that it
was late yesterday night, and I'd just come back from an Amsterdam Perl
Mongers meeting. Amusing evening, but it probably confused me into using the
least obvious of all the ways to do it ;)

> Seriously, I don't think I've ever seen a good use for reduce in Python --
> although I've certainly used it <wink>.

People still care about speed. I know I sometimes do, even though I know
intellectually that it isn't going to matter much in the end product. I
still like to make it as speedy as possible, even if that entails using
map/reduce or odd looking constructions like 'def spam(text, None=None)'. I
suspect the use of map/reduce will drop when python is optimisable enough
to make the for loop _feel_ faster than map(). Static typing and what not.

On the other hand, oneliners will always feel or look faster to some people.
I'm one of them, I must admit. Another thing is the geekiness and
playfulness of oneliners. 'look, mom, only one line.' Maybe this is where
list comprehensions come in ;)

[ <snip> Guido saying lambda's suck (compared to the rest of python) ]

> I suppose finally posting that will rekindle a few old flamewars!  Brrrr.

Weeellll, I never bothered much with flamewars (though I did see a spot of
red when Tom Christiansen made his appearance last year) but I was always
sceptical when I heard someone say such things. Surely lambda's are good in
some cases ! But lately I've been looking at a lot of things differently and
considering wether the good it brings in some cases justifies the pain it
brings in other cases (this goes for much more than just lambda, python,
computers or corporeality, though.)

So I hope potential flamers are willing to take a step back and do what I
originally did. Start with thinking, "Guido Is Right", then consider, "Why
is Guido right ? What is he saying, exactly ?" and ending up in argument
with yourself. If you do end up thinking lambda's are guido's gift to sane
coding, at least you'll be prepared for the flamewar, having heard all
arguments.

It's important, though, once started, to finish this process. Dont get stuck
at 'Guido Is Right', even if it's a very comfortable situation and nobody is
going to notice in the end. The bureau of AFT might pull a Waco on Guido,
after all :P

> can't-believe-he'd-drink-coffee-with-a-straw-ly y'rs  - tim

it's-an-old-dutch-delicacy---irish-coffee-ly y'rs.
-- 
Thomas Wouters <thomas at xs4all.net>

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



From skip at mojam.com  Wed Jan 12 14:51:47 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 12 Jan 2000 13:51:47 -0600 (CST)
Subject: Ping in Python
In-Reply-To: <5650A1190E4FD111BC7E0000F8034D26A0F250@huina.oceanic.com>
References: <5650A1190E4FD111BC7E0000F8034D26A0F250@huina.oceanic.com>
Message-ID: <14460.56147.2037.654537@beluga.mojam.com>

    Doug> I think on Windows you need to download the source to get it, but
    Doug> it was on my RedHat system under
    Doug> /usr/doc/python-demos-1.5.2/Demo/sockets/udpecho.py.

If you don't have sources and don't want to download the whole
kit-n-kiboodle, I maintain a web-available source tree here:

    http://www.musi-cal.com/~skip/python/Python/

I just sync'd it with the CVS repository, so it's up-to-date.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098




From slinkp23 at yahoo.com  Fri Jan 21 01:12:16 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Fri, 21 Jan 2000 01:12:16 -0500
Subject: Realtime scheduling??
References: <38858064.CDA4866C@yahoo.com> <87ogahtvnf.fsf@fun77.kivikko.hoas.net>
Message-ID: <3887F8C0.C2CDB7B0@yahoo.com>

Erno Kuusela wrote:
> 
> >>>>> "paul" == Paul Winkler <slinkp23 at yahoo.com> writes:
> 
>     paul> Hi, I'm working on a module I eventually hope to use to
>     paul> generate musical events in realtime. (These events will be
>     paul> sent to MIDI devices or to software synthesizers.)
> 
> i think there is some midi support in the stock sound drivers.
> you might be better off using that.

I _am_ using stock sound drivers (OSS/ Linux) ... that's how I get
messages to MIDI devices.
Maybe you misunderstood. I'm not writing a software synthesizer. I'm
using other people's synth software, principally csound. Csound can
read realtime events sent as text through a pipe, it's fun and easy
to do. My code simply generates events for controlling other
synthesizers, whether hardware synths via MIDI or soft-synths via
pipes or whatever.
 
>     paul> I've found that on my Linux system (pretty much stock RedHat
>     paul> 5.2), scheduling actions with sched.scheduler is not as
>     paul> reliable as I'd like. Events are anywhere from 2 to 20
>     paul> milliseconds late.
> 
> yeah, linux is not a real-time operating system.

BUT, for all intents and purposes, it CAN be with a patched 2.2
kernel.  Some people have demonstrated that linux can reliably
_beat_ the realtime performance of BeOS which is supposed to be the
best commercial OS for realtime multimedia work.
But you need 2.2 and some patches to it, and you need your C program
to be run SCHED_FIFO, and I don't know quite what that means or how
to make python do it, or even if it can.

> using a "real-time"
> scheduling class might, or might not, help. they are mostly a way to
> tell the scheduler to choose your task over other tasks when it's time
> to switch.

Do you have any examples of doing this with python? I'm searching
the python website for "real-time" and find nothing but rtlinux
stuff...
thanks!
 
>     paul> I'd like to improve the reliability, which brings me to my
>     paul> next question:
> 
>     paul> Is there any way to run a python script, or rather the
>     paul> python interpreter itself, with SCHED_FIFO priority on a
>     paul> linux system?  Or any other way to guarantee that something
>     paul> will happen when you really want it to happen??
> 
> there are programs that work like nice(1), one is named rtnice. a web
> search should find it for you. then you can say
> "rtnice python myprogram.py".

That's intriguing... I found some links... looks like this also
needs kernel 2.2, so it's REALLY time for an upgrade here.
 
Thanks for the reply!

................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
A member of ARMS    ----->    http://www.reacharms.com
or http://www.mp3.com/arms or http://www.amp3.net/arms
personal page   ---->    http://www.ulster.net/~abigoo


From claird at starbase.neosoft.com  Wed Jan 19 18:33:22 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 19 Jan 2000 17:33:22 -0600
Subject: [Newbie] Python, What is It ?
References: <865clq$qkc$1@nnrp1.deja.com>
Message-ID: <67211DAB9460A644.C7D18C730330BEAE.1889BE0748B20F59@lp.airnews.net>

In article <865clq$qkc$1 at nnrp1.deja.com>,  <kultur_of_void at my-deja.com> wrote:
>Hello
>
>I'm new to software design, but have a great hunger for learning and
>working to design my software. ( A musical stuff based on chaos )
>
>Searching for GUI design tools, I've been indicated (beside Fltk, MFC
>and ATL) Python/Tkinter, but also Perl/Tk and Tcl/Tk.
>
>I have knowledge in C, Win32 API, C++, MFC, DirectSound... I want a
>heavily graphic UI ( the front face of a synthesizer with knobs,
>sliders, leds, switchers etc...)
>
>Do you think Python is THE tool for my GUI design.
			.
			.
			.
Some will say "yes", some "no".  My own position is
that all the possibilities are sufficiently comparable
in technical merit that the true answer will depend on
how will each tool fits *you*.

That's the bad news.  It's bad news, because it dis-
courages you from deciding based on external authority.

The good news, though, is that with only about an hour
or two of your own time invested in these languages,
you can give each one a reasonable opportunity to make
a "first impression".  *Then* you can start to make a
reasoned judgment.

Other good news:  each language enjoys a wealth of
well-written books, illuminating Web sites, and helpful
volunteers to start you on your path.

In the meantime, I recommend you read
<URL:http://www.deja.com/=dnc/getdoc.xp?AN=571126529>
and the references you'll find there, to acquaint
yourself with some of the possibilities within your
reach.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From thomas at xs4all.net  Mon Jan 17 11:53:43 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 17 Jan 2000 17:53:43 +0100
Subject: Global help
In-Reply-To: <85vfbp$f3t@catapult.gatech.edu>; from gte838h@prism.gatech.edu on Mon, Jan 17, 2000 at 04:18:01PM +0000
References: <85vfbp$f3t@catapult.gatech.edu>
Message-ID: <20000117175343.E21652@xs4all.nl>

On Mon, Jan 17, 2000 at 04:18:01PM +0000, Shaun Allen Dishman wrote:

> I have something like the following situation:

> ----- x.py -----

> from y import *

> def x(filename):
> 	file = open(filename, 'w')
> 	y()

> ----- y.py -----
> 
> def y():
> 	# must be the same file object as in x.py
> 	file.write('blah blah blah \n')	
> 
> ----------------

> I have fiddled around with the global settings for the file object but
> cannot seem to get things to work.  All I need is to be able to access the
> file object from one module inside of another one, without passing it as a
> parameter.  Is this even possible?  Thanks in advance.

Well, if you _really_ need it as a global in y, you can do the following:


-------- x.py --------
import y

def x(filename):
	file = open(filename, 'w')
	y.y()
	
----------------------

-------- y.py --------

y.file = None

def y():
	y.file.write("spam spam spam\n")
----------------------

Or, if you must have the 'from y import *' statement:

-------- x.py --------
from y import *

def x(filename):
	setfile(open(filename, 'w'))
	y()
	
----------------------

-------- y.py --------

file = None

def setfile(newfile):
	global file
	file = newfile

def y():
	file.write("spam spam spam\n")
----------------------

But, in general, the from y import * statement is a Bad Thing. You can
overwrite your own functions/variables/classes, or the builtin ones, and not
notice it for weeks.

Think-from-perl-import-use-without-qw()-ly y'rs,

-- 
Thomas Wouters <thomas at xs4all.net>

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



From a.eyre at optichrome.com  Fri Jan 14 13:07:40 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Fri, 14 Jan 2000 18:07:40 -0000
Subject: class methods?
In-Reply-To: <85lf09$brl$1@nntp5.atl.mindspring.net>
Message-ID: <000101bf5eba$3f2af0d0$3acbd9c2@peridot.optichrome.com>

>> Company().print_stats()

> Huh.  That's interesting.  I wonder if there's any significant
> performance problem with this?

Problems arise if the constructor for the class throws an exception
under certain conditions.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From aahz at netcom.com  Sat Jan 22 19:30:51 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jan 2000 00:30:51 GMT
Subject: Python dumps core with this.. how come?
References: <000001bf651d$f05ac240$132d153f@tim>
Message-ID: <86di3r$tj0$1@nntp4.atl.mindspring.net>

In article <000001bf651d$f05ac240$132d153f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>Guido gave some excellent semi-exasperated advice (similar in spirit to
>parts of Michael's advice that I've snipped) many moons ago, when the whole
>newsgroup was temporarily filled with msgs of this character:  he invited
>people to just *try* writing a Python program without using any __xxx__
>hooks.  I expect some people were amazed at how much easier their lives got!
>If he were to get that exasperated again today, I bet he'd expand his
>invitation to include avoiding map, filter, reduce and lambda too.

I assume that Guido's advice specifically excluded __init__; otherwise,
it makes no sense to me.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From sholden at bellatlantic.net  Fri Jan  7 07:49:58 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Fri, 07 Jan 2000 12:49:58 GMT
Subject: Python not a Very High-Level Language?
References: <slrn86ur6a.fph.neelk@brick.cswv.com> <000801bf5802$dcca7d60$0aa0143f@tim> <853lss$tj8$1@nnrp1.deja.com>
Message-ID: <3875E10F.F24591A2@bellatlantic.net>

srenner at lycosmail.com wrote:
[...]
>                                                             As you
> are probably only dimly (if at all) aware, I've been using Python and
> PyOpenGL to test some realtime rendering ideas. At a certain point I
> began to grow increasingly frustrated my lack of control over the many
> layers involved (Python, PyOpenGL, Mesa library, X windows, Linux).

This is called "sitting on the shoulders of giants".  If your
alternative
is to produce all this stuff yourself, perhaps your situation might seem
less frustrating.  Could be worse: you might have chosen a Windows
platform,
or had no choice of platform at all ...

>     Again, Oz looks so cool. And Python is great. Just not C. Or
> anything that reminds me of C.
> 

I'm kind of on your side about C, which has always struck me as a
pragmatist's
language.  And you're probably right about the reason it's so much used
(though
criticizing Python for being written in it seems unfair: Guido should
have
chosen assembler, maybe?  Surely it's the implementation choices which
you
disagree with).

When I'm teaching class I occasionally take a sideswipe at "the evil
empire",
and someone replies "but you've got to use Microsoft stuff, everyone
does".
I wish I could remember where I got the reply I use, which is "In that
case
you should eat sh*t, since six hundred billion flies can't possibly be
wrong".

Good luck with your project!

regards
 Steve


From aahz at netcom.com  Fri Jan 28 19:37:11 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 29 Jan 2000 00:37:11 GMT
Subject: Call for Participation: Open Source Convention in Monterey (July)
References: <200001271623.LAA03648@eric.cnri.reston.va.us>
Message-ID: <86tcnn$7v6$1@nntp5.atl.mindspring.net>

In article <200001271623.LAA03648 at eric.cnri.reston.va.us>,
Guido van Rossum  <guido at CNRI.Reston.VA.US> wrote:
>
>I received the following request from O'Reilly.  The deadline is very
>tight (Feb 1); this is an exciting opportunity to teach other open
>sourcers about Python!

With the permission of my boss, I've submitted two proposals:

* Case study of Searchbutton.com and our mixed open-source/proprietary
environment

* Tutorial on Python threads
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From aa8vb at yahoo.com  Tue Jan 11 10:49:27 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 11 Jan 2000 10:49:27 -0500
Subject: LongInt-to-String % Bug (was Re: Python plugin)
In-Reply-To: <slrn87kk90.930.kc5tja@garnet.armored.net>
References: <38771491.BA08010@iqsoft.hu> <3877E637.D3C9CC11@callware.com> <slrn87kk90.930.kc5tja@garnet.armored.net>
Message-ID: <20000111104927.A40699@vislab.epa.gov>

Samuel A. Falvo II:
 |To be fair, Python isn't without its numerical faults either.  For example:
 |
 |>>> data = string.atol( stringArgument[0:8] )
 |>>> print "%08lX" % (data)
 |
 |will fault on any integral value greater than 0x7FFFFFFF.  Why?  Because
 |Python just plain can't understand the concept of an unsigned integer.

That's odd, especially considering str() and hex() can cut the mustard.

I don't see a bug report in the database for this; you might want to file one:

     http://www.python.org/search/search_bugs.html


    >>> str( 0x80000000L )
    '2147483648L'
    >>> "%ld" % 0x80000000L
    OverflowError: long int too long to convert

    >>> hex( 0x80000000L )
    '0x80000000L'
    >>> "%lX" % 0x80000000L
    OverflowError: long int too long to convert


LibCh2:
"The size specifiers h, l , or L may be present but are ignored"
No caveats about long ints though.

-- 
Randall Hopper
aa8vb at yahoo.com



From grant at nowhere.  Wed Jan 19 15:51:32 2000
From: grant at nowhere. (Grant Edwards)
Date: Wed, 19 Jan 2000 20:51:32 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com> <slrn88c42t.53m.grant@grante.comtrol.com>
Message-ID: <slrn88c8uk.5a8.grant@grante.comtrol.com>

In article <slrn88c42t.53m.grant at grante.comtrol.com>, Grant Edwards wrote:

A "message" as I find them in my inbox impliments the following
properties (shown with typical values):

               msg.Application: Collaboration Data Objects
                msg.Categories: None
                     msg.Class: 3
         msg.ConversationIndex: 01BF62A319835A310278CDB711D3B4D600609763ECBA
         msg.ConversationTopic: Testing 1,2,3
                msg.Importance: 1
                msg.MAPIOBJECT: <PyIUnknown at 0x7d4ff4 with obj at 0x16af420>
               msg.Sensitivity: 0
                      msg.Sent: 1
                   msg.Session: MS Exchange Settings
                      msg.Size: 390
                   msg.Subject: Testing 1,2,3
                 msg.Submitted: 0
                    msg.UnRead: 0

I'm assuming that stuff like the message body, sender's e-mail
address, etc. are all contained in the MAPIOBJECT object.

How do I find out what type of object that is?

-- 
Grant Edwards                   grante             Yow!  Today, THREE WINOS
                                  at               from DETROIT sold me a
                               visi.com            framed photo of TAB HUNTER
                                                   before his MAKEOVER!


From SpyderTL3 at yahoo.com  Thu Jan 13 17:00:13 2000
From: SpyderTL3 at yahoo.com (Allen Cleveland)
Date: Thu, 13 Jan 2000 22:00:13 GMT
Subject: Newbie
Message-ID: <NVrf4.1234$bQ1.17968@dfiatx1-snr1.gtei.net>

ok, i'm new to this and want to get into programming.  What do i need to
D/L?

etc..


-------------------------------------------------------------------
Homework is like a jucy steak, its rarely done




From mikael at isy.liu.se  Thu Jan 20 04:34:52 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 20 Jan 2000 10:34:52 +0100 (MET)
Subject: Python Rocks!
In-Reply-To: <8654iq$fcl@news.or.intel.com>
Message-ID: <XFMail.000120103452.mikael@isy.liu.se>

On 19-Jan-00 tye4 wrote:
 >  Perfect example.
 >     if (a > 3)
 >       if (b == 4)
 >          c = 2;
 >      else
 >       c = 3;
 >  
 >  If you add a space, the 'else' will belong to the inner 'if'. If you delete
 >  one, the 'else' will belong to the outer one. Highly error prone.

I would say error detection capability. The Python interpreter detects the 
problem and reports it. 

 >  if x = 5:
 >      print 'x is 5'
 >  end if

Hmm. Repetition is the mother of all knowledge:

if x = 5:
    print 'x is 5'
#end if

is even more backwards compatible, because it has worked ever since
comments were introduced.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    10:28:15

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From wtanksle at hawking.armored.net  Fri Jan  7 00:42:55 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 7 Jan 2000 05:42:55 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net>
Message-ID: <slrn87avnr.d7v.wtanksle@hawking.armored.net>

On 6 Jan 2000 21:54:56 GMT, Samuel A. Falvo II wrote:
>In article <3874FC58.69F1D59D at rubic.com>, Jeff Bauer wrote:
>>I have written a Python plugin that works with
>>Netscape Navigator on Win32.  It used to work with
>>Microsoft IE, but I think Microsoft has dropped
>>plugin support for their recent browsers.  Can 
>>anyone confirm?

>I can't answer this question, but there are still an awful lot of web pages
>which utilize the <object class="{...clsid...}"> tag to embed various
>objects.  Is it possible to re-write your plug-in as a COM object which can
>be embedded?

It probably is -- but that would drop the possibility of working with
Netscape.  Unless he wanted to maintain on of each, of course (which isn't
impossible).

>(Personally, I think the COM solution is a superior one anyway, but that's
>just my opinion.)

Sure it is -- but compatibility is even better.

Hmm...  JPython might be worthwhile.

>>As Mozilla becomes more stable, I think it will
>>be the ideal browser to use for Python client-side
>>scripting.

>Mozilla will become stable only after I'm old and grey.  And I'm only 24
>now.  I don't have much faith in Mozilla -- the codebase is just plain too
>big and too complicated.

Mozilla is right now (M12) almost as stable than Netscape 4.7 or IE 5.0
(although it still lacks some important features, and both of those
browsers are terrible for stability).  The codebase is big and
complicated, but I like the progress they're making.

>Samuel A. Falvo II

-- 
-William "Billy" Tanksley, in hoc signo hack


From see at below  Sat Jan 22 08:01:55 2000
From: see at below (Paul Foley)
Date: 23 Jan 2000 02:01:55 +1300
Subject: In defence of the two-namespace rule
References: <E12BheE-0005Zo-00@lsls4p>
Message-ID: <m2iu0muunw.fsf@mycroft.actrix.gen.nz>

On Fri, 21 Jan 2000 17:14:58 +0000, Edward Welbourne wrote:

> As to passing the default through globals, the problem then gets
> complicated by the fact that, of course, that dialog gets used by loads
> of bits of code, so someone writes this helper:

> def relativeto(root):
>     def result(*fragments, r=root):
>         return apply(os.path.join, (r,) + fragments)
>     return result

You don't need the (*args, var=...) thing.  You can do this, instead:

  class relativeto:
      def __init__(self, root):
         self.root = root
      def __call__(self, *fragments):
         return apply(os.path.join, (self.root,)+fragments)

-- 
C/C++/Java/Perl/etc are for people who want to make things that work.
Common Lisp is for people who want to make things that don't break.
                                                         -- Erik Naggum
(setq reply-to
  (concatenate 'string "Paul Foley " "<mycroft" '(#\@) "actrix.gen.nz>"))


From pauldubois at home.com  Sun Jan 16 15:47:09 2000
From: pauldubois at home.com (Paul F. Dubois)
Date: Sun, 16 Jan 2000 20:47:09 GMT
Subject: Numerical Python Project
Message-ID: <h7qg4.1288$Fb1.38931@news1.frmt1.sfba.home.com>

The Numerical Python Project is in the process of establishing itself at
SourceForge, http://numpy.sourceforge.net.

To register as a user of SourceForge go to http://sourceforge.net.

Use the Numerical Python webpage above to get to the Numerical Python
Project page at SourceForge. On that page you will find links to current
releases, the CVS repository, bug tracking facility, etc. We are just
getting up and running but you can check out the sources using anonymous CVS
by following directions given in the CVS link. (Note: unlike FTP, the
anonymous password required is blank -- just use the Enter key. They say
this but up above the lines that catch your eye).

The project developers are David Ascher, Travis Oliphant, and myself. Once
we are settled in we will entertain requests to become codevelopers.
However, for now we would like to keep it simple.

The remainder of the LLNL-distributed packages will be relocated as well
during the next two months. The current web server, xfiles.llnl.gov, is both
unstable and about to disappear behind a firewall. The ftp site
ftp-icf.llnl.gov will be redirected to a new machine at that time.  The
situation now is: Numerical only is located at SourceForge; other packages
haven't been updated and so the LLNLDistribution11.tgz file is the "latest"
for those.

Due to the new firewall plans, it is better not to try to distribute this
software from LLNL itself, and by some miracle SourceForge has appeared in
the nick of time and offers us a truly superior location. This switch does
not represent any diminution of LLNL's participation or interest in the
Python scene.

Parties willing to make and install binary releases for Windows and Mac will
be needed. For the moment, only source is available at the SourceForge site.






From nobody at nowhere.nohow  Sat Jan 22 22:48:46 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Sun, 23 Jan 2000 03:48:46 GMT
Subject: Python online docs are great!
References: <slrn88kouc.pej.nobody@isis.visi.com> <86dpnm$dtg$1@nntp9.atl.mindspring.net>
Message-ID: <slrn88kugu.r5l.nobody@isis.visi.com>

In article <86dpnm$dtg$1 at nntp9.atl.mindspring.net>, Aahz Maruch wrote:
>In article <slrn88kouc.pej.nobody at isis.visi.com>,
>Grant Edwards <nobody at nowhere.nohow> wrote:
>>
>>I just want to compliment GvR on the Python online documentation files.
>>They are amazingly well done.
>
>Fred Drake deserves much of the credit here. I can say from personal
>experience that the docs have improved greatly over just the last year.

I certainly didn't mean to slight anybody else who has worked on the
documentation or on the rest of Python.  I should have included an "et
al" in my original post.

As the next step in my never-ending battle with the BOFH, I am just
finishing my first Python program: a utility that fetches e-mail
(including attachments) from MS Exchange server using COM/MAPI,
generates an RFC822/MIME compliant message and forwards it to an SMTP
server.  I'd never used COM or MAPI before (heck, I'd never programmed
for Win32 at all), and the thing is basically working after a couple
days.  (And that includes a COM/MAPI simulator/test-stub module that
lets me test my application under Linux).

The richness of the Python modules in the standard distribution is
amazing, and the ease of development and debugging in Python makes
using those modules a pleasure.

I guess I'm preaching to the choir, eh?

-- 
Grant Edwards                   grante             Yow!  I had pancake makeup
                                  at               for brunch!
                               visi.com            


From johnm at magnet.com  Fri Jan 28 18:43:35 2000
From: johnm at magnet.com (John Mitchell)
Date: Fri, 28 Jan 2000 18:43:35 -0500
Subject: statements and blocks
In-Reply-To: <86t0ov$4en$1@news.psc.edu>
Message-ID: <Pine.SGI.4.05.10001281833490.13547-100000@lemur.magnet.com>

On 28 Jan 2000, John Kochmar wrote:

> OK, I haven't seen this in the FAQ (Yeah, I did look ;^), and maybe I 
> just missed it, but someone at the Python conference (sorry, I forget
> which talk) brought up that you couldn't do:
> 
> 	print "a"
> 	print "b"
> 		print "c"
> 		print "d"
> 		print "e"
> 	print "f"
> 
> to visually set off a block of statements.  And ahort of forcing a different 
> scope though a conditional, loop, or some other scope modifier, I haven't 
> been able to create a block of statements in a seperate context.

I'd say just do the "if 1" think and deal with it:

       print "a"
       print "b"
       if 1:
            print "c"
            print "d"
            print "e"
       print "f"

I use this technique a *lot*, but mostly for testing: use "if 0" to take
code out, "if 1" to re-use it.  Its also useful as "if 0: xxx else: yyy".
One code change swaps two segments of code.

One talk at the conference ("Python for Scene and Model Description for
Computer Graphics") used too many pushes and pops.  I suggested using
tuple- or list-format to make things a little better.  Now that I think
about it, it would probably suck.  Hmph.


- j





From aahz at netcom.com  Tue Jan 18 10:30:56 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jan 2000 15:30:56 GMT
Subject: Any Newbies?
References: <s2Wf4.974$sv3.3886@newsfeed.slurp.net> <20000118074115.A664@stopcontact.palga.uucp>
Message-ID: <8620vg$v7c$1@nntp2.atl.mindspring.net>

In article <20000118074115.A664 at stopcontact.palga.uucp>,
Gerrit Holl  <gerrit at nl.linux.org> wrote:
>TechSpood wrote on 947899478:
>>
>> Anyone besides me out there almost a complete newbie to all programming and
>> is learning python for a first language?
>
>I did almost a year ago. I'm 14 years old.

Not only that, but he's a big help with other newbies now.  Thanks,
Gerrit!
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From alan.gauld at gssec.bt.co.uk  Wed Jan  5 09:14:34 2000
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Wed, 05 Jan 2000 14:14:34 +0000
Subject: Python and ASP
References: <_ji34.8719$523.343739@news.chello.at> <bBk34.1288$Cl3.6689@news-server.bigpond.net.au> <slrn84t6n0.q3j.aa126@fan1.fan.nb.ca> <BZA34.2308$Cl3.11615@news-server.bigpond.net.au>
Message-ID: <387351CA.1EA77B30@gssec.bt.co.uk>

Mark Hammond wrote:
> No - you can do Python in ASP.

Care to elaborate?

> Not really.  Python programs running under ASP make extensive use of the ASP
> object model, and hence won't work outside an ASP environment.  

Again, how is that done? Any Web pages you can reference?
I don't see any ASP references on python.org

> Just like
> VBScript or JScript code designed for ASP wont work outside of ASP.

On a (tangentially) related theme has anyone written whatever 
plumbing is needed to make python work under 
Windows Scripting Host? MS do VBscript and JScript and 
mention a Perl version available for download. 
Is there a Python one?
I know the pythonwin extensions render it semi redundant 
but it would be nice to keep up with the Jones :-)

> including those sent to mailing lists).  4 spams per week, and I have been
> using a unprotected email address on Usenet for years and years.  What
> exactly are you afraid of?

Maybe the 7 or 8 spans I get per _day_... and thats after 
filtering which catches another 3 or 4!

But OTOH, I still don't protect my mail coz I find them such 
a pain. I just don't reply to protected addresses!

Alan G.
-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.


From nolan_d at bigfoot.com  Sun Jan  2 21:15:18 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Sun, 02 Jan 2000 20:15:18 -0600
Subject: Python language spec
Message-ID: <E124x1i-00053N-00@ethereal>

I have a long-standing tradition of taking up personal projects in
which I bite off much more than I can chew. The end result is that
several weeks later I emerge with a bit more knowledge, and either a
semi-working product or much less hair on my head. :) My most recent
attempt is a Python interpreter written to compile a subset of Python
into Muq (http://muq.org/~cynbe/muq/) bytecodes for execution natively
on the server platform.

As I've stated, this project is probably more than I can accomplish,
considering the fact that yesterday I knew nothing about Yacc, and
very little about compiler design. :) I've been cramming for the past
day or so though, and feel that I have made a few strides in
understanding what I'll have to do. Today I've been looking through
the Python language reference, working on rules for expressions.

Here's where my confusion lies. Maybe I'm studying the language
reference and taking it too literally. I'm working slowly through,
section-by-section, writing rules for each construct and testing
them. I'm confused by the expression_list definition, however. It
reads:

expression_list:      expression ("," expression)* [","]

I understand the syntax of the definition, but I don't know what the
syntax for an expression should be. Earlier rules define u_expr,
a_expr, etc. but not a regular expression (I don't mean regexp, just a
no-frills expression :). The only rule which defines 'expression' is:

expression:     or_test | lambda_form

which seems to imply that 'expression' is only a boolean result, which
I'm guessing it isn't. :) After thinking about the problem for a bit
though, I began to wonder if I'm taking rules too literally? For
example, is:

m_expr:         u_expr | m_expr "*" u_expr ...

meant to also describe an expression, but m_expr/u_expr are used
since:

expr:         expr | expr "*" expr ...

may seem less meaningful? This makes sense to me, though I don't
completely understand why:

expr: expr "*" expr { output }
expr: expr "/" expr { output } ...

wouldn't work? In other words, to me the first part of the m_expr rule
(m_expr: m_expr | ...) seems to be redundant. Am I making any sense
here? :) Is the rule in fact defining a standard expression, but
excluding all other operators besides * and /?

If anyone can make sense of my confused ramblings and help me out, I'd
greatly appreciate it. :) I've never done anything of this sort
before, nor have I really delved into studying a language
specification, so this project is a very big learning experience for
me, and hopefully something will actually come of it. :) Thanks.



From pj at sgi.com  Mon Jan  3 15:21:14 2000
From: pj at sgi.com (Paul Jackson)
Date: 3 Jan 2000 20:21:14 GMT
Subject: RPM-interface/module
References: <386B3CDB.FF843280@bibsyst.no> <3daemsrru3.fsf@amarok.cnri.reston.va.us> <iacg0wkrp3c.fsf@vier.idi.ntnu.no> <slrn871j2e.jg1.grant@grante.comtrol.com>
Message-ID: <84r0bq$ku92t@fido.engr.sgi.com>

|> IIRC, rpm was originally written in Python,

And the current Red Hat installer, called Anaconda,
is written in Python.

-- 

=======================================================================
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 kc5tja at garnet.armored.net  Mon Jan 24 17:43:15 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 24 Jan 2000 22:43:15 GMT
Subject: Python Rocks!
References: <000d01bf6312$77edd660$31a2143f@tim> <388BB4E1.974E63B7@prescod.net>
Message-ID: <slrn895cqh.rar.kc5tja@garnet.armored.net>

In article <388BB4E1.974E63B7 at prescod.net>, Paul Prescod wrote:
>Wouldn't it be easier if we just added an optional, (but
>compiler-recognized if present) block delimiter to Python 2? Then people

Sometimes you just have to stick up for what you believe in.  I say, "No,
let's not do this."

--
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From michelorengo at netscape.com  Wed Jan 19 10:06:22 2000
From: michelorengo at netscape.com (Michel Orengo)
Date: Wed, 19 Jan 2000 15:06:22 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com>
Message-ID: <3885D3E0.4F10F28D@netscape.com>

As I told you, I am still a beginner wrt MAPI. It appears that my code worked but not always! In fact, I ran myself into several problems I don't understand. The strangest one is that
some code run in Interactive window, but will not run otherwise!
Moreover, I am using the lotus cc-mail version of MAPI. You are using the one provided by Outlook.

In fact, you are lucky because your version is 1.2
I know that because I have access to the MSDN Library, and that's where you can find all the command/properties of objects of the CDO (Collaboration Data Object) the OLE wrapping of
the MAPI.

If you do not have access to the MSDN Library, you can use a COM browser: it will give you the properties/methods of the objects.
Another solution is to make Python aware of the CDO object by using makepy (see QuickStartClientCom.html)
In the file generated by makepy, you will find all the classes and attributes.

What you want to do though should be not very complex. Here is a sample code (I didn't test it!), just to provide some idea:

session = win32com.client.Dispatch("MAPI.Session")
session.Logon("MS Exchange Settings")
inbox = session.Inbox
collmsg = inbox.Messages
msg = collmsg.GetFirst()
while msg:
    print "subject: ", msg.subject
    msg = collmsg.GetNext()
session.Logoff()

Note that you can also take a look at testExchange.py under the 'test' directory.

Hope this help.

Michel

Grant Edwards wrote:

> In article <slrn889o9j.3oa.grant at grante.comtrol.com>, Grant Edwards wrote:
>
> >------------------------------------------------------------
> >import win32com.client
> >import win32com.mapi.mapitags
> >
> >session = win32com.client.Dispatch("MAPI.Session")
> >session.Logon("MS Exchange Settings")
> >outbox=session.Outbox
> >collmsg=outbox.Messages
> >msgo=collmsg.Add("subj text","message text",win32com.mapi.mapitags.PR_MESSAGE_CLASS,1)
> >msgo.Recipients.Add("grante at visi.com")
> >msgo.Sender.Address="grant.edwards at comtrol.com"
> >msgo.Send(1,0)
> >session.Logoff
> >------------------------------------------------------------
> >
> >It bombs at line msgo.Sender.Address="" line with the following
> >error:
> >
> >com_error: (-2147221233, 'OLE error 0x8004010f', (1271, 'Collaboration Data Objects', ' [Collaboration Data Objects - [MAPI_E_NOT_FOUND(8004010F)]]', None, 0, -2147221233), None)
>
> If I comment out the msgo.Sender.Address line, then I get an
> error when I make the msgo.Send(1,0) call:
>
>   File "U:\Grante\python\testSend.py", line 11, in ?
>     msgo.Send(1,0)
>   File "<COMObject <unknown>>", line 2, in Send
>   File "C:\Program Files\Python\win32com\client\dynamic.py", line 207, in _ApplyTypes_
>     result = apply(self._oleobj_.InvokeTypes, (dispid, LCID, wFlags, retType, argTypes) + args)
> com_error: (-2147467259, 'Unspecified error', (17389, 'Collaboration Data Objects', ' [Collaboration Data Objects - [E_FAIL(80004005)]]', None, 0, -2147467259), None)
>
> ???
>
> --
> Grant Edwards                   grante             Yow!  My Aunt MAUREEN was
>                                   at               a military advisor to IKE &
>                                visi.com            TINA TURNER!!



From katz at glue.umd.edu  Wed Jan 26 19:01:16 2000
From: katz at glue.umd.edu (Roey Katz)
Date: Thu, 27 Jan 2000 00:01:16 GMT
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com> <388F140B.C0AA1C99@erols.com>
Message-ID: <388f88ca.1925884@news.erols.com>

The 'virtual classes' that Sweeney describes look to me an awful lot
like something like this:


	class myClass:
	    def __init__( self ): pass
	    def doSomething( self ): print 'instantiating'

	class myClass2:
	    def __init__( self ): pass
	    def doSomething( self ): print 'instantiating (2)'
	
	class Base:
	   
	    virtualClass = None
	
	    def __init__( self ): 
	         self.inst = self.virtualClass()
	         self.inst.doSomething()
	
	
	class Derived1( Base ):
	    virtualClass = myClass
		
	class Derived2( Base ):
	    virtualClass = myClass2


-----------------
Isn't this the same thing?  I noticed that the virtual classes in
Sweeney's code sample were all of the same base class;  to 
me, it seems that Python's object model is one of policy, not rules:
if an object responds to a message, so be it, regardless of what
*kind* of object it is -- so we have a little more freedom with
Python.  Correct me if I'm redundant?

I enjoy these constructs because they allow me to design flexible
interfaces in the base class that I don't have to change at all as I
add new features (sounds coherent?) I'm using this for a role playing
game.


Roey


From A.Concept at wanadoo.fr  Sat Jan 29 11:52:45 2000
From: A.Concept at wanadoo.fr (Jessica Elkaim)
Date: Sat, 29 Jan 2000 17:52:45 +0100
Subject: CGI, Apache, and mxODBC
References: <6p1g4.35$iL4.1041@news.rdc1.tx.home.com> <gX1g4.48$iL4.1177@news.rdc1.tx.home.com>
Message-ID: <38931ADD.88FD9618@wanadoo.fr>

	Apparently you don't have a DSN named Allegro in your ODBC Setup.

	Add this DSN as a 'system DSN' because if Apache runs under a different
user id it may not find the correct DSN.

	Hope that helps,
	Cyril Elkaim

"Paul E. Bible" a ?crit :
> 
> Just some more information that I found relating to my problem.  I thought
> it might in the diagnosis...
> These are the error message lines from Apache:
> 
> File "c:/apache/cgi-bin/sql.py", line 13, in Main
> db = Odbc.Windows.Connect('Allegro','helistest','oilwell')
> mxODBC.OperationalError: ('IM002', 0, '[Microsoft][ODBC Driver Manager] Data
> source name not found and no default driver specified', 4265)
> 
> Thanks again,
> Paul
> 
> "Paul E. Bible" <pbible at home.com> wrote in message
> news:6p1g4.35$iL4.1041 at news.rdc1.tx.home.com...
> > Hi all,
> >
> > First of all, sorry for the length of this message...
> >
> > Okay, I've been successful in getting Python CGI scripts to run under
> Apache
> > 1.3.9 under Windows/NT.  Now it's time to add database functionality using
> > mxODBC to the script, but I'm encountering some problems.  I have provided
> > my script and the Apache environment debug listing below.
> >
> > The script, as written, gives me a Internal Server Error.  If I comment
> out
> > the lines dealing with mxODBC (i.e., db = Odbc.Win..., etc.), the script
> > works fine; just doesn't do much.
> >
> > Thanks in advance your help,
> > Paul
> >
> > --------------------------------------------------------------------------
> --
> > ---
> > #!c:\progra~1\python\python.exe
> >
> > import os, string, sys, cgi, re, Odbc
> >
> > def Main():
> >      sql = 'Select * from cashacct'
> >
> >      db = Odbc.Windows.Connect('TestDB','testid','testpw')
> >
> >      # Add the columns
> >      cols = ()
> >      c = db.cursor()
> >      c.execute(sql)
> >      cols = c.description
> >
> >      # Print the HTML header info
> >      print "Content-type: text/html"
> >      print
> >      print "<HTML>"
> >      print "<HEAD>"
> >      print "SQL Statement Test"
> >      print "</HEAD>"
> >      print "<BODY>"
> >
> >      # Add the columns to the Access table
> >      print '<table border="0">'
> >      print "<tr>"
> >      for idx in range(len(cols)):
> >           print "<td>" + cols[idx][0] + "</td>"
> >      print "</tr>"
> >      print "</table>"
> >
> >      print "</BODY>"
> >      print "</HTML>"
> >      sys.exit(0)
> >
> > Main()
> >
> > --------------------------------------------------------------------------
> --
> > ---
> >
> > Shell Environment:
> > COMSPEC
> > D:\WINNT\system32\cmd.exe
> > DOCUMENT_ROOT
> > c:/apache/htdocs
> > GATEWAY_INTERFACE
> > CGI/1.1
> > HTTP_ACCEPT
> > image/gif, image/x-xbitmap, image/jpeg, image/pjpeg,
> > application/vnd.ms-powerpoint, application/vnd.ms-excel,
> application/msword,
> > application/x-comet, */*
> > HTTP_ACCEPT_ENCODING
> > gzip, deflate
> > HTTP_ACCEPT_LANGUAGE
> > en-us
> > HTTP_CONNECTION
> > Keep-Alive
> > HTTP_HOST
> > XX.X.XX.XXX
> > HTTP_REFERER
> > http://XX.X.XX.XXX/blank.htm
> > HTTP_USER_AGENT
> > Mozilla/4.0 (compatible; MSIE 5.01; Windows NT)
> > PATH
> >
> C:\orant\bin;D:\WINNT\system32;D:\WINNT;C:\RECYCLER\S-1-5-21-862944694-15761
> >
> 02806-697575874-500\DC0;C:\RECYCLER\S-1-5-21-862944694-1576102806-697575874-
> > 500\DC0\Perl;c:\Program Files\Symantec\pcAnywhere;c:\Program Files\python
> > PYTHONPATH
> > c:\Program Files\python;c:\Program Files\Lib;c:\Program
> > Files\python\lib\tkinter;c:\Program Files\python\win32com;c:\Program
> > Files\python\win32
> > PYTHONUNBUFFERED
> > 1
> > QUERY_STRING
> > sql=select+*+from+cashacct&submit-button=Submit
> > REMOTE_ADDR
> > XX.X.XX.XXX
> > REMOTE_PORT
> > 1052
> > REQUEST_METHOD
> > GET
> > REQUEST_URI
> > /cgi-bin/sql.py?sql=select+*+from+cashacct&submit-button=Submit
> > SCRIPT_FILENAME
> > c:/apache/cgi-bin/sql.py
> > SCRIPT_NAME
> > /cgi-bin/sql.py
> > SERVER_ADDR
> > XX.X.XX.XXX
> > SERVER_ADMIN
> > webmaster at home.com
> > SERVER_NAME
> > XX.X.XX.XXX
> > SERVER_PORT
> > 80
> > SERVER_PROTOCOL
> > HTTP/1.1
> > SERVER_SIGNATURE
> > <ADDRESS>Apache/1.3.9 Server at 24.6.21.112 Port 80</ADDRESS>
> > SERVER_SOFTWARE
> > Apache/1.3.9 (Win32)
> > SystemRoot
> > D:\WINNT
> > WINDIR
> > D:\WINNT
> > None
> > These environment variables could have been set:
> > AUTH_TYPE
> > CONTENT_LENGTH
> > CONTENT_TYPE
> > DATE_GMT
> > DATE_LOCAL
> > DOCUMENT_NAME
> > DOCUMENT_ROOT
> > DOCUMENT_URI
> > GATEWAY_INTERFACE
> > LAST_MODIFIED
> > PATH
> > PATH_INFO
> > PATH_TRANSLATED
> > QUERY_STRING
> > REMOTE_ADDR
> > REMOTE_HOST
> > REMOTE_IDENT
> > REMOTE_USER
> > REQUEST_METHOD
> > SCRIPT_NAME
> > SERVER_NAME
> > SERVER_PORT
> > SERVER_PROTOCOL
> > SERVER_ROOT
> > SERVER_SOFTWARE
> > In addition, HTTP headers sent by the server may be passed in the
> > environment as well. Here are some common variable names:
> > HTTP_ACCEPT
> > HTTP_CONNECTION
> > HTTP_HOST
> > HTTP_PRAGMA
> > HTTP_REFERER
> > HTTP_USER_AGENT
> > None
> >
> >
> >


From darrell at dorb.com  Thu Jan  6 19:16:41 2000
From: darrell at dorb.com (Darrell)
Date: Thu, 6 Jan 2000 19:16:41 -0500
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com>
Message-ID: <011501bf58a4$7b0db540$0100a8c0@rochester.rr.com>

Using an exception as a "switch" statement is a cool idea.
A dictionary approach comes to mind, with so many states.

From: "Skip Montanaro" <skip at mojam.com>
>
> Running the above yields:
>
> Time: 5.52181506157
> Time: 4.23444199562
>

Skip you need a faster computer :)
I get these times, with a 400mhz PII.
Time: 2.65600001812
Time: 1.90699994564
Yes I know, who cares.

Increased size by 100x
size=100000
Time: 265.381999969
Time: 191.114000082


# Dictionary based switch statement
def action():
    pass

stateDic={}
s = string.letters + string.digits
for i in s:
    stateDic[i]=action

def defaultCase():
    pass

def func3(size, stateDic, default=defaultCase):
    s = string.letters + string.digits
    for i in xrange(size):
        for l in s:
            action=stateDic.get(l,default)

size=100000
t1=time.time()
func3(size, stateDic)
print 'Time:',time.time()-t1

I didn't call the action since our previous examples just did "pass"
Time: 8.51600003242

The -X switch was interesting. But I doubt it would be useful in general ???
It amazes me what many eyes can see.

--Darrell






From gmcm at hypernet.com  Sat Jan  1 14:20:50 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 1 Jan 2000 14:20:50 -0500
Subject: need help getting tk to work with python
In-Reply-To: <84lje7$hrj$1@news00.btx.dtag.de>
Message-ID: <1265368930-15727554@hypernet.com>

Eide wrote:

> I too have been having this problem.  Tk once worked, but does no
> longer. I run W98 with python at C:\program files\python and tcl
> at C:\program files\tcl
> 
> I tried the autoexec.bat fix as suggested at python.org to no
> avail.  

> My autoexec.bat reads:
> 
> PATH
> C:\WINDOWS;C:\WINDOWS\COMMAND;C:\PROGRA~1PYTHON;C:\PROGRA~1\PYTHO
> N\LIB;C:\PROGRA~\TCL\BIN

The intent of the last entry is correct, but it's spelled wrong. 
That should probably be "PROGRA~1".



- Gordon



From mhammond at skippinet.com.au  Thu Jan 27 11:52:39 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 27 Jan 2000 16:52:39 GMT
Subject: What is winfo_id() under Win32
References: <3890439A.1C1BA6D2@freemail.hu>
Message-ID: <86pt4k$f6h$1@nnrp1.deja.com>

In article <3890439A.1C1BA6D2 at freemail.hu>,
  Attila =?iso-8859-1?Q?Filet=F3th?= <tothfile at freemail.hu> wrote:
> Hi!
>
> class app(Toplevel):
>     def __init__(self, master):
>         self.master = master
>         Toplevel.__init__(self, master)
>         self.update_idletasks()
>         hwnd = self.winfo_id()
> #       hwnd = win32ui.GetParent(hwnd)  # Attribute error: GetParent
>         win32gui.ShowWindow(hwnd, SW_SHOWMAXIMIZED)  # It would only
> work after the GetParent() call

win32gui _should_ include these functions.  If I remember, I will add
them.

In the meantime, the following _should_ work:

wnd = win32ui.CreateWindowFromHande()
parent = wnd.GetParent()

Mark.


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


From ScrewMe at mypad.com  Tue Jan 18 19:28:48 2000
From: ScrewMe at mypad.com (TechSpood)
Date: Tue, 18 Jan 2000 18:28:48 -0600
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
Message-ID: <2t7h4.218$Pe6.151@newsfeed.slurp.net>

I agree totally with that...I am a newbie too, and I since I am used to
basic I am used to seeing an end of loop block and indention being
optional....but oh well
"tye4" <tye4 at yahoo.com> wrote in message
news:860ftk$bas at news.or.intel.com...
> I'm a python newbie.. but I like Python better than Perl (no ugly @$#@..
in
> syntax).
>
> Python does have one quirk: no end if or end loops
>
> if x == 10:
>     print 'x is 10'
> print 'Hello world'
>
> # easy to get confused whether this line is in 'if' block or not
> Having an 'end' block would help tremendously.
>
> if x == 10:
>     print 'x is 10'
> end if
> print 'Hello World'
>
> or
> for x in range(1, 10):
>     print x,
> end loop
>
> Indentation is a confusing and lame way to separate what is inside a block
> and what's outside it.
>
> tye4
>
>




From tjg at avalongroup.net  Wed Jan 19 17:34:19 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Wed, 19 Jan 2000 14:34:19 -0800
Subject: Mapping Column Names to results using DBI
Message-ID: <38863BEB.BCB9E15C@exceptionalminds.com>

Hi again,

Right now I have an app that works ok, but I forsee a problem in the
future.

As you know, the results from a database query are returned in a tuple.
I can access those results by position just fine, is it possible to map
the column names onto the results, so I can access columns by name
rather than by number?

While indexing by number works, it is incredibly fragile and will start
failing the minute the underlying DB changes.

Once again, it appears to me from the DBI spec that there should be a
way to do this, I just can't figure it out from the spec.

I'm using mxODBC if that helps.

Thanks.
-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From davidm at softdesigns.com  Fri Jan 21 09:22:35 2000
From: davidm at softdesigns.com (David Metzler)
Date: Fri, 21 Jan 2000 09:22:35 -0500
Subject: database connection
Message-ID: <s8grjo5jeh697@corp.supernews.com>

I would like to create some py programs that connect to a database like
mysql or postgres on either win 98 or Linux.
I could not find the file to include. The python.org directions talk about
compiling but I do not have MS C.
Is there a mysqldb available to install in Win 98 without compiling.
How about a rpm or binary for RH 6.0

If not for mysql perhaps another database?





From jorn.h.b.dahl at stud.jbi.hioslo.no  Fri Jan 28 07:14:29 2000
From: jorn.h.b.dahl at stud.jbi.hioslo.no (Jørn Helge B. Dahl)
Date: 28 Jan 2000 13:14:29 +0100
Subject: html parsing (htmllib)
References: <280120000015041211%zippy@cs.brandeis.edu>
Message-ID: <m3vh4etmu2.fsf@pc-432.p52s.hioslo.no>

* Patrick Tufts <zippy at cs.brandeis.edu>

> I want to grab the text from a set of web pages.  htmllib seems like
> what I want, except that want to assign the parsed text to a variable
> instead of printing it to stdout or a file.
> 
> What do I need to do?
> 
> And better yet, has someone already done it?

As a matter of fact, I have. Beware of ugly code. But it gives you an
idea. (I don't remember which module I snatched it from, may be
sgmmlib.py or formatter.py)

## -- code begins --

# Remember this:
import string

from htmllib import *
from formatter import *

class DumbVariableWriter(NullWriter):

    def __init__(self, variable, maxcol=72):
        self.variable = variable
        self.maxcol = maxcol
        NullWriter.__init__(self)
        self.reset()
        
    def reset(self):
        self.col = 0
        self.atbreak = 0

    def send_paragraph(self, blankline):
        self.variable.append('\n' + '\n'*blankline)
        self.col = 0
        self.atbreak = 0

    def send_line_break(self):
        self.variable.append('\n')
        self.col = 0
        self.atbreak = 0

    def send_hor_rule(self, *args, **kw):
        self.variable.append('\n')
        self.variable.append('-'*self.maxcol)
        self.variable.append('\n')
        self.col = 0
        self.atbreak = 0

    def send_literal_data(self, data):
        self.variable.append(data)
        i = string.rfind(data, '\n')
        if i >= 0:
            self.col = 0
            data = data[i+1:]
        data = string.expandtabs(data)
        self.col = self.col + len(data)
        self.atbreak = 0

    def send_flowing_data(self, data):
        if not data: return
        atbreak = self.atbreak or data[0] in string.whitespace
        col = self.col
        maxcol = self.maxcol
        for word in string.split(data):
            if atbreak:
                if col + len(word) >= maxcol:
                    self.variable.append('\n')
                    col = 0
                else:
                    self.variable.append(' ')
                    col = col + 1
            self.variable.append(word)
            col = col + len(word)
            atbreak = 1
        self.col = col
        self.atbreak = data[-1] in string.whitespace

## -- end code --

Usage:

# List to temporarily hold the string
list_of_output = []
parser = HTMLParser(AbstractFormatter(DumbVariableWriter(list_of_output))) 
parser.feed(open('test.html').read())

# Smash it all together in one string
result = string.join(list_of_output, '')
-- 
This goes to eleven


From soundhack at hotmail.com  Tue Jan 18 12:38:47 2000
From: soundhack at hotmail.com (Robert)
Date: Tue, 18 Jan 2000 12:38:47 -0500
Subject: PythonWin and Dialogs in DLLs
References: <38839d87$0$22473@senator-bedfellow.mit.edu> <388418ec.8757803@news.ricochet.net> <k5_g4.3484$iy5.225935@typ12.deja.bcandid.com>
Message-ID: <3884a59a$0$22486@senator-bedfellow.mit.edu>

Joel Lucsy wrote in message ...
>Nope, simply add /noentry in the link tab settings. No DllMain/WinMain/or
>main required.


hmm never got the message you replied to, but thanks! this is what I was
looking for (thanks to others who replied too, I'll check out the
PyResWizard
after I figure out how to do this myself!) . I was able to halve the size of
the release dll with this.

now one more question. I have gotten the subclassed dialog to show
up, but am having problems with assigning message handlers to buttons.
Following one of the demos, I did something like the following

(in the dialog's InitDialog method)
    self.button=self.GetDlgItem(IDC_CMD_BUTTON)
    self.button.HookNotify(self.OnButton,win32con.BN_CLICKED)

When i click on the button, the OnButton handler isnt getting called.
I tried self.button.HookCommand() too, with no results. The one thing
that sorta worked was
self.button.HookMessage(self.OnButton,win32con.WM_LBUTTONDOWN), which is not
exactly what
I wanted (sometimes the button click animation isnt shown)

Thanks again for all the help!
Robert





From michael.perlbach at uni-konstanz.de  Fri Jan 14 09:26:14 2000
From: michael.perlbach at uni-konstanz.de (onk)
Date: Fri, 14 Jan 2000 14:26:14 +0000
Subject: latex formatter usable with htmllib ?
Message-ID: <387F3206.330929FF@uni-konstanz.de>

I'm looking for a html to latex converter which can easily be adapted to
special makros -
does anyone know about a latex formatter which works with the htmllib
module?
Excuse me if this is a FAQ - mail response is preferred, since our news
server does not
regularely serve this news group.

Thank you,

Strubi




From hanche at math.ntnu.no  Fri Jan 28 00:36:03 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 28 Jan 2000 00:36:03 -0500
Subject: Wanted: A python icon for X11
References: <pcosnzl4bpf.fsf@math.ntnu.no> <388F005F.D5A0BE94@dcs.kcl.ac.uk>
Message-ID: <pcoya9azrjw.fsf@math.ntnu.no>

+ Laurence Tratt <tratt at dcs.kcl.ac.uk>:

| Harald Hanche-Olsen wrote:
| 
| > Would anybody perhaps have some pixmap lying about that might be
| > (or become) a suitable icon for IDLE windows under X11?  [...]
| 
| When I ported Python to RISC OS, I got Mark Hammonds permission and
| pinched the bitmap from the Windows port [...]

Ah, of course.  After a bit of digging around it turned out quite easy
to do:  Since I have a dual-boot Windows/FreeBSD machine (where I
never use the Windows bit, but nevertheless I have installed python
there too), all I needed to do is mount the Windows disk and do

; convert '/win98/c/Program Files/Python/py.ico' py.xpm

(Ain't Imagemagick great?)  The resulting 32x32 X icon is short enough
to include right here:
/* XPM */
static char *magick[] = {
/* columns rows colors chars-per-pixel */
"32 32 7 1",
"  c Gray0",
". c Blue",
"X c Green",
"o c Red",
"O c #808000",
"+ c Yellow",
"@ c None",
/* pixels */
"@@@@@@@  @   @@@@@@@@@@@@@@@@@@@",
"@@@@@@ OO OOO @@@@@@@@@@@@@@@@@@",
"@@@@@       OO @@@@@@@@@@@@@@@@@",
"@@@@@ ++ +++ O @@@@@@@@@@@@@@@@@",
"@@@@@ +. +.. O @@@@@@@@@@@@@@@@@",
"@@@@@ +. +.. O @@@@@@@@@@@@@@@@@",
"@@@@@ ++ +++ O  @@@@@@@@@@@@@@@@",
"@@@@    O   OOO    @@@@@@@@@@@@@",
"@@  OOOOOOOOOOO OOO  @@@@@@@@@@@",
"@ OOOOOOOOOOOOOX OOOO @@@@@@@@@@",
" OOOOOOOOOOOXOXXX OOOO @@@@@@@@@",
" O OO OOOOXOXXXXX OOOO @@@@@@@@@",
" OOOOOOOXOXX  XXX OOOOO @@@@@@@@",
" XXOXOXOXXX  XOX XOOOOO @@@@@@@@",
" XXXXXXXX   OXX XXXOOOO @@@@@@@@",
"@ XOXOXO   XOX  OOXOOOO @@@@@@@@",
"@@   oo   OXX @ XXXOOOO @@@@@@@@",
"@@  oo  OXXX @@ OXOOOO @@@@@@@@@",
"@@ ooXOXXX  @@ XXOOOOO @@@@@@@@@",
"@@ooXXXX  @@ OOXOOOOO @@@@@@@@@@",
"@oo     @@@ XXXOOOOO @@@@@@@@@@@",
"o at o@@@@@@ OOXXOOOOO @@@@@@@@@@@ ",
"@@o@@@@  XXXOOOOOO @@@@@@@@@@@  ",
"@@@@@@ OOOXOOOOOO    @@@   @@ O ",
"@@@@@@ XXXOOOOOO OOOO @ OOO  OO ",
"@@@@@ XXXXOOOOO OOOOOO OOOOOOOO ",
"@@@@@ OOOXOOOOOOOOOOOOOOOOOOOO @",
"@@@@@ XXXXOOOOOOOOOOOOOOO  XX @@",
"@@@@@ XXXXXOOOOOXOOOOOOO @@  @@@",
"@@@@@@ OOOXXXXX  XXXXX  @@@@@@@@",
"@@@@@@@ XXXOX  @@     @@@@@@@@@@",
"@@@@@@@@     @@@@@@@@@@@@@@@@@@@"
};

I can do the others too (there is pyc, where the snake has closed its
eyes, and three identical looking icons with the snake framed inside a
window), if there is any interest.

but-this-is-just-what-I-wanted-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From robert_kuzelj at my-deja.com  Wed Jan  5 05:22:50 2000
From: robert_kuzelj at my-deja.com (robert_kuzelj at my-deja.com)
Date: Wed, 05 Jan 2000 10:22:50 GMT
Subject: [ANN] gumby.py 0.1.5 (was java python console)
Message-ID: <84v61h$jg9$1@nnrp1.deja.com>

hi,

here is the java-console for all the gumbies out there ;-)

i dont have much time at the moment for
further development (max 2-4 hours a week).

but i guess its has been getting more stable
and more useable (let me know if you disagree <g>).

you find it under http://www.itemj.com/produkte.htm.
ciao robertj

ps: any help for development/testing is appreciated.

itemj
http://www.itemj.com
Robert Kuzelj          mobil 0177  5302230
Ramonvillestr.6        tel   06039 930223
61184 Karben           fax   06039 2224


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


From aa8vb at yahoo.com  Wed Jan  5 09:18:21 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 5 Jan 2000 09:18:21 -0500
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
In-Reply-To: <uBCWdBAkv0c4Ewdy@jessikat.demon.co.uk>
References: <841d0c$j0n$1@violet.singnet.com.sg> <387f6d37.25522789@news.demon.co.uk> <19991230065516.A778039@vislab.epa.gov> <W8TkoEATH4a4Ewpu@jessikat.demon.co.uk> <387378c0.4549271@news.demon.co.uk> <xc1BqAAkwpb4EwWW@jessikat.demon.co.uk> <20000105082703.A9441@vislab.epa.gov> <uBCWdBAkv0c4Ewdy@jessikat.demon.co.uk>
Message-ID: <20000105091821.D9441@vislab.epa.gov>

Robin Becker:
 |Randall Hopper:
 |>Robin Becker:
 |> |how arithmetic errors are handled using built in IEEE exceptions and
 |> |how does one gracefully crash a python script? (eg an abend/abort).
 |>
 |>raise ArithmeticError, "Oops"
 |>raise ZeroDivisionError, "Ouch"
 |>
 |>?
 |>
 |yes that's python, but what's the best way to do it in an existing C
 |library?

And in the function that returns control to Python, do this:

    PyErr_Format( PyExc_ArithmeticError, "Oops\n" );
    return NULL;

The first statement can be invoked elsewhere if desired.

-- 
Randall Hopper
aa8vb at yahoo.com



From fredrik at pythonware.com  Fri Jan 14 03:22:59 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 14 Jan 2000 09:22:59 +0100
Subject: file width and height
References: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com> <ubqf4.3018$hN5.92483@typhoon.nyroc.rr.com>
Message-ID: <004b01bf5e68$92965a40$f29b12c2@secret.pythonware.com>

Darrell <news at dorb.com> wrote:
> Sposhua wrote:
> >Is there a module to find gif/jpeg width and height?
> 
> I hacked this out of PIL and it should work for gif and jpg.

"""
hacked for PIL by Darrell Gallion
"""

where "hacked" means "copied 150 lines of source
code, and removed the copyright notice"...

</F>





From tmick at mail.com  Wed Jan  5 19:52:51 2000
From: tmick at mail.com (tmick at mail.com)
Date: Thu, 06 Jan 2000 00:52:51 GMT
Subject: Compiling on Alpha-Win2K/WinNT (Python 1.5.2)
Message-ID: <850p0u$q6a$1@nnrp1.deja.com>

I am trying to compile Python 1.5.2 (as extracted from py152.tgz on
www.python.org) for Win2000 on an Alpha. Has anyone done this? I
presume that it should be similar to compiling for Alpha/WinNT.

This is my method (please correct me if I am wrong):
 - load "Python-1.5.2\PCbuild\pcbuild.dsw" in MS Dev Studio 5.0
 - set the "python15" project as the active project
 - build

My first problem is that all of the build icons/menu-entries in
DevStudio are grayed out (i.e. I can't build the project). Does anyone
know what this indicates?

I then tried this:
 - export a makefile for the "python15" project
 - run nmake on it: "nmake python15.mak"

This got me further (the .c files were being compiled) but the
compilation broken with the following error:

--------------------------------------------
G:\trentm\Python-1.5.2\PCbuild>nmake python15.mak

Microsoft (R) Program Maintenance Utility   Version 1.62.7128
Copyright (C) Microsoft Corp 1988-1997. All rights reserved.

No configuration specified. Defaulting to python15 - Win32 Debug.

cl.exe /nologo /MDd /W3 /Gm /GX /Zi /Od /I "..\Include" /I "..\PC" /D "_
DEBUG" /D "USE_DL_EXPORT" /D "WIN32" /D "_WINDOWS" /Fp".\temp-
debug\python15.pch
" /YX /Fo".\temp-debug\\" /Fd".\temp-
debug\\" /FD /c ..\Modules\fpectlmodule.c
fpectlmodule.c
..\Modules\fpectlmodule.c(167) : fatal error C1083: Cannot open include
file: 'machine/fpu.h': Permission denied
NMAKE : fatal error U1077: 'cl.exe' : return code '0x2'
Stop.
--------------------------------------------

I do not know of a file called "fpu.h" and cannot find one on my system.


Any help would be appreciated. Thanks in advance.

Trent
tmick at mail.com


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


From boud at rempt.xs4all.nl  Sat Jan  8 02:26:32 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 8 Jan 2000 07:26:32 GMT
Subject: Getting SWIG to work with C++/Python
References: <9mad4.42353$Ke.27158@news21.bellglobal.com>
Message-ID: <856or8$dib$1@news1.xs4all.nl>

Jason Maskell <backov at nospam.csolve.net> wrote:
> Well, I bit the bullet and reinstalled VC5, and got SWIG to compile. Now, I
> try "swig -c++ -python interf~1.cpp" and it chokes, hard. I try it with the
> h, same thing.. I've tried variations of that, and I keep getting loads of
> syntax errors, etc. Did it compile wrong? Is there some hidden syntax I
> should be using? The other possibility is that since the files I'm swigging
> are Borland C++ and SWIG itself was compiled with VC++.. But I can't see
> that being a problem.

I thought I'd just mention sip, which is a small swig especially
for C++ - Phil Thompson, the author, developed it for the Qt and
KDE bindings - but its scope is wider than that:

   Some Background When I started to develop these bindings I (like
   everybody else) had a look at  SWIG. I came to the conclusion that,
   while the concept is excellent, the C++ support isn't really up to it.
   I then wrote SIP (so called because it is a small SWIG) which does
   a similar job but is specifically designed for generating Python
   bindings for C++ class libraries.

   SIP works in a similar way to SWIG in that it generates compilable C++
   code from a set of specification files. The format of a specification
   is similar (but not identical) to a C++ header file. SIP also includes
   a run-time support library for the generated code - so you need a
   copy of SIP even if you are not generating your own code.

   One of my design goals was to make all the features of underlying
   class library available to the Python programmer in exactly the same
   way as they would be available to the C++ programmer - this has (for
   the most part) been achieved.

You can find it at: http://www.river-bank.demon.co.uk/software/



-- 

Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt


From tjg at avalongroup.net  Tue Jan 11 15:40:32 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Tue, 11 Jan 2000 12:40:32 -0800
Subject: ODBC and mxODBC
Message-ID: <387B9540.BE617781@exceptionalminds.com>

Hi,

Has anyone got any tips on configuring mxODBC, or ODBC in general on a
Linux box?

I have a Linux box running PostgreSQL with the ODBC system installed.

I can access it just fine from a Windows machine running mxODBC, but I
have yet to be able to figure out how to get it to work from my own
development machine which is a Linux notebook running RH6.1

I can access the database just fine using the PyGreSQL module, but I
really want to use mxODBC so that I can do cross platform development.

This is probably more of a ODBC question than a Python question, but any
help would be appreciated.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From evan at tokenexchange.com  Tue Jan 18 15:44:34 2000
From: evan at tokenexchange.com (Evan Simpson)
Date: Tue, 18 Jan 2000 14:44:34 -0600
Subject: Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us> <m12AMt3-00007rC@vortigen.demon.co.uk> <al8u2kbycss.fsf@sirppi.helsinki.fi> <m3embfz6p5.fsf@greebo.nodomain.de>
Message-ID: <s89kep3peh6124@corp.supernews.com>

Bernhard Herzog <herzog at online.de> wrote in message
news:m3embfz6p5.fsf at greebo.nodomain.de...
> > Edward Welbourne <eddy at vortigen.demon.co.uk> writes:
> [snip]
> > > Three new kinds of argument come into being (with my given spelling):
> > >   * safe tunnel: after both * and **
> > >   * keyword-only: after * but not **
> > >   * positional-only: after ** but not *
[objections to proposal]
>
> I'll second that. Turning a kludge that works because of an
> implementation detail into an officially supported feature isn't a good
> idea, IMO. Support for lexical scoping is the correct approach.
[snip]
> lexical scoping would take care of that, unless I'm missing something.

Now that two people have asserted that lexical scoping is the right way to
handle this, I'd like to know what they *mean* by lexical scoping.

How would names get bound to (or just initialized from?) names in lexically
enclosing scopes?  Implicitly? Explicitly?
Would we still declare module-global variables, or would they use the same
mechanism as names in intermediate scopes?

I rather like the idea of having to explicitly bring names into an inner
scope, and with semantics identical to parameter defaults.  Why is this so
objectionable?  Is it just the syntax?

Cheers,

Evan @ 4-am




From emile at fenx.com  Wed Jan 19 11:17:36 2000
From: emile at fenx.com (Emile van Sebille)
Date: Wed, 19 Jan 2000 08:17:36 -0800
Subject: Sucking e-mail from Exchange Server vis HTTP?
Message-ID: <025301bf6298$b6dc38a0$01ffffc0@worldnet.att.net>

Grant,

Take a look at http://www.slipstick.com/exchange

Hanging out at the exchange newsgroup pretty much requires
that you adopt a degree of religion, or at least come to
understand it.  I've found in the past that slipstick.com
will show you where the backdoors are even though advising
against it.

Good luck!

--

Emile van Sebille
emile at fenx.com
-------------------


Grant Edwards <grant at nowhere.fenx.com> wrote in message
news:slrn888v95.31r.grant at grante.comtrol.com...
> That's what I used to do, but there are two problems with that:
>
>  1) MickySoft Exchange f@#ks up the headers on bounced mail so
>     that you can't "reply" to it, and in some cases you can't
>     even tell what the e-mail address of the sender was.
>
>  2) The system admin shut that feature off to try to force
>     everybody to use Outlook.
>
> I've also considered setting up a rule to save mail in a local
> mailbox that's on an SMB mountable filesystem and writing a
> program to periodically suck mail out of that mailbox.  I've no
> idea what the format of MS Outlook mailboxes is...
>
> --
> Grant Edwards                   grante             Yow!  Then, it's
off to
>                                   at               RED CHINA!!
>                                visi.com
> --
> http://www.python.org/mailman/listinfo/python-list
>






From mikael at isy.liu.se  Thu Jan 20 08:29:44 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 20 Jan 2000 14:29:44 +0100 (MET)
Subject: A modest Error (I mean Proposal)
In-Reply-To: <MODh4.1808$D4.62243@newsread1.prod.itd.earthlink.net>
Message-ID: <XFMail.000120142944.mikael@isy.liu.se>

On 20-Jan-00 Arinte wrote:
 >  I am getting this weird error???
 >  Traceback (innermost last):
 >    File "Beer", line 17, in ?
 >    File "D:\POSSApps\anotherposs app\PossDevice.py", line 37, in IOCTL
 >      if type(command)==type(2):
 >  TypeError: len() of unsized object

Maybe you have accidentally redefined type to len...

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    14:28:11

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From erno at iki.fi  Tue Jan 18 11:07:41 2000
From: erno at iki.fi (Erno Kuusela)
Date: 18 Jan 2000 18:07:41 +0200
Subject: re-opening stdin in raw (binary) mode?
References: <388388c9.12001026@news.iol.ie> <slrn8876la.bos.kc5tja@garnet.armored.net>
Message-ID: <874scbuzw2.fsf@fun77.kivikko.hoas.net>

>>>>> "Samuel" == Samuel A Falvo <kc5tja at garnet.armored.net> writes:

    Samuel> under Linux, you can perform an ioctl() on file descriptor
    Samuel> 1 (or is it 0?) to set a binary-mode flag.

There is no binary mode in unix/linux, it's a windoze (and mac?) thing.

   -- erno


From aahz at netcom.com  Tue Jan  4 19:42:29 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 5 Jan 2000 00:42:29 GMT
Subject: is variable a list?
References: <38729120.7616054D@simplypets.com>
Message-ID: <84u41l$c5s$1@nntp9.atl.mindspring.net>

In article <38729120.7616054D at simplypets.com>,
James Punteney  <jamesp at simplypets.com> wrote:
>
>Is there some way to determine whether a variable is list or a string?
>Basically I have a variable that I would like to be able to be a string
>or a list, but I need some way to tell which it is.

type(var)
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From helge.hess at mdlink.de  Sun Jan  9 19:37:03 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Mon, 10 Jan 2000 01:37:03 +0100
Subject: no return values for __init__ ??
References: <1264659013-22815850@hypernet.com>
Message-ID: <387929AF.AC816307@mdlink.de>

Gordon McMillan wrote:
> Is inheritance a matter of taste?

Strange questions ?!

> You claim to have patched your interpreter. Try this:
> 
> class A:
>   def __init__(self):
>     pass
> 
> class B:
>   def __init__(self):
>     return A()
> 
> class C(B):
>   def __init__(self):
>     self = B.__init__(self)
>   def method(self):
>     print "C.method called"
> 
> c = C()
> c.method()
> 
> Produces an AttributeError, doesn't it?

(First, *no*, it doesn't, since C.__init__ doesn't return a new object,
but only reassigns 'self').

Why should I do the above ? It's a completly constructed example. It's
similiar to this:

class A:     def __init__(self):  pass
class B(A):  def __init__(self):  return self.method()

Produces an AttributeError as well ...

The problem in your example is that you are misusing return values, eg

  class C(B):
    def __init__(self):
      s = B.__init__(self)
      if (s is not None) & (s is not self): return s

will work perfectly.

Helge
-- 
SKYRIX-OS  Web Application Environment - http://www.skyrix.com
SKYRIX-IAS Groupware Application Suite - http://www.skyrix.com



From aahz at netcom.com  Wed Jan 19 01:07:02 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Jan 2000 06:07:02 GMT
Subject: Suggestions for python 2
References: <000501bf623e$18b56a20$9a2d153f@tim>
Message-ID: <863ka6$7ej$1@nntp9.atl.mindspring.net>

In article <000501bf623e$18b56a20$9a2d153f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>[Evan Simpson]
>>
>> I rather like the idea of having to explicitly bring names
>> into an inner scope,
>
>Me too -- except in the cases that bite most often, where a trivial little
>lambda wants to refer to one or two measly little names from its context.
>People stumble on that so often it's a FAQ.

....and there's an easy suggestion for Python 2 that fixes the problem,
right?  ;-)
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From jsc at zoo.co.uk  Fri Jan 21 09:14:04 2000
From: jsc at zoo.co.uk (Jethro Cramp)
Date: Fri, 21 Jan 2000 14:14:04 +0000
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <3dzouke2v7.fsf@amarok.cnri.reston.va.us> <200120002033497688%zippy@cs.brandeis.edu>
Message-ID: <388869AB.7EF93B0E@zoo.co.uk>

I have both books: 'Programming Python' and 'Learning Python'. I value the
'Learning Python' book way and above 'Programming Python'.

There is a lot in 'Programming Python' but it is not ideal for learning the
language and you will have to make a heavy investment in time to get a
decent return from the book. My major dislike with the book is that the
chapter titles are all very humorous, but tell you little about what is in
the chapter.

Jethro Cramp

Patrick Tufts wrote:

> In article <3dzouke2v7.fsf at amarok.cnri.reston.va.us>, Andrew M.
> Kuchling <akuchlin at mems-exchange.org> wrote:
>
> > Bala <engbpa at liquidinformation.com> writes:
> > >         Python Essential Reference (OTHER NEW RIDERS)
> > >         by David M. Beazley, Guido Van Rossum
> > > or
> > >         Learning Python (Help for Programmers)
> > >         by Mark Lutz, David Ascher, Frank Willison (Editor)
> >
> > The Essential Reference is just what its title says: a reference
> > guide.  You'd have to be *really* motivated to learn Python from it.
>
> The Essential Reference is an excellent reference.  Get it anyway.  But
> you won't learn the language from it -- it's too terse for that.
>
> Oh, and don't bother with "Programming Python" (also by Mark Lutz).
> It's not a good for learning or reference.  I've read reviews that say
> "Learning Python" is much better.
>
> --Pat



From bachlmayr at solve-it.de  Tue Jan 25 08:33:21 2000
From: bachlmayr at solve-it.de (Gerald Bachlmayr)
Date: Tue, 25 Jan 2000 14:33:21 +0100
Subject: python <==> ado
Message-ID: <1375BA7ABFD1D3119F840050DA41E448D343@mail2.solve-it.de>

hi!

i would like to connect to an sqlserver 7.0 database with ado.
the database is on a different pc on the network.
does anyone know the command to get a connection?
(i just found odbc etc.)
gerry

-----Urspr?ngliche Nachricht-----
Von: Richard Brodie [mailto:R.Brodie at rl.ac.uk]
Gesendet: Tuesday, January 25, 2000 12:10 PM
An: python-list at python.org
Betreff: Re: tabs do WHAT?



"Thomas Hamelryck" <thamelry at vub.ac.be> wrote in message
news:86joac$l2s$1 at mach.vub.ac.be...

> BTW, people here seem to have adopted the idea that if you don't like
> a particular feature of a language, you should abandon it (e.g., "You
> don't like indentation? Use Perl!").

Adding features to a language always seems easy but that is the best
reason for not doing so. There is the temptation to make the language
the union of all the desired features. It's all been talked around before,
and Guido has said what his feelings are. In the end, I think you've
either got to respect that decision, or fork the language.

> You can find a very nice description of this process with respect to C++
> in "The Evolution of C++ : Language Design in the Marketplace of Ideas",
> by Jim Waldo (Editor), James Waldo (Editor).

C++ has to be the classic case of a language that evolved rather than
was designed: PDP-11 assembler with OO bits tacked on. If we can
learn something from its evolution, it should be not to do the same
again.

'Oh, that way madness lies; let me shun that.' - King Lear


-- 
http://www.python.org/mailman/listinfo/python-list



From wjk at wjk.mv.com  Wed Jan  5 00:35:47 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Wed, 05 Jan 2000 00:35:47 -0500
Subject: is variable a list?
References: <38729120.7616054D@simplypets.com>
Message-ID: <3872D831.84A2BF20@wjk.mv.com>

Hi:
   As a newbie (still am) - I asked similar question:

  There is a function called type()

  so if you had a variable:  x = "This is a string"
  then use type                  type(x)
  you would get back:         <type 'string'>
  or    if type(x) == types.StringType: print "This is a String"
  see also module types
Hopefully this is a start .. and hopefully have not led you
down the wrong path.

cheers



James Punteney wrote:

> Hello,
>
> Is there some way to determine whether a variable is list or a string?
> Basically I have a variable that I would like to be able to be a string
> or a list, but I need some way to tell which it is.
>
> Thanks for the help,
> --James Punteney
>
> jamesp at simplypets.com



From aahz at netcom.com  Tue Jan  4 10:00:21 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 4 Jan 2000 15:00:21 GMT
Subject: String Handling
References: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com>
Message-ID: <84t1u5$7dv$1@nntp8.atl.mindspring.net>

In article <iuv37sktc8d9io2chho7i322vcb0am3k80 at 4ax.com>,
Simon Faulkner  <News at Titanic.co.uk> wrote:
>
>Can anyone give me the Python line/command to change all spaces in a
>string to +s?

string.translate() is faster than string.replace()
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From grant.griffin at iowegian.com  Fri Jan 21 16:32:14 2000
From: grant.griffin at iowegian.com (Grant Griffin)
Date: Fri, 21 Jan 2000 21:32:14 +0000
Subject: horrible MATLAB indexing (from the POV of DSP)
References: <HUrd4.2410$I5.253221@typhoon1.gnilink.net> <85ivj6$1cv$1@mozo.cc.purdue.edu> <387EBB59.3CEC@viconet.com> <3882B39D.734E@viconet.com> <85vuet$vqb$1@news.ccit.arizona.edu> <3885609F.152A@viconet.com> <wkbt6ii7z1.fsf@math.uio.no> <38863B43.7E0F@iowegian.com> <3888621A.230943CE@mitre.org>
Message-ID: <3888D05E.1A44@iowegian.com>

Stan Pawlukiewicz wrote:
> 
...
> The only real annoyance I have with  python is the mechanism for taking
> the real part of a complex array can be somewhat cumbersome.  I spent
> about half a Saturday trying to track down why
> 
>   a=b.real
>   print a.flat
> 
> wouldn't work.  I don't if this a bug, or by design. In Guido's tutorial
> he explicitly says that abs(a) is the only built in conversion.  I
> really have no idea why he thinks that real(a) would be ambiguous.
> 

(cross-posted from comp.dsp to comp.lang.python)

I guess I don't really know.  (Anybody else?)

But I'm curious, Stan: What sorts of DSPish things have you been using
Python for?

=g2
-- 
_____________________________________________________________________

Grant R. Griffin                                       g2 at dspguru.com
Publisher of dspGuru                           http://www.dspguru.com
Iowegian International Corporation	      http://www.iowegian.com


From sposhua at my.pc  Mon Jan 10 04:48:24 2000
From: sposhua at my.pc (Sposhua)
Date: Mon, 10 Jan 2000 09:48:24 +0000
Subject: CGI problem
In-Reply-To: <855k3e$agm$1@nnrp1.deja.com>
References: <855k3e$agm$1@nnrp1.deja.com>
Message-ID: <Pine.WNT.4.21.0001100945530.162-100000@wks_3.e-shopwork.com>

On Fri, 7 Jan 2000, sp00fD wrote:

> I'm creating a CGI program, and I keep getting an error from apache:
> 
> "(8)Exec format error exec of /usr/local/apache/htdocs/...
> Premature end of script headers: /usr/local/apache/htdocs/..
> 
> I've put the print "Content-type: text/html\n" (with anywhere from 0 to
> 8 newlines after it) in there.
> 
> What am I doing wrong?

Either
a) The script has errors
or
b) You have to add 'HTTP:/1.1 200 OK\n' at the begining of the header. Don't ask
me why, but on my server I can't run anything without that.



From tseaver at aftershock.neosoft.com  Tue Jan 11 18:18:16 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 11 Jan 2000 17:18:16 -0600
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov> <85fnbk$j6h$1@nntp9.atl.mindspring.net>
Message-ID: <E812D8E3D435F586.05792F5FF8BB0F78.50ECE477F9A61A79@lp.airnews.net>

In article <85fnbk$j6h$1 at nntp9.atl.mindspring.net>,
Aahz Maruch <aahz at netcom.com> wrote:
>In article <20000111113656.A42270 at vislab.epa.gov>,
>Randall Hopper  <aa8vb at yahoo.com> wrote:
>>Doug Hellmann:
>> |
>> |Could one of you post a brief description of what a multimethod is, or
>> |maybe just a pointer to a definition somewhere?
>>
>>I had the same question yesterday and did a little surfing.  Basically,
>>instead of switching off just the type of 'self' to determine which method
>>to call, you switch off the type of other arguments as well.
>
>IOW, similar to C++-style multiple dispatch?

Not quite, I think -- C++ uses overloading and/or templates to achieve static
(compile-time) selection of a function; Lisps's multi-methods are virtual
(dispatched at runtime), but selected the types of all the arguments, not
just of this->. (In fact, in Lisp, methods aren't declared within classes
at all!)

Only matters if you are inheriting, but it makes something like the Visitor
pattern practically unnecessary.

As an example (stealing from Scott Meyer's C++ World talk on double dispatch),
think of a game where players, NPC's, ships, and other objects are careening
about in zero-gee;  the effect of a collision between two of them depends on
all their types, and not merely the first one the game engine picks up.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From tekhir at yahoo.com  Mon Jan 10 13:18:38 2000
From: tekhir at yahoo.com (Tekhir)
Date: Mon, 10 Jan 2000 13:18:38 -0500
Subject: ODBC
Message-ID: <387A227E.EF84B01E@yahoo.com>

  At work I'm finishing up work on a equipment checkout database.  I've been
using tab delimited files (from MS Access) for all of my work.  But I'm getting
tired of all the work currently required.  So I'm going to make one master ODBC
and do all my work from therer.  Ok now for my questions what files do this?  Do
they come with Python or do I need to d/l them and if so from where?  And are
there any good tutorials out there?

Thanks,
Jeremy


From news at dorb.com  Wed Jan  5 23:52:46 2000
From: news at dorb.com (Darrell)
Date: Wed, 5 Jan 2000 23:52:46 -0500
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com>
Message-ID: <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com>

from filename import *

Doesn't create another namespace. Avoid this in most cases.

Use import filename and filename.someVar the "from" was intended to help
when your running Python interactively.

--
--Darrell
"John Nielsen" <nielsenjf at my-deja.com> wrote in message
news:8515bs$2vn$1 at nnrp1.deja.com...
> Hopefully, there's a simple answer to a problem I am having.
>
> I am using import to include a file, however, it looks like the module
> then has another namespace, which means it can't access some variables.
>
> Is there a way to simply include the raw text and insert it into your
> own code, as if it was not in a another file?
>
> Thanks for any help,
>
> john
>
>
>
> --
> nielsenjf at my-Deja.com
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list




From mal at lemburg.com  Tue Jan 18 17:44:56 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 18 Jan 2000 23:44:56 +0100
Subject: Python and MS SQL (Microsoft SQL server)
References: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in> <3884C60B.BE820782@psychosis.com>
Message-ID: <3884ECE8.DFB83340@lemburg.com>

Dave Cinege wrote:
> 
> "George Thom at s" wrote:
> >
> > Hi Dave,
> > The odbc module that comes as part of the win32 extensions bundled with
> > Python 1.5.2 for the various flavours (!) of Windows should do just fine.
> 
> So you're saying python has to be runnign on NT for this to work?
> 
> > Additional functionality (with some tweaking ... ymmv ;-) can be obtained
> > with marc Lemburg's mxODBC module.
> 
> What about this? we're looking to run on solaris or linux.

Then you should probably try out the Sybase interface together
with the Sybase dev libs. mxODBC will only connect you to
local databases unless you let it talk to the DB through
an ODBC-ODBC bridge.

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





From matthias.huening at univie.ac.at  Thu Jan 13 13:49:56 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Thu, 13 Jan 2000 19:49:56 +0100
Subject: Python Pre-processor (like php/asp)
References: <387D7349.DCF6567A@sprint.ca>
Message-ID: <85l6ua$3v38$1@www.univie.ac.at>

Orlando Vazquez schrieb in Nachricht <387D7349.DCF6567A at sprint.ca>...
>Hi I'm looking for a Python equivalent of php or asp.
>

You don't need an equivalent, you can use ASP. Have a look at:
http://starship.python.net/crew/pirx/asp/py_asp.html

Matthias

- - - - -
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/





From helge.hess at mdlink.de  Mon Jan 10 07:36:53 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Mon, 10 Jan 2000 13:36:53 +0100
Subject: no return values for __init__ ??
References: <1264652208-23225161@hypernet.com>
Message-ID: <3879D265.3440A2AF@mdlink.de>

Hi,

Gordon McMillan wrote:
> Helge Hess wrote:
> > > class C(B):
> > >   def __init__(self):
> > >     self = B.__init__(self)
> > >   def method(self):
> > >     print "C.method called"
> > >
> > > c = C()
> > > c.method()
> > >
> > > Produces an AttributeError, doesn't it?
> >
> > (First, *no*, it doesn't, since C.__init__ doesn't return a new
> > object, but only reassigns 'self').
> 
> And how does "method" become an available to 'self'?

'c' is of class C() because __init__ did no return a new object
(B.__init__ does, but this is thrown away by not returning 'self').

> > Why should I do the above ?
> 
> To prove that returning a value from __init__ is a sane thing to
> do.

I think we are getting into a loop here. You can do insane stuff with
any construct, returning a value from __init__ is no exception of
course.
In practice the above construct won't appear and is a semantic
programming error like my example below.

> > It's similiar to this:
> >
> > class A:     def __init__(self):  pass
> > class B(A):  def __init__(self):  return self.method()
> >
> > Produces an AttributeError as well ...
> 
> Of course it does. You didn't define "method". So yours is a
> programmer error. I did define it.

You did a programming error as well by not returning the correct object.

> And C().method() will produce what result with your interpreter?

AttributeError due to programming error.

> Or by insisting that __init__ is "just another method", have
> you succeeded in producing a special kind of class that looks,
> smells and tastes like "just another" class, except it may
> blow up on you at run time?

If you like you can 'blow up' your runtime, like you can do with any
other factory approaches I was shown. The question is why I should want
to blow up my runtime - doesn't make a lot of sense to me ..

We should really stop discussing the topic now, it is getting weird.

Helge
-- 
SKYRIX-OS  Web Application Environment - http://www.skyrix.com
SKYRIX-IAS Groupware Application Suite - http://www.skyrix.com



From aahz at netcom.com  Sat Jan  8 10:12:54 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jan 2000 15:12:54 GMT
Subject: RegExp, Python and strings
References: <857hhb$3kmi$1@www.univie.ac.at>
Message-ID: <857k5m$q3h$1@nntp6.atl.mindspring.net>

In article <857hhb$3kmi$1 at www.univie.ac.at>,
Matthias Huening <matthias.huening at univie.ac.at> wrote:
>
>In PERL I can do things like this (in one line):
>$A = "Rossum, Guido van; Harms, Daryl; Python, Franz-Josef";
>$A =~ s/([A-Z])[\w]+(?![ \w\-]+,)([ ;-]|$)/$1.$2/g;
>This RegExp results in: "Rossum, G. van; Harms, D.; Python, F.-J."
>
>Now I am trying to rebuild this in PYTHON, but I can't get it to work.
>Any hints? Should I keep trying? Or should I stick to PERL for those
>kinds of string-manipulations?

You should be able to do that with re.sub(), but the syntax is slightly
different.  Don't forget to use raw strings, though (e.g. r'\w').
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From scherbi at bam.com  Tue Jan  4 07:50:07 2000
From: scherbi at bam.com (Bill Scherer)
Date: Tue, 04 Jan 2000 07:50:07 -0500
Subject: Zope: how to search for files?
References: <387113CF.75DE590F@bart.nl>
Message-ID: <3871EC7F.961AB52D@bam.com>

Generally, this question would be better asked on a Zope mailing list.
However, ZCatalog  should be able help you.  If you're using a recent
version of Zope, you already have it.  See the ZCatalog tutorial:

http://www.zope.org/Documentation/How-To/ZCatalogTutorial

Another possibility for you:
You can loop through the contents of a folder with code like the
following:

<dtml-in "(foldername).objectValues(['Folder', 'DTML Document'])">
  <li><a href="<dtml-var id>"><dtml-var title></a>
</dtml-in>

replace (foldername) with the name of your folder.  In the list argument
passed to objectValues, list the types of objects you're interested in.

One of these methods should be able to help.

Good luck,

Bill

Richard Smol wrote:

> Hi there,
>
> I have been fiddling around with Zope for a while and am trying to do a
> simple thing: find out if a certain file-object exists within a folder.
> How do I do this? I don't see any methods that achieve this...
>
> TIA!
>
> RS
>
> --
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000104/bc11733d/attachment.vcf>

From pbible at home.com  Wed Jan 12 07:38:33 2000
From: pbible at home.com (Paul E. Bible)
Date: Wed, 12 Jan 2000 12:38:33 GMT
Subject: Python and Oracle
Message-ID: <dB_e4.1177$hJ1.22100@news.rdc1.tx.home.com>

Has anyone had success installing the Python Oracle Call Interface (OCI)
library written by Digital Creations (DCOracle)?  I'm currently having
compile problems with the C modules, which is a library reference.  I just
curious as to whether this library is worth the effort to resolve the
current issues that I'm having with it's installation.

My platform is:
Python           1.5.2
Oracle           7.3.4
OS                Unixware 7.0.1


Thanks in advance for your help,
Paul




From thor at localhost.localdomain  Sat Jan  8 12:46:58 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 8 Jan 2000 17:46:58 GMT
Subject: Python ring info
Message-ID: <slrn87f1jk.md.thor@localhost.localdomain>

Python ring Info
-------------------------

This is a piece of information about the existing python ring in:
http://www.ctv.es/USERS/irmina/pythonring.html
(now with Statistics)

What is it?
>From a functional point of view, it's a list of linked web sites
that share a common interest or information, in this case python
related stuff.

How does it work?
The HTML code included in each node of the ring calls a CGI that
performs the movement thru the ring.

How can I join?
In the above web site , fill in your data and that's all.

Why I should ? 
If you've done a web site, I think it's worth you spend 15 minutes
promoting it, after your web site has been included in it you'll get
more hits than ever.
Moreover, your information may be interesting for other people, or
you can be interested in other people's work, a big Webring provides
the community a lot of unusual but somehow interesting data.

How can I get more hits? 
Because the Webring is a democratic way of organization, that is,
all its members tend to receive the same ammount of visits, the
big ones as much as the small ones.

Well, I don't have a Web page.., so?
So  you must (  :) ) start doing one right now. Seriously, we
( at least me,MGA) are interested in people's work, perhaps you've
been collecting data or even your libraries or methods of work are
clearly defined, the ammount of time required for a fast compilation
of that info is not too much. And I remark that you can learn only
if all ( or many) want to show or teach what they know.

Of course we admit commercial Web sites also, *any* Web site about
python.

Thanks for your time and patience reading this.
MGA




From doughellmann at home.com  Tue Jan 11 06:45:19 2000
From: doughellmann at home.com (Doug Hellmann)
Date: Tue, 11 Jan 2000 11:45:19 GMT
Subject: More Tkinter questions
References: <3878BF45.7C2A0DA0@exceptionalminds.com>
Message-ID: <387B193F.CD284117@home.com>


Timothy Grant wrote:
> 
> 
> However, once I start adding various widgets to the various frames
> things go south really fast. It appears that I don't get a new grid when
> I start adding objects to a frame, but I work on some sort of master
> grid. When this happens all of the rows and columns shown above start
> losing there relationships.
> 
> Can someone give me a good solution to this?
> 
> BTW: it doesn't help that none of the frames hold a specified size when
> widgets start getting added to them.

That's by design.  The Frames will collapse to just contain their
contents, depending on how you set the sticky values of the contents. 
Gridding within each Frame will use a new grid, not a global one, but
adding a child to the Frame causes the Frame to resize.

What sorts of children are you adding to the frames?  Post some code,
maybe we can help with the layout if we have more specifics.

Doug


From volumeregeling at hotmail.com  Mon Jan 24 05:14:10 2000
From: volumeregeling at hotmail.com (volumeregeling)
Date: Mon, 24 Jan 2000 10:14:10 GMT
Subject: Exiting program
References: <38873769.5048399@122.0.0.250>
Message-ID: <388c259a.489634@122.0.0.250>

Thanks for the answers, I could use them ubt it wasn?t really the
problem. The primary problem is how to pass the command_button object
to another function so i can do command_button.quit from that other
function ??????

Thanks.
Volumeregeling

On Thu, 20 Jan 2000 16:34:52 GMT, volumeregeling at hotmail.com
(volumeregeling) wrote:

>Hi,
>
>I am not very good in searching through the online documentation for
>python. I am trying some things which keep generating errors on
>running.
>
>I have the following program :
>
>I am making a menubar which uses functions from a cupple of COM
>objects. I want to exit my program through a stop_program function not
>from any COM object.
>
>def Exit_program():
>    Editor.mnuFile_Exit()
>    sys.exit()                        <-- for some reason this also
>doesn?t work. but that on problem jet.
>
>Editor in the above is a COM object, with the method from that COM
>object to close it.
>
>When the editor.mnuFile_Exit() method is invoked , python crashes with
>some OLE error. But when I run the COM object from WScript it exits
>fine so the problem is in python.
>
>I hooked the Editor COM object the following way :
>
>Editor= win32com.client.Dispatch("Comeditor.Classeditor")
>
>Can anyone tell me how to "release" the COM object ??
>Editor.release() doesn?t work.
>
>I also can?t seem to get a variable COMMAND_Button to the Exit_program
>procedure. I need to close the menubar with command_button.quit(), ubt
>then it must be invoked from the procedue where command_button is
>created
>
>
>Can anyone tell me how to pass-throug this caommand_button so i can
>invoke command_button.quit() from the Exit_program function.



From scherbi at bam.com  Tue Jan  4 09:38:17 2000
From: scherbi at bam.com (Bill Scherer)
Date: Tue, 04 Jan 2000 09:38:17 -0500
Subject: String Handling
References: <iuv37sktc8d9io2chho7i322vcb0am3k80@4ax.com>
Message-ID: <387205D9.61E40934@bam.com>

>>> import string
>>> replace = string.replace
>>> a = "Lucy by Railway Carraige"
>>> b = replace(a, " ", "+")
>>> b
"Lucy+by+Railway+Carraige"
>>>

- Bill

Simon Faulkner wrote:

> Hi All,
>
> Can anyone give me the Python line/command to change all spaces in a
> string to +s?
>
> ie
>
> "Lucy by Railway Carraige"
>
> becomes
>
> "Lucy+by+Railway+Carraige"
>
> Muchos gratcias etc
>
> Simon
> Simon Faulkner
> --
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000104/1b19f775/attachment.vcf>

From shouldbe at message.com  Thu Jan 20 10:18:51 2000
From: shouldbe at message.com (Arinte)
Date: Thu, 20 Jan 2000 15:18:51 GMT
Subject: A modest Error (I mean Proposal)
References: <XFMail.000120142944.mikael@isy.liu.se>
Message-ID: <uHFh4.1875$L4.54845@newsread2.prod.itd.earthlink.net>

"Mikael Olofsson" <mikael at isy.liu.se> wrote in message
news:XFMail.000120142944.mikael at isy.liu.se...

On 20-Jan-00 Arinte wrote:
 >  I am getting this weird error???
 >  Traceback (innermost last):
 >    File "Beer", line 17, in ?
 >    File "D:\POSSApps\anotherposs app\PossDevice.py", line 37, in IOCTL
 >      if type(command)==type(2):
 >  TypeError: len() of unsized object

Maybe you have accidentally redefined type to len...

I would entertain that thought, but this problem only occurs on the 2nd call
to this function???




From jstok at bluedog.apana.org.au  Thu Jan  6 14:37:28 2000
From: jstok at bluedog.apana.org.au (Jason Stokes)
Date: Thu, 06 Jan 2000 19:37:28 GMT
Subject: Origin of double underscore convention?
Message-ID: <Y96d4.1764$oF2.5206@ozemail.com.au>

I'm relatively new to Python.  This is possibly a dumb question, but
I'm wondering about the double underscore convention to mark an inbuilt
or internal method.  I find it pretty ugly and I'm wondering if it was
chosen for any historical reason, eg analogy to another language.  I
know C programmers sometimes use the underscore convention.

-- 
Jason Stokes: jstok at bluedog.apana.org.au


From david.maslen at iname.com  Thu Jan 27 21:00:03 2000
From: david.maslen at iname.com (David Maslen)
Date: Fri, 28 Jan 2000 02:00:03 GMT
Subject: Charts
References: <3890C74D.786411BC@acm.org>
Message-ID: <87n1pr6m0i.fsf@iname.com>

Thierry Thelliez <thierry at acm.org> writes:
> Is there any packages for creating charts (histograms, scattered plots,
> line charts,...) in Python ?

Yes. There seem to be a few. Do a search on
http://www.vex.net/parnassus/


Just to quickly summarise what I discovered.

Gnuplot, stable, mature and the python module to access is quite good,
but not as feature packed as it might be. The best thing I found.

gdcharts. Very pretty. I'd consider it for generating nice graphs for
a webpage or perhaps a report, but not if I really wanted to look at
the lines in detail.

graphite. Based on the piddle library. I think this will be very nice,
but it didn't do the financial type charts I was after.

gist stuff in python_graphics. Interesting, and interactive, not quite
what I was after. 


From jjlucsy at concentric.net  Fri Jan 28 10:39:30 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Fri, 28 Jan 2000 10:39:30 -0500
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com> <MPG.12fa1256cc30caf0989682@or.news.verio.net> <8EC991745duncanrcpcouk@194.238.50.13>
Message-ID: <zJik4.2042$CX6.158724@news4.usenetserver.com>

> I find it strange that you find this long delay, but others (myself
> included) find the standard DLL loads quickly every time (possibly a bit
> slower the first time). I can think of a few possible reasons for this.
> Perhaps the python DLL loads other DLLs which take time to load, and which
> are preloaded into memory on my system, but not on yours. The python
> dependencies are KERNEL32, USER32, ADVAPI32, WSOCK32, WINMM, MSVCRT, most
> of which would be loaded on any booted windows system. WSOCK32 might be
> unloaded if you do not normally have a network connection, in which case
> running up a network application before running python should reduce the
> wait.

Only if the dlls are loaded at the same base address.

> The other possibility that occurs to me might be a conflicting DLL already
> loaded in the system at the same address. This would mean that the python
> dll would have to be relocated every time it loads, however the delayed
> loading DLL shouldn't be faster in this case unless it has a different
> relocation base.

On my system and another I've talked to with email, it's the winmm that
causes the problem. Most likely it's caused by improperly written sound
drivers. I have a laptop which probably exacerbates the problem.

As I said in another post, delayload is probably not for everyone, or every
project. I'm embedding python in a particular plug-in which works within a
huge memory hog of an application. I definately noticed the delay in loading
the standard python.dll. Now that I've rebuilt python15.dll with delayload
for my project, it loads darn quick. I don't expect to use any functions
within the delayloaded modules, so all the better for me.

-
Joel Lucsy (jjlucsy at concentric.net)







From aa8vb at yahoo.com  Wed Jan  5 08:27:04 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 5 Jan 2000 08:27:04 -0500
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
In-Reply-To: <xc1BqAAkwpb4EwWW@jessikat.demon.co.uk>
References: <841d0c$j0n$1@violet.singnet.com.sg> <387f6d37.25522789@news.demon.co.uk> <19991230065516.A778039@vislab.epa.gov> <W8TkoEATH4a4Ewpu@jessikat.demon.co.uk> <387378c0.4549271@news.demon.co.uk> <xc1BqAAkwpb4EwWW@jessikat.demon.co.uk>
Message-ID: <20000105082703.A9441@vislab.epa.gov>

Robin Becker:
 |Is there an obvious way to allow an extension module to write messages to
 |various fd type objects?

Do you mean, change a module that's written to dump to stdout or stderr so
that it writes to objects?  You can always replace sys.stdout/stderr
before/after calling the extension module.  Not pretty, but it works.

 |Also under certain extreme circumstances my code would like to report a
 |total failure eg attempt to divide by error. I would like to ask more
 |experienced pythoneers how arithmetic errors are handled using built in
 |IEEE exceptions and how does one gracefully crash a python script? (eg
 |an abend/abort).

raise ArithmeticError, "Oops"
raise ZeroDivisionError, "Ouch"

?

-- 
Randall Hopper
aa8vb at yahoo.com



From gerrit.holl at pobox.com  Sat Jan 22 09:48:11 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 22 Jan 2000 15:48:11 +0100
Subject: 7/8 bit email (Was: Re: OT: Uitgelijnde email verzenden)
In-Reply-To: <3888B8E4.77B50FA2@kub.nl>; from paai@kub.nl on Fri, Jan 21, 2000 at 08:52:04PM +0100
References: <001101bf6445$64d51370$43c7a8c0@tentacleloco.nl> <3888B8E4.77B50FA2@kub.nl>
Message-ID: <20000122154811.A6625@stopcontact.palga.uucp>

Hans Paijmans wrote on 948484324:
> Nee, dank je. Wat niet in zeven bits ASCII geschreven kan worden 
> is niet waard om gelezen te worden. Je mag me citeren.

Wat is er mis met 8 bits? Zijn er nog terminals die geen 8 bits
supporten? Linus heeft 'es iemand die dit zei geflamed. Vooral
voor een aantal Europese talen is dat niet goed te doen.

groeten,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From dcinege at psychosis.com  Wed Jan 19 10:58:04 2000
From: dcinege at psychosis.com (Dave Cinege)
Date: Wed, 19 Jan 2000 10:58:04 -0500
Subject: Python and MS SQL (Microsoft SQL server)
References: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in> <3884C60B.BE820782@psychosis.com> <3884ECE8.DFB83340@lemburg.com>
Message-ID: <3885DF0C.819CB0FB@psychosis.com>

"M.-A. Lemburg" wrote:
> 
> Dave Cinege wrote:
> >
> > "George Thom at s" wrote:
> > >
> > > Hi Dave,
> > > The odbc module that comes as part of the win32 extensions bundled with
> > > Python 1.5.2 for the various flavours (!) of Windows should do just fine.
> >
> > So you're saying python has to be runnign on NT for this to work?
> >
> > > Additional functionality (with some tweaking ... ymmv ;-) can be obtained
> > > with marc Lemburg's mxODBC module.
> >
> > What about this? we're looking to run on solaris or linux.
> 
> Then you should probably try out the Sybase interface together
> with the Sybase dev libs. 

So you're saying the sybase interface works with 

mxODBC will only connect you to
> local databases unless you let it talk to the DB through
> an ODBC-ODBC bridge.

That sounds like a better solution. Can I do that using the unix ODBC thing
that I've seen listing in the python modules page?



From cjc26 at nospam.cornell.edu  Thu Jan 27 18:28:59 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: 27 Jan 2000 23:28:59 GMT
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de> <869uel$suj$1@nntp1.atl.mindspring.net> <slrn88hnpb.brp.quinn@auto.ugcs.caltech.edu> <slrn88n40a.2qh6.cjc26@bridget.mindriot.net> <948708715.22839@news.adtranzsig.de>
Message-ID: <slrn891l2f.5tk.cjc26@bridget.mindriot.net>

Pada Mon, 24 Jan 2000 10:09:21 +0100, Dirk-Ulrich Heise bilang:
| >
| >My semantics prof always made us call currying "Schoenfinkelization",
| >because "Schoenfinkel came up with it first".
| >Do you know how hard it is to spell Schoenfinkelization? ;)
| 
| [..]
| And does the word "closure" play an important role outside
| of the functional programming world?

I'm assuming you mean outside of programming in general, because I
mentioned semantics above.  Well, traditionally semanticists who try
to represent the meaning of a sentence such as "John kicked the ball"
would say that "kicked" represents a two-place predicate (IOW, a
function that takes two arguments).  So it would be written in
predicate logic as:

	kicked(John, the ball)

However, there is evidence that "kicked the ball" forms a seperate
phrase or unit in the sentence (for example, you can say "John
kicked the ball, and Mary did too" where the word "did" is sort of
like a variable which references "kicked the ball").  So then you
might want to represent the sentence as:

	kicked_the_ball(John)

or even:

	(kicked(the ball))(John)

In that last one, "kicked" represents a curried (or
Schoenfinkelized :) function; it is a function which takes a noun
phrase X as its argument and which returns another function which
takes a noun phrase Y as its argument and returns true if Y
kicked X.  I.e.:

	kicked = lambda(X) (lambda(Y) (if Y kicked X: 1
	                               else: 0))

This would mean that kicked(the ball) would represent a closure,
where the variable X is bound to the entity "the ball".

Semantics is a lot like programming, actually. :)


| (When you have curried a lot, this would probably mean in
| german: "Heute hab ich viel geschoenfinkelt." Try that!)

but-I-can't-even-say-it-<wink>-ly yr's,  Cliff


-- 
cliff crawford  -><-  http://www.people.cornell.edu/pages/cjc26/
                     "All I need is a small town
                      with good Internet connectivity" - Nadine


From tim_one at email.msn.com  Mon Jan 31 20:36:34 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 31 Jan 2000 20:36:34 -0500
Subject: Dynamic class construction?
In-Reply-To: <3895B829.39067306@math.okstate.edu>
Message-ID: <000601bf6c54$c6418aa0$d709143f@tim>

[Adrian Eyre]
>> [...]
>>
>> There's already a module which does this...
>>
>> import new
>>
>> def anOp(self):
>>         return 'This is an instance of the class %s' % \
>>                 self.__class__.__name__
>>
>> someOps={'__str__':anOp}
>> G=new.classobj('G', (Group,), someOps) # Group defined elsewhere

[David C. Ullrich]
>     I sorta thought there might be a module that does this, but I
> couldn't find it.
>
>     Um, I still can't find it. It appears to be a builtin module -
> new.__doc__ says
>
> '     Functions to create new objects used by the interpreter.
> You need to know a great deal about the interpreter to use this!',
>
> which is helpful, I guess, but I can't find any more about it in
> the docs (or I'm not looking in the right place? It's not in the
> "module index", I don't find any interesting files contaning
> the text "classobj", etc.)

The "new" module is deliberately undocumented.  As the docstring says, you
need to understand the internals of Python inside out to use it safely; but
if you do know the internals sufficiently well to use "new" at all, you
don't need any docs.  The module is not safe for general use (it provides
unchecked access to the internals, and it's very easy to crash the
interpreter by misusing it, and it's impossible not to misuse it if you
don't know the internals ...).

I'm not sure what the problem here is (haven't followed this thread), but
"new" isn't the answer <0.9 wink>.

If that doesn't dissuade you, study Modules/newmodule.c in the source
distribution.

the-code-*is*-the-docs-in-this-case-ly y'rs  - tim





From hnowak at cuci.nl  Mon Jan 24 16:13:02 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Mon, 24 Jan 2000 22:13:02 +0100
Subject: Dictionary from a file
In-Reply-To: <388BF0C4.F0E08005@mail.bulkley.net>
Message-ID: <200001242113.WAA11012@dionysus.fw.cuci.nl>

On 23 Jan 00, at 22:27, Peter Schopfer (aka Dr. J. -  wrote:

> Hello
> 
> I have a dictionary that is too big to fit right in my program, so i was
> going to have it as a seperat file, labeled dic.py. how would i be able to
> retrieve this dictionary. wuld this work:
> 
> from dic.py import dictionary
> mydic=dic.dictionary
> 
> i am a newbie, so any help would be appreciated
> thanks

See my snippets page (http://tor.dhs.org/~zephyrfalcon/snippets), snippet 
#169, for an example. 

I uploaded some new stuff yesterday and I'm not sure yet if everything 
works as it should, so I'll copy & paste the example for you:

---begin code-----

# writeabledict.py
"""A writeable dictionary."""

from UserDict import UserDict

class WriteableDict(UserDict):

    def __init__(self, other=None):
        """Provides copy-constructor."""
        UserDict.__init__(self)
        if type(other) == type({}):
            self.data = other.copy()
        elif isinstance(other, UserDict):
            self.data = other.data.copy()

    def write(self, filename, dictname="data"):
        if not filename[-3:] == ".py":
            filename = filename + ".py" # duh!
        f = open(filename, "w")
        f.write("# %s generated by writeabledict.py\n\n" % filename)
        f.write("%s = {\n" % dictname)
        for key in self.data.keys():
            f.write("\t%s: %s,\n" % (repr(key), repr(self.data[key])))
        f.write("}\n")
        f.close()


if __name__ == "__main__":

    telnr = WriteableDict()
    telnr['hans'] = 4540
    telnr['huub'] = 4774
    telnr['uszo'] = [telnr['huub'], telnr['hans']]
    telnr['misc'] = {'fred': 3456, 'jaap': 9233, 'rik': 4521}
    print telnr
    telnr.write("telnrdict.py")

    import telnrdict
    print telnrdict.data

---end code---

The WriteableDict class writes a dictionary to a Python file, which can be 
imported, as you can see in the last two lines.

Alternatively, you could take repr(dict), which is a string, and write it 
to a file. You can read it again by reading the string and using eval() on 
it. (Didn't test this though)

Hope this helps,


--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
You call me a masterless man. You are wrong. I am my own master.



From skip at mojam.com  Tue Jan 11 10:47:56 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 11 Jan 2000 09:47:56 -0600 (CST)
Subject: Newbie simple question - can't find answer
In-Reply-To: <85fhha$6js$1@nnrp1.deja.com>
References: <85fhha$6js$1@nnrp1.deja.com>
Message-ID: <14459.20652.930067.750483@beluga.mojam.com>

    Rachel> For example if I type:
    >>>> import math
    >>>> exp (4)

Try this instead:

    import math
    math.exp(4)

Import brings one or more names into the current namespace.  The simple form
you used above associates a name, "math", with the math module object.  If
you had executed

    from math import exp

or

    from math import *

you would have had a variable called "exp" associated with the math.exp
function in your local namespace and could have executed a bare exp(4).

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From clemens.wyss at gmx.ch  Fri Jan 28 08:48:07 2000
From: clemens.wyss at gmx.ch (Clemens Wyss)
Date: Fri, 28 Jan 2000 14:48:07 +0100
Subject: PythonWin 2.0?
References: <388f3957_1@spamkiller.newsfeeds.com>
Message-ID: <38919E17.741BC63D@gmx.ch>

See
http://starship.python.net/crew/mhammond/registration/RegisteredUsers.html
    and then register at
http://starship.python.net/crew/mhammond/registration/
or wait for another 3 months, when 2.0 will be released to "the rest".

Clemens

David Max wrote:

> Does anybody know where I can download PythonWin 2.0? The latest I seem to
> be able to find is build 128, yet I saw version 2.0 demonstrated on Monday
> at the Python conference.
>
> Thanks,
>
> - David



From hgg9140 at skewer.ca.boeing.com  Fri Jan 14 09:42:09 2000
From: hgg9140 at skewer.ca.boeing.com (Harry G. George)
Date: Fri, 14 Jan 2000 14:42:09 GMT
Subject: XML-RPC Server
References: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de>
	<slrn87ne3i.qla.kc5tja@garnet.armored.net>
Message-ID: <xqxr9fkk94u.fsf@skewer.ca.boeing.com>

kc5tja at garnet.armored.net (Samuel A. Falvo II) writes:

[snip]
> Indeed -- I too would like more information on this.  I checked out
> www.develop.com, since they seem to be the biggest mover and shaker in the
> SOAP world, but I haven't found anything just yet.  Maybe I'm just not
> looking in the right areas?

What is the python community's position re SOAP and XML-RPC?  As I
understand it, XML-RPC is standards-base, platform-independent, while
SOAP uses MS-sponsored extensions (including, e.g., XML Data).  Can I
assume that *both* will be supported in the future?

> 
> -- 
> KC5TJA/6, DM13, QRP-L #1447
> Samuel A. Falvo II
> Oceanside, CA

-- 
Harry George                E-mail:  harry.g.george at boeing.com 
The Boeing Company          Renton:  (425) 237-6915 
P. O. Box 3707  OY-89       Everett: (425) 266-3149 
Seattle, WA 98124-2207      Page:    (425) 631-8803



From andres at corrada.com  Fri Jan  7 15:16:30 2000
From: andres at corrada.com (Andres Corrada)
Date: Fri, 07 Jan 2000 15:16:30 -0500
Subject: CGI problem
References: <855k3e$agm$1@nnrp1.deja.com>
Message-ID: <3876499E.E7D7E023@corrada.com>

Any error in your script could be causing the problem. Retreat to simple
scripts!

------------------------------------------------------
Andres Corrada-Emmanuel   Email: andres at corrada.com
Owner                     http://www.corrada.com/mamey
Mamey                     Phone: (413) 587-9595
------------------------------------------------------



From hjwidmai at foxboro.com  Wed Jan 19 09:21:27 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Wed, 19 Jan 2000 15:21:27 +0100
Subject: Python Rocks!
In-Reply-To: <8631qm$26p@news.or.intel.com>
Message-ID: <20000119152127.A6919@foxboro.com>

"tye4" wrote:
> I'm more concerned with how many maintenance headaches will result if I were
> to accidently add or delete a space on a line. Also, I would to be very
> careful and choosy about what editor to use.. Some editors expand tabs while
> others don't.. What indentation you see in one editor is sometimes different
> to what is seen in another...

Accidently add or delete a space on a line? To me, it looks like you're looking
at it the wrong way: The whitespace in front of the line isn't "spaces" and "tabs",
it's *indentation*. And it's added und deleted not by hitting <BackSpace> or <Del>
4 times but by an editor command "Indent" or "Dedent".
Yes, I'm mixing tabs and spaces, without any problems. I keep tabs at 8, since
every tool I've come across defaults to that. But my editor (vim) distinguishes
between tabs and indentation and keeps me happy all the time. In the rare case
that somebody wants my code, it's easy to filter it through expand.

Every sane programmer indents his or her programs anyway. And all the C programmers
I know were bitten by this old example:

  if (a > 3)
    if (b == 4)
       c = 2;
  else
    c = 3;

Indentation is a way to show the structure of a program to the human reader.
Why do we need a second way to show it to the compiler/interpreter? This only
makes the code ambiguos. What was the programmer's intention in the example
above?
I see the structuring by indentation as one of python's strongest points.

-- 
Hans-Joachim Widmaier         E-Mail:  hjwidmai at foxboro.com
Foxboro Eckardt GmbH          Telefon: +49 711 502 - 397
Pragstrasse 82                Fax:     +49 711 502 - 300
D-70376 Stuttgart 



From tlu at bmo.com  Thu Jan 27 10:48:53 2000
From: tlu at bmo.com (Tao Lu)
Date: Thu, 27 Jan 2000 10:48:53 -0500 (EST)
Subject: lib reference in Unix man format
Message-ID: <200001271548.KAA04215@trm.bmo.com>

Where can I find the lib reference in Unix man format? Forgive me if this has 
been asked before.

Have been programming in Perl for years, I found those perlfunc/perlop/perlfaq 
are very convenient. Any similar work done for Python developers/newbies?

Thanks,
--
Tao





From scherbi at bam.com  Mon Jan 10 09:00:09 2000
From: scherbi at bam.com (Bill Scherer)
Date: Mon, 10 Jan 2000 09:00:09 -0500
Subject: Zope and NT databases
References: <387956BD.CFDF17DD@bby.com.au> <3879CE9B.BE77787C@bam.com> <3879E4E4.B4BD96FD@ipro.de>
Message-ID: <3879E5E9.AEA3CC6B@bam.com>

oops, I missed the "on Unix" part....

Jochen Knuth wrote:

> Hi,
>
> Bill Scherer wrote:
> >
> > You want ZODBCDA:
> >
> > http://www.zope.org/Products/DA/ZODBCDA
> >
> > Sarah Burke wrote:
> >
> > > Hello...
> > > I want to connnect up to a NT Access 97 Database through Zope running on
> > > Unix. Is the mxODBC module in an External Method the way to go?
> > > Thanks in advance
> > >
>
> if i remember correctly ZODBCDA only works on Win32 platform.
>
> also Access 97 is not a client/server database, so you cant connect to
> it from unix directly to it.
>
> perhaps the ODBC Socket Server from http://odbc.linuxbox.com/index.html
> will work (haven't tried it).
>
> You have to write a python client for the ODBC Socket Server (if noone
> else has written it, don't know).
>
> Ciao,
> Jochen
> --
> --------------------------------------------------
> Jochen Knuth          WebMaster http://www.ipro.de
> IPRO GmbH             Phone ++49-7152-93330
> Steinbeisstr. 6       Fax ++49-7152-933340
> 71229 Leonberg        EMail: J.Knuth at ipro.de
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Bell Atlantic Mobile
914 365 7265


-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000110/5e80afdb/attachment.vcf>

From dheise at debitel.net  Sat Jan 29 21:48:56 2000
From: dheise at debitel.net (Dirk Heise)
Date: 29 Jan 2000 21:48:56 MEST
Subject: Case Sensitivity and Learnability
References: <B4191059635E81B1.8AFDD386EFACF4A3.6FF161EA0151B5EB@lp.airnews.net> <86u4uf$26lq$1@thoth.cts.com>
Message-ID: <01bf6a99$fafba6c0$LocalHost@dreadzone>


Will Rose <cwr at crash.cts.com> schrieb im Beitrag <86u4uf$26lq$1 at thoth.cts.com>...
[...]
> Your arguments are reasonable, but they are arguments against systems
> which are both case-insensitive _and_ case-preserving.  The trouble with
> CI/CP is that the compiler sees a different set of tokens from those seen
> by the programmer, and it's sometimes difficult for the programmer to
> realise this.  If unskilled programmers find case too difficult or
> distracting, the cure is to force a single case, either upper or lower. 
> This could be done by an (optional?) compiler switch.

PLEASE DON'T THINK I'M SHOUTING. I'M JUST NOSTALGIC AND
I WANT MY C64 BACK.

DIRK.

("TEACHER, WHY IS MY PROGRAM ALL BIG LETTERS?"
"YOU'D BE TOO DUMB TOO READ IT OTHERWISE.")


From scherbi at bam.com  Wed Jan 12 07:56:26 2000
From: scherbi at bam.com (Bill Scherer)
Date: Wed, 12 Jan 2000 07:56:26 -0500
Subject: converting tuple to a list
References: <tHZ8OL+ftbNGsmad91aonRmEC1nh@4ax.com>
Message-ID: <387C79FA.FE7E8830@bam.com>

Not really.  Try the list function:

>>> t = (1,2,3,4)
>>> l = list(t)
>>> l
[1, 2, 3, 4]
>>>

Bill Scherer

Anders M Eriksson wrote:

> Hello!
>
> Is this the 'best' way to convert a tuple to a list?
>
> t = (1,2,3,4,5)
> l = []
>
> for i in range(len(t)):
>         l.append(t[i])
>
> // Anders
>
> --
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000112/97825d6f/attachment.vcf>

From bjorn at roguewave.com  Tue Jan 25 14:41:12 2000
From: bjorn at roguewave.com (bjorn)
Date: Tue, 25 Jan 2000 12:41:12 -0700
Subject: Python compiler?
References: <20000123092040.A2010@stopcontact.palga.uucp> <000401bf661e$da1c4100$902d153f@tim> <20000125163234.C2497@stopcontact.palga.uucp>
Message-ID: <388DFC57.30C80573@roguewave.com>

> [...]

> > This is probably not what J
> > Dubal had in mind, but that's life.  A Python VM is needed to execute Python
> > bytecodes just as a Java VM is needed to execute Java bytecodes, and that's
> > what "the interpreter" is.
>
> But isn't it possible to write a *compiler* for Python?

Sure, but "nobody"[*] seems to think it will be smaller/faster without introducing
limits on dynamic behavior and/or types...

-- bjorn

[*] although Viper can apparently do some special caseing by performing a
whole-program analysis...




From pinard at iro.umontreal.ca  Thu Jan 13 15:05:46 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 13 Jan 2000 15:05:46 -0500
Subject: International Python
Message-ID: <oqk8ldg2jp.fsf@titan.progiciels-bpi.ca>

Hi, people.

Things are improving in the area of Python internationalisation! :-) After
having made PO mode able to mark strings for translation in a collection
of Python sources, I finally got `xpot' to extract those strings into a
neater POT file.  I'm grateful to Barry Warsaw for his `pygettext' tool,
which has been useful to cross-check the results of 'xpot' and debug it.

`xpot' uses the same principles that guided PO mode scanning for Python.
Adjacent strings (those only separated by whitespace or comments)
are correctly concatenated at extraction time.  Strings are considered
translatable if they are used within _(STRING) or gettext(STRING) constructs,
other keywords may be added, of course.  Also, to palliate the lack of a
pre-processor, strings which translation should be delayed may be marked
as translatable by using one of ''"text", ''r"text", ""'text' or ""r'text'.
I might address ''"""text""" and ""'''text''' as well if deemed useful.

One option to `xpot' allows for automatic extraction of all doc strings,
which never need explicit marking for this reason.  I think I got it right:
doc strings should be correctly found even after very complex initialisation
of keyword parameters.  I do not expect this to be common use, but this
might be useful in highly interactive programs giving access to Python
internal code, a bit like Emacs already does for its own LISP.

`xpot' is also able to extract translatable strings from Emacs LISP files,
as well as from other pre-existing PO files.  To find out the language,
it looks for hints in file name, in `#!PATH/env PROGRAM', `#!PATH/PROGRAM',
`-*-mode:MODE-*-' or `-*-MODE-*-' in the first two lines.  I did not recheck
Emacs LISP or PO recently, but found out on the whole Emacs, that `xpot'
fails on raw-Mule files.  My plan is to complete `xpot' so it can also handle
C/C++ (`xgettext' makes this one less urgent), and later `awk', Perl and
`bash', or in fact, anything that could help internationalisation! :-)

`xpot' is itself written in C, Flex and Bison, and I would expect it to be
rather fast even for big projects.  I intend to put out an alpha version
for pretest, probably soon, and then tell you where to find it.

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





From breiter at usf.Uni-Osnabrueck.DE  Fri Jan 14 05:43:09 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 14 Jan 2000 10:43:09 GMT
Subject: Python and MapObjects (Was: Python and ArcView GIS)
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>
    <Pine.SGI.3.95.1000112172541.19328B-100000@clavey.des.ucdavis.edu>
    <85kci6$i9l$1@newsie2.cent.net>
    <85kq6t$4a4$1@newsserver.rrzn.uni-hannover.de>
    <85l4p2$s4h$1@newsie2.cent.net>
Message-ID: <85mujt$7n4$2@newsserver.rrzn.uni-hannover.de>

In article <85l4p2$s4h$1 at newsie2.cent.net>,
	"Norman Vine" <nhv at cape.com> writes:
> Bernhard Reiter wrote

>> > http://mapserver.gis.umn.edu/
>> Mapserver still has to clarify its license (though they are aiming
>> towards BSD style, AFAIK.)

> 
> Here is the license from the current mapserver distribution
> seems to be Python compatable to me

> Copyright (c) 1996-1999 Regents of the University of Minnesota.
> 
> Permission is hereby granted, free of charge, to any person
> obtaining a copy of this software and associated documentation
> files (the "Software"), to deal in the Software without
> restriction, including without limitation the rights to use, copy,
> modify, merge, publish, distribute, sublicense, and/or sell copies

Yes, you are right this is in the README of the 
MapServer 3.3.003 (beta version). It must have been released a
couple of days ago. Thanks for pointing it out, we will adapt the
freegis.org pages.

	Bernhard

-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From gerrit.holl at pobox.com  Fri Jan 28 07:40:10 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 28 Jan 2000 13:40:10 +0100
Subject: nesting class definitions?
In-Reply-To: <slrn891nqh.bt0.neelk@brick.cswv.com>; from neelk@brick.cswv.com on Fri, Jan 28, 2000 at 12:13:57AM +0000
References: <20000127155551.A2069@stopcontact.palga.uucp> <slrn891nqh.bt0.neelk@brick.cswv.com>
Message-ID: <20000128134010.C1005@stopcontact.palga.uucp>

Neel Krishnaswami wrote on 949014837:
> Gerrit Holl <gerrit.holl at pobox.com> wrote:
> > Hello,
> > 
> > I do see the point in nesting function definitions, namely as an alternative
> > on places where one could use lambda. But what's the point in nesting
> > class definitions? I know it's possible, but I've never seen it used.
> 
> I have used it, though for several reasons it's definitely not a
> standard idiom.

[why not use them]

[why use them]

> def class_factory(*mixins):
>     class Anonymous:
>         # some various routines
>     Anonymous.__bases__ = tuple(mixins)
>     return Anonymous
> 
> so that I could call class_factory() to create the appropriate class
> at runtime.

Ah, but what I meant was:

class Jan:
    class Piet:
        ...
    class Henk:
        ...

Is there any point in using this feature?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From mikael at isy.liu.se  Tue Jan 11 03:13:02 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 11 Jan 2000 09:13:02 +0100 (MET)
Subject: starting a cgi with params
In-Reply-To: <387a686e.3028554@nntpserver.swip.net>
Message-ID: <XFMail.000111091302.mikael@isy.liu.se>

On 10-Jan-00 Anders Eriksson wrote:
 >  To make everything easy. The developing platform is WinNT+Xitami and
 >  the target platform (where the script will run) is Solaris + Apache

Try importing os and check out what you've got in the dictionary 
os.environ. I'm using Apache and Solaris, and I've got the argument in
os.environ['QUERY_STRING'].

Lycka till!

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    11-Jan-00
Time:    09:07:35

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From schiotz1 at yahoo.com  Thu Jan  6 11:57:08 2000
From: schiotz1 at yahoo.com (Jakob Schiotz)
Date: Thu, 6 Jan 2000 08:57:08 -0800 (PST)
Subject: Changes to UserList.py
Message-ID: <20000106165708.20071.qmail@web3002.mail.yahoo.com>



--- Yesterday, I <schiotz1 at yahoo.com> wrote:
> Dear Python Gurus,
> 
> Apparently, classes derived from UserList can be initialized with
> real
> lists and with instances themselves derived from UserList but not
> with
> other list-like objects (i.e. objects defining the methods you would
> expect of a list).

The solution is probably to fix UserList.py.  I have attached a fix I
have submitted to GvR.  Here is a short description:

The problem
was that UserLists could not be initialized from any sequence, only
from true lists and other UserLists.  Furthermore, __init__
incorrectly assumed that anything not a true list is a UserList,
causing a confusing AttributeError.  __setslice__, __add__ and
__radd__ are more tolerant, they test if their argument is a
UserList or a true list, and if not they attempt to convert it to a
list with list(s).  The extend method, however, had the same problem
as __init__.

__init__ was changed from 

    def __init__(self, list=None):
        self.data = []
        if list is not None:
            if type(list) == type(self.data):
                self.data[:] = list
            else:
                self.data[:] = list.data[:]

to

    def __init__(self, lst=None):
        self.data = []
        if lst is not None:
            if isinstance(lst, UserList):
                self.data[:] = lst.data[:]
            elif type(lst) == type(self.data):
                self.data[:] = lst
            else:
                self.data = list(lst)

extend was changes similarly.

> [ ... ] and then
> the build-in function list() should be extended to work on any object
> having a __getitem__ method.  The latter change should probably be
> done
> anyway in a later version of Python.

Ooops, I managed to fool myself into thinking that list(x) didn't work,
when the problem was that I used it in __init__ above, but had kept the
argument name as "list" thus hiding the function.  



 Best regards,
 
 Jakob Schiotz
 



=====
Jakob Schiotz, CAMP and Department of Physics, Tech. Univ. of Denmark,
DK-2800 Lyngby, Denmark.  http://www.fysik.dtu.dk/~schiotz/
This email address is used for newsgroups and mailing lists
(spam protection).  Official email: schiotz @ fysik . dtu . dk
When spammed too much, I'll move on to schiotz2 at yahoo.com
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: UserList.diff
Type: application/x-unknown
Size: 1326 bytes
Desc: UserList.diff
URL: <http://mail.python.org/pipermail/python-list/attachments/20000106/ede08438/attachment.bin>

From AKiss at GEOMETRIA.hu  Tue Jan 18 05:49:01 2000
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Tue, 18 Jan 2000 11:49:01 +0100
Subject: Python code for documenting Python classes?
Message-ID: <D148AB3A9D8FD311A99A008048C79D1104B6AE@MAIL>

Hi,
Someone at starship.python.net(i don't remember his name) has a nice tool
for documenting Python sources.
Arpad Kiss

Michael Hudson <mwh21 at cam.ac.uk> wrote in message
news:<m3g0vvadwx.fsf at atrus.jesus.cam.ac.uk>...
> T. Bridgman <wtbridgman at radix.net> writes:
> 
> > Just so I don't try to re-invent the wheel...
> > 
> > Does anyone know of a Python program for documenting Python classes?
> 
> No, but I've whipped something up. The script at the end of this post
> scans through the methods of a class and collects a list of attributes
> that are assigned to self.
> 
> E.g:
> 
> >>> class A:
> 	def __init__(self,a):
> 		self.b = a
> 		self.c = 1
> 		self.e = "orange"
> 	def meth2(a,b): # even works if you don't call it self!
> 		a.confuse = b
> 
> 		
> >>> classdoc.do_class(A)
> ['b', 'c', 'confuse', 'e']
> 
> It uses bytecodehacks, so get that here: (where I've just uploaded a
> new version with a couple fixes)
> 
> http://starship.python.net/crew/mwh/
> 
> (it could be written just using dis, but not as quickly) 
> 
> It only does the class in question; it could be trivially modified to
> recurse into base classes (but only Python ones, I'm afraid).
> 
> This is a fairly maximal set of data members; you could just scan
> __init__ for a more minimal set. Trying to work out which members will
> always be set is a job for someone else (maybe someone who can solve
> the halting problem...).
> 
> It also makes no attempt to determine the type of member (that would
> be *hard*), or format the output (this should be easy).
> 
> enough-caveats-for-one-post-ly y'rs
> Michael
> 
> --- this is classdoc.py:
> 
> from bytecodehacks.code_editor import InstanceMethod
> from bytecodehacks.ops import *
> import types
> 
> def do_method(meth):
>     code = InstanceMethod(meth).im_func.func_code
>     ops = code.co_code
>     varnames = code.co_varnames
>     names = code.co_names
> 
>     ans = []
>     
>     for i in range(len(ops)):
>         op = ops[i]
>         if op.__class__ is LOAD_FAST:
>             if op.arg == 0: # self is the first posarg
>                 if ops[i+1].__class__ is STORE_ATTR:
>                     name = names[ops[i+1].arg] 
>                     if name not in ans:
>                         ans.append(name)
> 
>     return ans
> 
> def do_class(klass):
>     ans = []
>     for m in dir(klass):
>         a = getattr(klass,m)
>         if type(a) is types.MethodType:
>             for i in do_method(a):
>                 if i not in ans:
>                     ans.append( i )
> 
>     ans.sort()
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From News at Titanic.co.uk  Sun Jan  2 11:54:50 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Sun, 02 Jan 2000 16:54:50 GMT
Subject: MySQL
Message-ID: <pj0v6sg2d06cjuih0ofjhnoqjm5lp6q8lr@4ax.com>

Confused - I am!

I have written a lovely little database stored in MySQL that has a
Python front end and uses MySQLdb through _mysql on SuSE with MySQL
3.27 - Fine.  All on my Sony laptop development machine.

Now I am trying to put it onto my RH6 production server I try to build
the MySQLdb-0.1.1 and it reports that it needs _mysql.  Where do I get
this from?

If it is part of MySQL then where is it?  I have the client, server,
benchmarks and devel loaded.

Is there a simple overall explanation of databases in Python anywhere?

Sorry to whine :-)

TIA

Simon
Simon Faulkner


From adh at ice.ucdavis.edu  Wed Jan 12 20:42:07 2000
From: adh at ice.ucdavis.edu (Allan Hollander)
Date: Wed, 12 Jan 2000 17:42:07 -0800
Subject: Python and MapObjects (Was: Python and ArcView GIS)
In-Reply-To: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca>
Message-ID: <Pine.SGI.3.95.1000112172541.19328B-100000@clavey.des.ucdavis.edu>

The ArcView extension sounds very interesting.    

Along these lines of interfaces to ESRI software, has anybody done any
work with getting Python to talk to the MapObjects set of components?  I'd
love to have some pointers on how to do this.  Right now I'm stuck on
how to create/access/etc. the Map ActiveX control, which seems basic to
getting much farther with this.

desperately-trying-to-avoid-visual-basic y'rs -- 

 -- allan  





From mgushee at havenrock.com  Mon Jan 17 17:21:24 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: Mon, 17 Jan 2000 17:21:24 -0500 (EST)
Subject: Tkinter: wm_protocol() error on RedHat Linux 6.0
In-Reply-To: <005101bf6136$a0670ea0$f29b12c2@secret.pythonware.com>
References: <m3snzwtq5u.fsf@havenrock.com>
	<005101bf6136$a0670ea0$f29b12c2@secret.pythonware.com>
Message-ID: <14467.38372.747358.211606@gargle.gargle.HOWL>

Fredrik Lundh writes:
 > Matt Gushee <mgushee at havenrock.com> wrote:

 > >  >   File "./sngui.py", line 164, in __init__
 > >  >     self.wm_protocol('WM_DELETE_WINDOW', self.quit)
 > >  > AttributeError: wm_protocol
 > > 
 > > He has tkinter-1.5.1-10.i386.rpm.
 > 
 > pre-1.5.2 releases don't have "wm_protocol".
 > 
 > for backwards compatibility, use the "protocol"
 > alias instead.

Oops. Thank you!

Matt Gushee



From helge.hess at mdlink.de  Thu Jan  6 12:24:53 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Thu, 06 Jan 2000 18:24:53 +0100
Subject: no return values for __init__ ??
Message-ID: <3874CFE5.765A40EA@mdlink.de>

Hi there,

recently I was heavily wondering why __init__ does not allow return
values ! Is there any special reason not to do this ?

For example consider this
  
  C_singleton = None
  
  class C:
    def __init__(self):
      if C_singleton is None:
        c_singleton = self
      return C_singleton

  c1 = C()
  c2 = C()
  'if c1 is c2' will be true

It took me 5 minutes to change the Python interpreter to allow return
values, I'm just wondering why this wasn't implemented in the first
place ?!!

Hm,
  Helge
-- 
SKYRIX-OS  Web Application Environment - http://www.skyrix.com
SKYRIX-IAS Groupware Application Suite - http://www.skyrix.com


From pbible at home.com  Mon Jan  3 19:20:51 2000
From: pbible at home.com (Paul E. Bible)
Date: Tue, 04 Jan 2000 00:20:51 GMT
Subject: CGI Script using Python
Message-ID: <D1bc4.40$hJ1.1161@news.rdc1.tx.home.com>

Hi there,

I'm currently developing a CGI script using Python under Apache Web Server.
I've written a simple script, placed it in the /cgi-bin subdirectory and
perform a chmod 755 on it.  Here is the script:

-----------------------------------------------------------
#!/usr/local/bin/python
import os, string, sys
from cgi import *

def Main():
    print "Content-type: text/html\n\n"
    print "<HTML>"
    print "<HEAD>"
    print "This is a test"
    print "</HEAD>"
    print "<BODY></BODY>"
    print "</HTML>"
    sys.exit(0)

Main()
-----------------------------------------------------------

When the above script is executed on Apache, I get the following message:

-----------------------------------------------------------
500 Internal Server Error
Internal Server Error
The server encountered an internal error or misconfiguration and was unable
to complete
your request.

Please contact the server administrator, webmaster at domain.com and inform
them of the time the error occurred,
and anything you might have done that may have caused the error.
More information about this error may be available in the server error log.
-----------------------------------------------------------

The Apache error log indicates a 'Premature end of script headers:
/usr/local/apache/cgi-bin/test.py'.

My platform is:  Redhat Linux 6.2, Apache Web Server 1.3.9, Python 1.5.2

Any help with this matter would be greatly appreciated.

Thanks in advance,
Paul

----------------------------------------
Paul E. Bible
pbible at littlefishsoftware.com
www.littlefishsoftware.com
----------------------------------------







From sholden at bellatlantic.net  Mon Jan 10 11:11:02 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Mon, 10 Jan 2000 16:11:02 GMT
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu>
	 <slrn87fol9.15t.kc5tja@garnet.armored.net>
	 <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk>
	 <slrn87imad.thc.wtanksle@hawking.armored.net>
	 <slrn87j11s.vc0.kc5tja@garnet.armored.net> <Y$u7SGAdYae4EwJv@jessikat.demon.co.uk>
Message-ID: <387A04AC.886DD88C@bellatlantic.net>

In Samuel Falvo's defence, and without wishing to start a flame war,
might I point out that he specifically mentioned his news server didn't
show him the original article.  It might, therefore, be appropriate to
suggest that he misunderstood your points.  To read his post as an
attack
shows a thinness of skin which won't get you by in the average newsgroup
without daily psychological trauma.  Pythonistas are, in my limited
experience, rather more polite than most.  With the exception of one
particular group member, who can only be described as a complete
<wink>er.  But Tim appears to poke fun with a dry sense of humour
which, as a Yorkshireman, reminds me of home.  [You had COM? You
were lucky.  When I were a lad we had to pull objects out of the
labraries with our bare hands and shout at them to find out how they
worked.
Etc.]

I didn't read his post as attacking you, just expressing his (apparently
well-founded) opinions on Corba and COM.  The nicer parts about COM are
the ways it allows you to discover the interfaces of the objects you
want
to control.  The fact that it works so well with Python lends some
credibility to the assertion that it's one of Microsoft's better ideas.
That and, I would suggest, ODBC, which while not perfect has become a
de facto standard primarily due to its ubefulness and ubiquity.

regards
 Steve

Robin Becker wrote:
> 
> In article <slrn87j11s.vc0.kc5tja at garnet.armored.net>, Samuel A. Falvo
> II <kc5tja at garnet.armored.net> writes
> ...
> 
> Don't get me wrong, but to attack me for saying I treat COM as too
> object oriented and then attack me for being an OO neophyte seems a bit
> silly. I'm from the fortran era, I treat COM objects as exactly that;
> blobs with interfaces, that doesn't mean I don't appreciate nice OO
> languages like Python. The reason I like Python is that it allows me to
> "just do stuff".
>


From wtanksle at hawking.armored.net  Sun Jan  9 22:50:45 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 10 Jan 2000 03:50:45 GMT
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu> <slrn87fol9.15t.kc5tja@garnet.armored.net> <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk>
Message-ID: <slrn87imad.thc.wtanksle@hawking.armored.net>

On Sun, 9 Jan 2000 07:16:22 +0000, Robin Becker wrote:
>In article <slrn87fol9.15t.kc5tja at garnet.armored.net>, Samuel A. Falvo
>II <kc5tja at garnet.armored.net> writes

>I tried com for many years without coming close to understanding it. If
>M$ continues to have huge leverage it will continue to be arcane and for
>wizards only!

Keep in mind the fact that although working with COM may be simple (and
Sam is the authority on this, since he has implemented a reasonably
complete implementation, with a start at documentation), working with
Windows involves more than COM -- you've gotta work out the whole
API/framework classes.

>Python at least allowed me in without a lot of exotic
>verbiage.

That's Python's strength.

>I'm not a whizz at COM, but Bill Tutt, Hark Hammond and the
>other COMerati regularly throw up their hands at the complexity and
>opacity of it all.

Microsoft pays good money to the best minds in the business to design
stuff for them -- and then they hire grads fresh out of college and pay
them below the industry average and expect them to code up the brilliant
designs.

The results are obvious -- an OS, for example, which is nearly uncrashable
by an application, but which regularly bluescreens itself; or a design
framework which embodies a really cutting-edge idea, but which is
destroyed by tons of special cases and bugs.

>Robin Becker

-- 
-William "Billy" Tanksley, in hoc signo hack


From phd at phd.russ.ru  Mon Jan  3 10:55:38 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 3 Jan 2000 15:55:38 +0000 (GMT)
Subject: setting time?
In-Reply-To: <84qfvq$ba4$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.21.0001031552040.18647-100000@fep132.fep.ru>

On Mon, 3 Jan 2000 moonseeker at my-deja.com wrote:
> > > Noone an advice to set the system time?
> >
> >    It is TOOOO system-dependent...
> 
> Tell me for your favorite system:
> Win, AIX, Solaris, IRIX, OS/2

   I cannot say for Win or OS2...

   On all unicies your program MUST run under superuser priviledges (under
root account or be setuid root) - and even there all you can do is actually

import os
os.system("/bin/date '%s'" % date_time_string)

   But Python is good to calculate date_time_string :)

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From glen at electricorb.com  Fri Jan 28 07:47:47 2000
From: glen at electricorb.com (Glen Starchman)
Date: 28 Jan 2000 06:47:47 -0600
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com> <86np80$jhp$1@harvee.billerica.ma.us> <86ntlg$2tc$1@news.ycc.yale.edu> <86pi88$n3n$1@harvee.billerica.ma.us>
Message-ID: <38918EE7.38A43687@electricorb.com>

class test:
indent
def test_function():
indent
print "foo"
dedent
dedent

or, even:

class test: indent(1) def test_function():indent(2) print "foo" dedent(1)
dedent(2)

;-)

"Eric S. Johansson" wrote:

> Paul Magwene <paul.magwene at yale.edu> wrote:
>
> > Eric S. Johansson <esj at harvee.billerica.ma.us> wrote in message news:86np80
> > [snip[
> >> anyone trying to program using speech recognition.  Why would they
> >> do this you might ask?  Probably because they were injured from
> >> programming using keyboards.
>
> > Yes, but isn't it possible to set the speech recognition software so when
> > you say "indent" it tabs right, and when you say "dedent" it tabs left?
>
> yes it is possible but then you lose the words "indent" and "dedent"
> for most uses.  One needs to be a little more careful when picking phrases
> for command use.
>
> But as it turns out, indent vs. outdent commands are among the more minor
> problems when programming by voice.  More major problems include navigation,
> high precision spacing, lots of funny characters such as . & __ and of course
> spelling.  Speech recognition is extremely good at spelling ordinary
> English words.  It gets seriously cranky when you try to force it to use
> English pronunciations with non-English spellings.
>
> a project working on developing speech driven programming technologies
> can be found at: http://ai.iit.nrc.ca/II_public/VoiceCode/
>
> --- eric
>
> --
> Eric S. Johansson    esj at inguide.com    esj at harvee.billerica.ma.us
> This message was composed almost entirely using NaturallySpeaking



From rsmol at bart.nl  Mon Jan  3 16:25:35 2000
From: rsmol at bart.nl (Richard Smol)
Date: Mon, 03 Jan 2000 22:25:35 +0100
Subject: Zope: how to search for files?
Message-ID: <387113CF.75DE590F@bart.nl>

Hi there,

I have been fiddling around with Zope for a while and am trying to do a
simple thing: find out if a certain file-object exists within a folder.
How do I do this? I don't see any methods that achieve this...

TIA!

RS



From skip at mojam.com  Fri Jan 21 13:18:14 2000
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 21 Jan 2000 12:18:14 -0600 (CST)
Subject: Whitespace delimiters suck
In-Reply-To: <868bbv$lug@news.or.intel.com>
References: <000001bf63a1$832e5740$6ea0143f@tim>
	<868bbv$lug@news.or.intel.com>
Message-ID: <14472.41702.276290.678719@beluga.mojam.com>

    tye4> I don't oppose indentation either. It's good for program
    tye4> readability, and readability only. It shouldn't be part of the
    tye4> syntax.

If it's not part of the syntax its use won't be enforced.  It then won't be
applied either consistently or enough and will therefore be useless as an
aid to readability.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From zippy at cs.brandeis.edu  Sat Jan 29 01:57:40 2000
From: zippy at cs.brandeis.edu (Patrick Tufts)
Date: Fri, 28 Jan 2000 22:57:40 -0800
Subject: Web Board/Forum CGI in python.  Does it Exist?
References: <s94tshnfq0s78@corp.supernews.com>
Message-ID: <280120002257407015%zippy@cs.brandeis.edu>

In article <s94tshnfq0s78 at corp.supernews.com>, Joseph J
<islandbase at yahoo.com> wrote:

> Hi, I'm wondering if anyone has developed web board/forum type CGI scripts in 
> Python.  If so, can anyone point me in the right direction.  Thanks.

Zope has a forum module.  www.zope.org

Check out the "squishdot" slashdot clone in Zope.

--Pat


From cgw at fnal.gov  Wed Jan  5 15:42:00 2000
From: cgw at fnal.gov (Charles G Waldman)
Date: Wed, 5 Jan 2000 14:42:00 -0600 (CST)
Subject: Simple cartesian product
Message-ID: <14451.44184.280947.37437@buffalo.fnal.gov>


Magnus L. Hetland writes:
 > > 
 > > What I'm missing here is a simple expression which will make a list
 > > like [1, 2, 3, 4] into [1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4] (if
 > > len(b)==3)...
 > 

If you use Numeric, the "repeat" function does exactly this:

from Numeric import *

x = array([1,2,3,4])
xx = repeat(x,len(x)*(3,))


Of course if you're going to use Numeric, you could (ab)use "dot" to
do what you want:

#!/usr/bin/env python

from Numeric import *
class X:
    def __init__(self,val):
        self.val = val
    def __mul__(self,other):
	return (self.val,other.val)
    def __repr__(self):
        return repr(self.val)
    def __str__(self):
        return str(self.val)

l1=['a','b','c']
l2=['d','e','f']

v1=array(map(X,l1))
v2=array(map(X,l2))

v1.shape=3,1
v2.shape=1,3

print dot(v1,v2)




From Martin.Kew at vsl.com.au  Thu Jan 20 21:37:03 2000
From: Martin.Kew at vsl.com.au (Martin Kew)
Date: Fri, 21 Jan 2000 13:07:03 +1030
Subject: Install problem WNT4.0 & TCL
Message-ID: <3887C64F.AB67C453@vsl.com.au>

Hi,

I have just installed Python a WNT4.0 machine and are having problems
running 'idle'.  When I try to use the Idle icon the machine processed
for a while and then returned with out a message.  I then attempted to
run Idle manually and got the following message.  If I new a bit more
about Tcl I would try to fix it but I figure that you might be
interested in the default install not working.


C:\Python>python c:\Python\Tools\idle\Idle.py
Traceback (innermost last):
  File "c:\Python\Tools\idle\Idle.py", line 3, in ?
    PyShell.main()
  File "C:\Python\Tools\idle\PyShell.py", line 611, in main
    root = Tk()
  File "C:\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
    {} ./lib/tcl8.0 C:/tcl8.0/library C:/library



This probably means that Tcl wasn't installed properly.

C:\Python>

I also tried

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 _tkinter
>>> import Tkinter
>>> Tkinter._test()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "C:\Python\Lib\lib-tk\Tkinter.py", line 1947, in _test
    root = Tk()
  File "C:\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
    {} ./lib/tcl8.0 C:/tcl8.0/library C:/library



This probably means that Tcl wasn't installed properly.

>>>


Noting that I installed using py152.exe and changed the install
directories to c:\Python and c:\Tcl. and also included c:\Python and
c:\Tcl\bin in an effort to fix the problem. Can anyone help ?


--

Regards

Martin Kew

Vision Abell Pty Ltd
Second Avenue, Technology Park
Mawson Lakes SA 5095 Australia
Tel +61 8 8300 4544
Fax +61 8 8349 7420
email Martin.Kew at vsl.com.au
http://www.vsl.com.au

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000121/ce91f2bc/attachment.html>

From just at letterror.com  Tue Jan 18 17:10:49 2000
From: just at letterror.com (Just van Rossum)
Date: Tue, 18 Jan 2000 23:10:49 +0100
Subject: Python Rocks!
In-Reply-To: <vndn1q340ez.fsf@betelgeuse.ccs.neu.edu>
References: Nick Belshaw's message of "Tue, 18 Jan 2000 13:12:52 +0000"
 <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de>
 <388466D4.CC797CFC@earth.ox.ac.uk>
Message-ID: <l03102803b4aa8c9e7bfe@[193.78.237.141]>

[ ... ]
>> > Get used to it, and don't use tabs.
>
>> ! - whats wrong with tabs?
>
>They don't look the same in every editor, or with each individual's
>tab-width settings.

But it's easy to turn that argument around:

Tabs can be nice *because* they "don't look the same in every editor, or
with each individual's tab-width settings": they allow the *viewer* of the
source code to decide what layout looks best for him/her. Tabs are also far
superior if you prefer a proportional width font (yes, people do that,
too!).

Apart from the solid fact that _mixing_ tabs and spaces is evil, there's no
absolute truth here, it's just a matter of preference. So please stop
telling people not to use tabs!

Just





From jeffg at provenance.com.au  Mon Jan 10 02:54:56 2000
From: jeffg at provenance.com.au (jeffg at provenance.com.au)
Date: Mon, 10 Jan 2000 07:54:56 GMT
Subject: getservbyport() in Python?
Message-ID: <85c38c$ksb$1@nnrp1.deja.com>

Does anyone know of any Python implementation of the Linux standard call
getservbyport()? This function is given a network socket port number and
returns the service name. eg if passed 80, will return 'http'. This
information is taken from the /etc/services file.

If nothing exists (& I've looked around pretty thoroughly), then I plan
to scan the services file directly, using re to search for the port
number I want. If I do this, I'll post the function here.

Regards,
Jeff Gray


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


From algaba at my-deja.com  Tue Jan 18 14:58:34 2000
From: algaba at my-deja.com (Manuel Gutierrez Algaba)
Date: Tue, 18 Jan 2000 19:58:34 GMT
Subject: Porn python programm. The ultimate action
Message-ID: <862gl1$m3t$1@nnrp1.deja.com>


Being a bit annoyed with the fact that "anything sucessful in
Internet tends to suck", and considering python a superb (and a bit
forgotten in the midst of thousand of languages) I've being meditating
if a major action would be nice to be taken. Admitting the fact that
Porn sites (sex) are the most succesful and willingly to retain a bit
of dignity, I think it'd be nice to develop a python program that
generates sex histories, for this kind of programm OO would be superb,
and we easily would attract attention, as I think this is a bit
extreme decision to be taken, I think it's fair to ask for your opinion:

What do you think about a "PornStories.py" ?

This is not a joke, It's a serious stuff. Bad words could be expressed
not directly but with invented (obvious words ) or long phrases.

I think it's possible to deal with it with enough dignity and
mock sex sites, at the same time....

-------------
Manolo


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


From ilya at glas.net  Wed Jan 19 05:10:50 2000
From: ilya at glas.net (ilya at glas.net)
Date: 19 Jan 2000 10:10:50 GMT
Subject: Realtime scheduling??
References: <38858064.CDA4866C@yahoo.com>
Message-ID: <8642ja$4do$1@news.glas.net>

To schedule a really small intervals nanosleep() may help (it doesn't
put process to sleep hence may not get scheduled out, though).

Maybe it's worth looking at the realtime Linux kernel or a way to employ
a hardware timer for scheduling.

-ilya

> Is there any way to run a python script, or rather the python
> interpreter itself, with SCHED_FIFO priority on a linux system?
> Or any other way to guarantee that something will happen when you
> really want it to happen??


From loewis at informatik.hu-berlin.de  Fri Jan 28 14:56:12 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Jan 2000 20:56:12 +0100
Subject: Smalltalk blocks in Python
References: <388F8E03.389BC83A@acm.org>
Message-ID: <p6qln5aj7hf.fsf@informatik.hu-berlin.de>

Thierry Thelliez <thierry at acm.org> writes:

> I come from the Smalltalk world and I am trying to understand Python.
> 
> Is there a Smalltalk blocks equivalent in Python ?

Not really. You can do a lot with local functions, though:

def while_do(cond,block):
    while cond():
        block()

class C:
    def __init__(self,n):
        self.counter = 0
        self.values = []
        def more_to_do(self=self,max=n):
            return self.counter<max
        def step(self=self):
            self.counter=self.counter+1
            self.values.append(self.counter)
        while_do(more_to_do,step)

c = C(6)
print c.values
        
I'm not certain that this is a good example, but it shows how
Smalltalk-like things could be done. I hope it also shows that things
shouldn't be done the Smalltalk way all the time :-)

Anyway, you have functions as first-class objects and can pass them
around. The problem is that functions only have their local scope, not
the scope where they've been declared in. So to pass values to the
function, you'll have to bind some parameters with default values.

Regards,
Martin



From gmcm at hypernet.com  Mon Jan 17 19:24:20 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 17 Jan 2000 19:24:20 -0500
Subject: PYTHONPATH newbie question
In-Reply-To: <00091c0e.fbb4d1d4@usw-ex0102-012.remarq.com>
Message-ID: <1263967430-11066578@hypernet.com>

JohnM wrote:

> How does one make changes to PYTHONPATH permanent?

The Right Way To Do It is to create a file <anything>.pth 
somewhere on your existing python path and put the 
directories you want added to the path in there (one line per 
directory).

- Gordon



From grante at visi.com  Mon Jan 17 14:19:46 2000
From: grante at visi.com (grante at visi.com)
Date: Mon, 17 Jan 2000 13:19:46 -0600
Subject: Sucking e-mail from Exchange Server vis HTTP?
In-Reply-To: <3883681A.DBD1B3C1@roguewave.com>; from bjorn@roguewave.com on Mon, Jan 17, 2000 at 12:06:03PM -0700
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com> <slrn886oku.2ac.grant@grante.comtrol.com> <3883681A.DBD1B3C1@roguewave.com>
Message-ID: <20000117131945.A2415@visi.com>

On Mon, Jan 17, 2000 at 12:06:03PM -0700, bjorn wrote:

> > >> I just found the http server page on Exchange server that
> > >> allows access to e-mails via a web browser.  What I'd really
> > >> like is to write a Python program that just sucks the mail out
> > >> of the inbox and delivers it locally.
> >
> > My other option would be to write a MAPI client that runs under
> > Win32 and then gives the mail to the SMTP server on my Linux
> > machine.  There's a MAPI module for Python, isn't there?
> > Perhaps that would be easier -- but I've never programmed under
> > Win32 before...
> 
> If you do this, let us know (I know I would be interested ;-)

I'd rather get mail from the Exchange server directly, but
apparently the protocol used between Exchange Server and
Outlook isn't documented anywhere (it's rumored to be some form
of LDAP-like thing).  I might try getting mail via the Web
interface first.  I can play with that without having to
install Windows. ;)

-- 
Grant Edwards
grante at visi.com



From aahz at netcom.com  Mon Jan 24 11:09:12 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 24 Jan 2000 16:09:12 GMT
Subject: Python Rocks!
References: <000d01bf6312$77edd660$31a2143f@tim> <388BB4E1.974E63B7@prescod.net>
Message-ID: <86htf8$jqu$1@nntp9.atl.mindspring.net>

In article <388BB4E1.974E63B7 at prescod.net>,
Paul Prescod  <paul at prescod.net> wrote:
>
>Wouldn't it be easier if we just added an optional, (but
>compiler-recognized if present) block delimiter to Python 2? Then people
>would have to stop complaining that Python is broken and instead admit
>that Python programmers choose to program without delimiters.

But that's precisely the point: Guido wants (and rightly so) to
completely remove much of the freedom to choose different code layouts.
This makes it much easier for people to read each other's code.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From vincent_berg at hotmail.com  Sun Jan 23 08:17:11 2000
From: vincent_berg at hotmail.com (Vinnieboy)
Date: Sun, 23 Jan 2000 13:17:11 -0000
Subject: New install. Getting ImputError
References: <388AAE5A.B20A7BCE@slip.net>
Message-ID: <OgJxCSaZ$GA.224@net025s>

Hi, maybe I can help,

I'm a newbee too, but..
I think vtest must be a function (or?????) He can't find vtest because there
isn't a function called like that. So create a function called vtest, and in
that function your typing:
print 'this is a test'
In the code from your program you'll have to write..

vtest ()

Greetz from Vinnieboy (the Netherlands),

-I know, my English isn't very well.


Thomas D'Urso <tjdurso at slip.net> schreef in berichtnieuws
388AAE5A.B20A7BCE at slip.net...
> Just installed / started with Python and reading Learning Python. When I
> try to execute this code:
>
>    import test
>    from test import vtest    # <-- error on this line
>    print vtest
>
> file "test" contents:
>
>    vtest = 'this is a test'
>
> I get debugger error for 2nd line:    ImputError: cannot input name
> vtest
> The global variable says ImputError is:     Input can't fine module, or
> can't find name in module.
>
> It give no other errors as far as opening the file. Working on a Mac G4
> with OS 9. Python 1.5.2.
> Doing the "import test.autotest" reports no errors.
>
> TIA Thomas
>
>
>
>




From teep at danaicorp.com  Tue Jan 25 02:38:12 2000
From: teep at danaicorp.com (Prateep Siamwalla)
Date: Tue, 25 Jan 2000 14:38:12 +0700
Subject: newbie question about self and eval
Message-ID: <86jjkj$jmu$1@news.inet.co.th>

Hello o' wise pythoners,

I have a simple question to ask any generous pythoners out there:

I have a simple tab delim text file:
test.txt
---------
fee    harry
fi    joe
fo    jack
fum    jim
fee    john
fo    sam

I trying to make a class to read in these values and stuff them into lists
which this class keeps

(warning, spacing migth be off)
-------
class myclass:

    def __init__(self,filename='c:\\test.txt'):
        self.fee = []
        self.fi = []
        self.fo = []
        self.fum = []
        f = open(filename,'r')
        rl = f.readlines()
        f.close()
        for lines in rl:
            token = string.split(lines,'\011')
# do something more efficient than this !!!
            bucket = token[0]
            if bucket =='fee':
                self.fee.append(token[1])
            elif bucket == 'fi':
                self.fi.append(token[1])
            elif bucket == 'fo':
                self.fo.append(token[1])
            else:
                self.fum.append(token[1])

---------
I know there must be a more effective way of appending items into the
various class lists, (which probably involves eval or something similar, but
I've tried some feeble ideas, none of which seem to fly.

I've tried self.eval(token[0]).append(token[1])  but that obviously failed
in a big way.  Could someone please show me the way?

Thanks in Advance,
teep






From neelk at brick.cswv.com  Sun Jan 30 13:13:45 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 30 Jan 2000 18:13:45 GMT
Subject: Case Sensitivity and Learnability
References: <000501bf6a9f$8f501e00$78a0143f@tim>
Message-ID: <slrn898vr0.ffp.neelk@brick.cswv.com>

Tim Peters <tim_one at email.msn.com> wrote:
> 
> The great thing about newbies is that they pay no attention whatsoever to
> arguments about how they should behave <0.9 wink>.  I recently posted a link
> to Matt Conway's doctoral dissertation (see DejaNews), where his project's
> difficulties in using Python in a newbie-oriented system are explained.  The
> Alice system ships with a modified interpreter that ignores case (and also
> treats int/int as a floating-point operation).

Well, I'd prefer Python to be case-insensitive, just as a matter of my
own preference. Eyeball grep is for me basically case-insensitive, and
I have spent hours trying to find bugs caused by case errors, because
my eyes just slide over the difference.

Python is pretty good about reporting where the error occurred, so
it's rarely more than 10 or 20 minutes of wasted time here, even when
I just give up and rewrite all the code in question. It's still not
fun though.


Neel





From skaller at maxtal.com.au  Sun Jan  9 08:45:26 2000
From: skaller at maxtal.com.au (skaller)
Date: Mon, 10 Jan 2000 00:45:26 +1100
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <011501bf58a4$7b0db540$0100a8c0@rochester.rr.com> <3877513B.73B762F4@maxtal.com.au> <004901bf5a08$11a35be0$0100a8c0@rochester.rr.com>
Message-ID: <387890F6.2D1136F8@maxtal.com.au>

Darrell wrote:
> 
> From: "skaller" <skaller at maxtal.com.au>
> > What would be 'generally' useful is a genuine switch
> > statement -- a sort of unrolled dictionary whose values
> > are pieces of code. Syntax?
> >
> > switch x:
> > case 1: ..
> > case 2: ...
> > default: ...
> >
> > like C, except that each handler has an implicit
> > break at the end of the handler suite (no fall through,
> > like in C).
> >
> > Comments?
> 
> This is a common thing, does this express the idea ?
> 
> def one(arg):
>     print 'ARGV:', arg,
>     return 'RetVal'
> 
> def two(arg):
>     print 'ARGV:', arg,
>     return 'RetVal'
> 
> states={1:one, 2:two, 3:3}
> 
> def switch(state, args, stateTable, default=None):
>     val=stateTable.get(state,default)
>     if callable(val):
>         return val(args)
>     return val

Yes, that provides the required functionality, but has a serious
problem: cases have to be defined 'out of line', and therefore
do not have access to the current scope. 

So I'm looking for a syntax for providing the above
functionality _inline_.

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From thomas at xs4all.net  Thu Jan 13 07:19:31 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 13 Jan 2000 13:19:31 +0100
Subject: inverse of __repr__ ?
In-Reply-To: <yqtn1qa8aga.fsf@toad.scripps>; from ttnttn@toad.scripps on Thu, Jan 13, 2000 at 03:42:13AM -0800
References: <yqtn1qa8aga.fsf@toad.scripps>
Message-ID: <20000113131931.B21652@xs4all.nl>

On Thu, Jan 13, 2000 at 03:42:13AM -0800, Thien-Thi Nguyen wrote:

> i've looked through _programming python_ for the inverse of __repr__, but
> could not find anything.  perhaps i'm missing some clue?

> 	read : write
> 	??   : __repr__

> or is there no way to extend the reader in a python repl?

What do you want the inverse __repr__ to do, re-instantiate the object ? You
can use 'eval' for that, if the __repr__ function of that particular object
allows it. The basic types do, for instance, but for items like file and
socket objects it's impossible. Some standard python classes implement it,
but usually only those who do not rely on on other objects or
unrepresentable data (like filehandles, locks, etc.) When doing such tricks
you also have to make sure you imported the right modules the right way, so
that the eval(__repr__) actually makes sense.

Basically, your read/write analogy isn't functional. __repr__ isn't like
write, it's more like read, and the 'write' side of the equation is simply
typing in the code in an editor/python prompt. __repr__ is more or less a
way to 'print' an object with some technical info attached. __str__ is
usually used to produce 'user-friendly' output.

If you want to reliably dump an object into a file or string and be able to
reinstantiate it, take a look at the shelve and pickle modules (and
cPickle.) They do such work.
-- 
Thomas Wouters <thomas at xs4all.net>

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



From kputland at servicemagic.com  Thu Jan 27 12:27:59 2000
From: kputland at servicemagic.com (Karl Putland)
Date: Thu, 27 Jan 2000 10:27:59 -0700
Subject: DCOracle on Oracle 8.1.x
References: <86klsi$ktb$1@nnrp1.deja.com>
Message-ID: <004301bf68eb$dbb4b680$8901010a@w3denver.com>

Brief testing with DCOracle on NT using Oracle 8.0.5

everything seems to work just fine after I changed the attribute
substitution from the '?' style of mxODBC to the ':p1' ':pn' style of
DCOracle.

-Karl
----- Original Message -----
From: <R.Fehrenbacher at iname.com>
Newsgroups: comp.lang.python
To: <python-list at python.org>
Sent: Tuesday, January 25, 2000 10:18 AM
Subject: DCOracle on Oracle 8.1.x


> Did anybody have success with this combination on
> Solaris 2.6 (or other OS) and Python 1.52.
>
> Got the module compile all right, and most things
> I checked so far work. However some don?t:
>
> Example:
>
> cursor(select * from v$session)
> raises ORA-01455 error, whereas
> cursor(select sid,serial#,username,osuser from
> v$session)
> works just fine.
>
> I had a similar problem with another select * from
> a v$xxxxx table which I can't remember now. Again,
> when I specified the fields explicitly things
> worked.
>
> Could anybody veryfy this problem on Oracle 8.0.5
> e.g., or on other installations.
>
> Cheers,
>
> Roland
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list




From xyz at spud.dink.net  Tue Jan 11 19:31:13 2000
From: xyz at spud.dink.net (RootBoy)
Date: Wed, 12 Jan 2000 00:31:13 GMT
Subject: Default values for function args
References: <slrn87i04q.ad.xyz@sparc5.localnet> <38795F8D.C8C9342A@webamused.com> <slrn87kqp3.c4.xyz@sparc5.localnet> <387AAA57.F672471A@webamused.com>
Message-ID: <slrn87ninc.b5.xyz@sparc5.localnet>

On Tue, 11 Jan 2000 04:05:49 GMT, Joshua Macy <amused at webamused.com> wrote:
 >> 
 >>   Maybe you're thinking that assignment will create a copy of an object,
 >> instead of a copy of a reference to an object?  In Python assignment
 >> always stores references to objects, not copies of the objects
 >> themselves.  In a statement like
 >> 
 >> target = expression
 >> 
 >> expression yields an object, and the assignment stores a reference to
 >> the object in target.  If the expression is simply the name of a
 >> variable, after the assignment both variables have references to the
 >> same object.  This is what's happening in the list example you created.
 >>   To create a copy of an object, you have to take an extra step, as in
 >> the example I gave where I used the slice operation ([:]) to create a
 >> copy of the sequence that contained everything the original sequence
 >> did, and then assigned the reference to that new object to my variable. 
 >> For instance, if explicitly created a copy of local_list before
 >> returning the reference, your function would have behaved as you
 >> expected:
 >> 
 >> 
 >> >>> def fn(a, local_list=[]):   # local_list object gets created here
 >> ... 	local_list.append(a)
 >> ... 	return local_list[:]    # a copy of local_list is created and
 >> returned
 >> ... 
 >> >>> list1 = fn('xyzzy')
 >> >>> print list1
 >> ['xyzzy']
 >> >>> list2 = fn(3.14)
 >> >>> print list1, list2
 >> ['xyzzy'] ['xyzzy', 3.14]
 >> >>> list3 = fn(1000)
 >> >>> print list1, list2, list3
 >> ['xyzzy'] ['xyzzy', 3.14] ['xyzzy', 3.14, 1000]
 >> >>> 
 >> 
 >>   Does that help at all?

Yes, it's finally clear.  I had overlooked an important footnote in the
tutorial:

   
"Actually, call by object reference would be a better description, since 
if a mutable object is passed, the caller will see any changes the callee
makes to it (e.g., items inserted into a list). "

Although this refers to passing arguments into a function, it also apparently
applies to return values.  I had thought I was returning an entire list object
by passing the entire list by value on the stack.  What actually happens is
that a reference to the object is what gets passed back to the caller by value
on the stack.

Thanks again....

   


From w6uv at hotmail.com  Mon Jan 24 19:02:31 2000
From: w6uv at hotmail.com (Jerry Gardner)
Date: Tue, 25 Jan 2000 00:02:31 GMT
Subject: Multidimensional Dictionaries: Possible?
Message-ID: <slrn88ppsg.49t.w6uv@voyager.home.com>

Are multidimensional dictionaries possible in Python? I'm sure they're
possible, but I can't figure out the syntax.

Here's how I do it in Perl:


%dict = ();  # create empty dictionary


$dict{$key1}{$key2} = $string;       # make an entry
$newstring = ${dict{$key1}{$key2}};  # access an entry


-- 
Jerry Gardner     | "Bill Clinton has all the steely resolve
w6uv at hotmail.com  | of a kamikaze pilot on his 37th mission."


From moeller at networksplus.net  Sun Jan 23 03:18:32 2000
From: moeller at networksplus.net (Derek Moeller)
Date: Sun, 23 Jan 2000 02:18:32 -0600
Subject: TCL_GetFile removed
In-Reply-To: <20000123013648.B27888@networksplus.net>; from moeller@networksplus.net on Sun, Jan 23, 2000 at 01:36:48AM -0600
References: <20000123013648.B27888@networksplus.net>
Message-ID: <20000123021831.C27888@networksplus.net>

On Sun, Jan 23, 2000 at 01:36:48AM -0600, Derek Moeller wrote:
> I've been compiling 1.5.2 of Python, and it stops compiling when it gets
> to the TCL bit. There are undefined ref's to TCL_GetFile.
> 
> I searched Deja, and found that that feature (TCL_GetFile) had been
> removed somewhere in the 8.* line.
> 
> Which TCL version is the latest safe way to go?
> 

Well, I "fixed" the problem. A simple 'make clean' apparently cleared the
problems up.

And I suppose that Usenet post was incorrect... (?)

-- 
Derek Moeller
  moeller at networksplus.net
  http://dophnic.yi.org



From ivanlan at callware.com  Thu Jan 27 00:00:37 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 26 Jan 2000 22:00:37 -0700
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
References: <816010E2456BD111A48700805FBBE2EE01C60408@ex-quebec-u1.baan.com> <388FC511.C776341A@one.net.au> <00a101bf6882$aa7cb540$9600a8c0@execpc.com>
Message-ID: <388FD0F5.B890FCFD@callware.com>

Hi All--

Stuart Ford wrote:
> 
> How did this message make it into the list in the first place?
> 
> Did the spammer fake his/her sending address as that of a valid
> list member, or is the list not sent to allow only registered members
> to post?
> 

Because it's a gatewayed newsgroup, the Python mailing list can't reject
non-subscribers.

<sad-sad-story>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From tim_one at email.msn.com  Wed Jan 26 01:02:55 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 26 Jan 2000 01:02:55 -0500
Subject: Usability reports (was RE:tabs do WHAT?)
In-Reply-To: <200001241707.JAA14172@seanb.sttln1.wa.home.com>
Message-ID: <000301bf67c2$fda182c0$592d153f@tim>

[Sean Blakey, discovers some old ABC reports available for order
 from http://www.cwi.nl/~steven/abc/publications.html
]

> ...
> 	I have heard many claims about Python's usability and
> the ease of learning python.  While my own experience strongly
> concurs with these claims, I am eager to get my hands on some
> concrete information about this.  Perhaps eventually such
> information would be available from the python.org website?

That presumes there's "concrete information" to be had, though.  AFAIK, the
ABC project was both the first and the last to get funding to study
programming language design as a human factors problem.  Python took some of
its ideas, but is not proceeding the same way.  Guido seems to have trouble
enough getting funding for flashy, whiz-bang ideas like CP4E (let alone
mundane ones like mere UI tweaking <wink>).

The only other concrete info out there I can recall is Matt Conway's
doctoral dissertation, available from

    http://research.microsoft.com/~mconway/dissertation.htm

This is mostly about the Alice system (teaching 3D graphics programming
techniques to newbies), and is a darned good read; a variant of Python
serves as Alice's programming language.

for-personal-research-try-teaching-one-newbie-python-and-
    another-one-perl-ly y'rs  - tim

PS:  I would object to my tax dollars being used to fund such a predictable
study <wink>.





From jhauser at ifm.uni-kiel.de  Sat Jan  8 15:56:08 2000
From: jhauser at ifm.uni-kiel.de (Janko Hauser)
Date: 08 Jan 2000 21:56:08 +0100
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
References: <20000105082703.A9441@vislab.epa.gov> <1265040530-3892638@hypernet.com> <ONAe+FAsO4c4Ew5h@jessikat.demon.co.uk> <pqKd4.2780$Ef6.704898@news.shore.net> <IN2z0BAYY5d4EwgO@jessikat.demon.co.uk>
Message-ID: <87embs8eqf.fsf@ifm.uni-kiel.de>

Sorry to jump in so latly. I haven't used it myself, but I understand
that the ``fpectlmodule.c'' from standard distribution has some of the
needed infrastructure to catch exceptions from the C-code and handle
them.

But the code talks about hardware specific things. Have you looked at
this module already or am I offtopic?

HTH
__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 Sven.Drescher at dlr.de  Mon Jan 31 03:49:16 2000
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Mon, 31 Jan 2000 09:49:16 +0100
Subject: problems with blocking pipes, threading
References: <86rqr7$gn2$1@news.go.dlr.de> <86shaa$15g2$1@nntp6.u.washington.edu>
Message-ID: <873i5r$jhi$1@news.go.dlr.de>

> If you want to see whether this is really the problem, you might
> write a sample program like  sys.stdout.write('Hello\n');
time.sleep(1000.0)
> and verify that its pipe blocks this way.  Then add sys.stdout.flush()
> after the write, and see if anything is different.

Thanks for the hint! That's the problem. The programm works correct, if
using 'sys.stdout.write()'.
But in my further work I start processes, where I cannont know, how the
standard output is realized. Is there a possibility to flush the child's
stdout??? Or is there a complete other way to catch the child's stdout in
'real-time'???

Thanks!
Sven




From thomas at xs4all.net  Fri Jan 28 12:20:32 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 28 Jan 2000 18:20:32 +0100
Subject: nesting class definitions?
In-Reply-To: <20000127155551.A2069@stopcontact.palga.uucp>; from gerrit.holl@pobox.com on Thu, Jan 27, 2000 at 03:55:51PM +0100
References: <20000127155551.A2069@stopcontact.palga.uucp>
Message-ID: <20000128182031.C16093@xs4all.nl>

On Thu, Jan 27, 2000 at 03:55:51PM +0100, Gerrit Holl wrote:

> I do see the point in nesting function definitions, namely as an alternative
> on places where one could use lambda. But what's the point in nesting
> class definitions? I know it's possible, but I've never seen it used.

> There aren't functions in the default library that take a class as
> argument, are there?

Nesting doesn't really *do* anything in python, except bind the class to a
different namespace. And there is no difference between classes, modules,
functions, lists, dictionaries or any of the other objects. For instance:

class Viking:
	class Spam:
		pass

is effectively equivalent to:

class private_Spam:
	pass

class Viking:
	Spam = private_Spam

Because python only has 2 'active' namespaces, local and global, it does not
matter where a class or function is created, other than in which module. It
will only ever 'see' it's own namespace, which is private to itself, and the
module toplevel namespace.

And dont forget that you can pass classes and functions around just like any
other object:

>>> class Viking:
...         class Spam:
...                 def __init__(self, counter):
...                         self.counter = counter
...                 def __call__(self):
...                         print "spam "*self.counter
...         def __init__(self):
...                 self.counter = 0
...         def __call__(self):
...                 self.counter = self.counter + 1
...                 return self.Spam(self.counter)
...
>>> foo = Viking()
>>> foo()
<__main__.Spam instance at 80d4dd8>
>>> foo()
<__main__.Spam instance at 80d4ac0>
>>> foo()()
spam spam spam 
>>> foo()()
spam spam spam spam 

>>> class Spam:
...     def __init__(self, counter):
...             pass
...     def __call__(self):
...             print "I'm NOT the messiah!"
...
>>> Viking.Spam = Spam
>>> foo()()
I'm NOT the messiah!

Python makes no distinction between the original Spam class and the new one.
All existing instances of the old Spam keep their parent, of course, but any
new instances generated by Viking's __call__ method will use the new one.
The 'toplevel' Spam also doesn't overwrite the Viking.Spam until the
actual assignment. (The internal 'name' for Viking.Spam, __main__.Spam,
notwithstanding.)

So, in a sense, there is no actual nesting. The definition of a class or a
function does 2 things: creating the actual class or function object, and
assigning it to a name in the local namespace. In what namespace you execute
the 'class' or 'def' statement only affects where the name is bound, not how
it's treated.

And why should we use nesting then, if it's all namespace-magic ? Well,
because it's namespace magic, and python is cool like that. Perhaps you do
not want to pollute the main namespace with auxillary functions and classes,
or perhaps you want to have your auxillary class or function 'close', to
speed up accesses to them. Or perhaps it's just more obvious, easier to
read.

(There are, by the way, functions that take a class as argument... think
'isinstance()'. But wether or not class definitions are nested does not in
any way affect it. You can call isinstance() on a locally defined class just
as you can do it on a globally defined class, or qualified with a namespace
(Viking.Spam, myModule.Spam, etc)

To miss-quote Tim,
Namespaces-are-one-honking-great-idea----let's-do-more-of-those!-ly y'rs
-- 
Thomas Wouters <thomas at xs4all.net>

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



From Alex.Martelli at think3.com  Fri Jan 14 12:17:36 2000
From: Alex.Martelli at think3.com (Alex Martelli)
Date: Fri, 14 Jan 2000 18:17:36 +0100
Subject: Python Pre-processor (like php/asp)
Message-ID: <6D8A17398E28D3119F860090274DD7DB4B3DD7@pces.cadlab.it>

Garry Hodgson writes:

> Patrice Girard wrote:
> > 
> > Yes, ASP do the job.  Just to install Pythonwin and put SCRIPT=python
> inside
> > your .ASP and it's running!
> 
> is this limited to microsoft web servers/clients?
> 
ASP requires ISAPI, which AFAIK, so far, is only implemented
in Microsoft's PWS and IIS (I'd love to be proved wrong...:-).

There is absolutely no limitation to Microsoft *clients*, of course --
ASP's can be visited from clients of whatever stripe.


Alex




From thomas at xs4all.net  Fri Jan 21 06:48:35 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 21 Jan 2000 12:48:35 +0100
Subject: Whitespace delimiters suck
In-Reply-To: <868bbv$lug@news.or.intel.com>; from tye4@yahoo.com on Thu, Jan 20, 2000 at 05:04:34PM -0800
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com>
Message-ID: <20000121124834.A10685@xs4all.nl>

On Thu, Jan 20, 2000 at 05:04:34PM -0800, tye4 wrote:
> Whitespace end delimiters are invisible and hence confusing and prone to
> error. I've used some stupid Unix shell languages which fail if I add
> whitespace:

> I thought they gotten rid of all such languages until I came across Python.
> Good programming languages should be whitespace tolerant, and most of them
> are.
> (C++, Java, Ada, BASIC, HTML, Eiffel, Perl, Pascal, ML etc... really long
> list.)

Yeah, right. You dont honestly believe all of those languages are
white-space tolerant ? In C (or C++), what does this mean:

y+++x;

Is that y++ + x, or y + ++x ? If you add whitespace, the compiler will do
something different. Worse, this is invalid:

y+++++x;

Eventhough there IS only one possiblity, whitespace is NEEDED to make the
distinction. If you do not use whitespace, the compiler will think you mean
'y++ ++ +x' and will complain. The same goes for other languages with
autoincrement/decrement, by the way.

> Why is my question about whitespace raise so many eyebrows? This feature is
> commonplace.

Whitespace is a GOOD thing. it makes things readable. You aren't, however,
objecting to the use of whitespace, you're objecting to the use of
indentation. No, wait, not even that, you're objecting to the lack of
end-block markers, which you would use _in addition_ to indentation. Well,
you can do that, using the 2nd neatest invention of Guido's, called
'comments'. They're so cool, every language implemented them right away,
sometimes even using Guido's time machine to correct the oversight years
ago !

This whole discussion and your plan to add end-block markers reminds me of
how the bourne shell (the original one) was written. In C, but with a bunch
of macros to make the language look more like another language (Pascal, i
think.)

#define IF if (
#define THEN ) {
#define ELSE } else {
#define ENDIF }

and similar ones for while/do-while, for, switch, etc. Apparently, it was a
lot easier for Steve Bourne to program in, because he was used to these
constructs, but it made maintaining the original bourne shell a living
nightmare for everyone who was not used to these constructs (and, i imagine,
for Steve himself aswell, after he got used to C and forgot about Pascal :)

Steve probably didn't have much choice, he had to program the bourne shell
in C. But you DO have a choice. There are so many programming languages
about, nowadays, and a lot that you are free to download, install and use in
any environment. Dont like the indentation as block delimiters ? It's one of
the core features of Python, and it reflects quite accurately Python's
philosophy. I'd venture to guess you wont get the same satisfaction out of
Pythons (to us) obvious features, and it's great intuitivity. Perhaps you
can find everything you need in some other language. I doubt molding and
mauling the language into something you do like, making it incompatible with
the 'core' python, will get you that satisfaction.

(auto-increment and bourne-shell examples blatantly stolen from Peter v/d
Lindens' excellent "Deep C Secrets", by the way, which my SO misplaced
again, which is why i'm not clear in the details of the last story ;P
Frustrating how she always misplaces my favorite books.)

			-ly y'rs,

-- 
Thomas Wouters <thomas at xs4all.net>

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



From doughellmann at home.com  Tue Jan 11 06:48:18 2000
From: doughellmann at home.com (Doug Hellmann)
Date: Tue, 11 Jan 2000 11:48:18 GMT
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim>
Message-ID: <387B19F2.AC17FA8D@home.com>


Tim Peters wrote, in response to Neel Krishnaswami's post on multimethods:
> 
> Can't say I was surprised -- few people had thought about multimethods then.
> They're better known now, but also much easier to get at via Dylan now
> (which people should definitely look into if they like what they see in your
> emulation).

Could one of you post a brief description of what a multimethod is, or
maybe just a pointer to a definition somewhere?

Thanks,
Doug


From tjg at avalongroup.net  Sun Jan  9 12:03:01 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Sun, 09 Jan 2000 09:03:01 -0800
Subject: More Tkinter questions
Message-ID: <3878BF45.7C2A0DA0@exceptionalminds.com>

Hi all,

I'm having a great time playing with Tkinter, and am quite pleased with
some of the stuff I've accomplished, however, I'm having quite a bit of
difficulty understanding a couple of concepts on a particularly complex
screen display.

I'm basically clueless about how to accomplish what I want to accomplish
and how the various layout managers work. Ready for some ASCII art....

+---------------------------------------+
| Frame A                               |
+--------------+----------+-------------+
|Frame B       | Frame C  | Frame E     |
|              |          |             |
|              |          |             |
|              |          |             |
|              +----------+             |
|              | Frame D  |             |
|              |          |             |
|              |          |             |
|              |          |             |
|              |          |             |
+--------------+----------+-------------+

Of course each one of the frames contains a number of widgets.

When using the grid layout manager it seems to me that each frame should
have the following attributes:

Frame A  row=0, column=0, columnspan=3
Frame B  row=1, column=0, rowspan=2
Frame C  row=1, column=1
Frame D  row=2, column=1
Frame E  row=1, column=2, rowspan=2

However, once I start adding various widgets to the various frames
things go south really fast. It appears that I don't get a new grid when
I start adding objects to a frame, but I work on some sort of master
grid. When this happens all of the rows and columns shown above start
losing there relationships.

Can someone give me a good solution to this?

BTW: it doesn't help that none of the frames hold a specified size when
widgets start getting added to them.

Thanks very much!

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From news at dorb.com  Thu Jan 13 15:05:21 2000
From: news at dorb.com (Darrell)
Date: Thu, 13 Jan 2000 15:05:21 -0500
Subject: file width and height
References: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com>
Message-ID: <ubqf4.3018$hN5.92483@typhoon.nyroc.rr.com>

Sposhua wrote:
>Is there a module to find gif/jpeg width and height?

I hacked this out of PIL and it should work for gif and jpg.

--
--Darrell


begin 666 imageSize.py
M(R$O=7-R+V)I;B]E;G8@<'ET:&]N#0H-"B(B(@T*2&%C:V5D(&9O<B!024P@
M8GD at 1&%R<F5L;"!'86QL:6]N#0HB(B(-"@T*:6UP;W)T('-Y<PT*:6UP;W)T
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M+"!M87)K97(I. at T*(" @(",-"B @(" C($%P<&QI8V%T:6]N(&UA<FME<BX@
M(%-T;W)E('1H97-E(&EN('1H92!!4% @9&EC=&EO;F%R>2X-"B @(" C($%L
M<V\@;&]O:R!F;W(@=V5L;"UK;F]W;B!A<'!L:6-A=&EO;B!M87)K97)S+ at T*
M#0H@(" @<R ]('-E;&8N9G N<F5A9"AI,39J<&<H<V5L9BYF<"YR96%D*#(I
M*2TR*0T*#0ID968 at 4T]&*'-E;&8L(&UA<FME<BDZ#0H@(" @(PT*(" @(",@
M4W1A<G0@;V8 at 9G)A;64@;6%R:V5R+B @1&5F:6YE<R!T:&4@<VEZ92!A;F0@
M;6]D92!O9B!T:&4-"B @(" C(&EM86=E+B @2E!%1R!I<R!C;VQO=7(@8FQI
M;F0L('-O('=E('5S92!S;VUE('-I;7!L90T*(" @(",@:&5U<FES=&EC<R!T
M;R!M87 @=&AE(&YU;6)E<B!O9B!L87EE<G,@=&\@86X at 87!P<F]P<FEA=&4-
M"B @(" C(&UO9&4N("!.;W1E('1H870@=&AI<R!C;W5L9"!B92!M861E(&$@
M8FET(&)R:6=H=&5R+"!B>0T*(" @(",@;&]O:VEN9R!F;W(@2D9)1B!A;F0@
M061O8F4 at 05!0(&UA<FME<G,N#0H-"B @("!S(#T@<V5L9BYF<"YR96%D*&DQ
M-FIP9RAS96QF+F9P+G)E860H,BDI+3(I#0H@(" @<V5L9BYS:7IE(#T@:3$V
M:G!G*'-;,SI=*2P@:3$V:G!G*'-;,3I=*0T*(" @(')E='5R;B!S96QF+G-I
M>F4-"@T*(" @('-E;&8N8FET<R ](&]R9"AS6S!=*0T*(" @(&EF('-E;&8N
M8FET<R A/2 X. at T*"7)A:7-E(%-Y;G1A>$5R<F]R+" B8V%N;F]T(&AA;F1L
M92 E9"UB:70@;&%Y97)S(B E('-E;&8N8FET<PT*#0H@(" @<V5L9BYL87EE
M<G,@/2!O<F0H<ULU72D-"B @("!I9B!S96QF+FQA>65R<R ]/2 Q. at T*"7-E
M;&8N;6]D92 ]("),(@T*(" @(&5L:68@<V5L9BYL87EE<G,@/3T@,SH-"@ES
M96QF+FUO9&4@/2 B4D="(@T*(" @(&5L:68@<V5L9BYL87EE<G,@/3T at -#H-
M"@ES96QF+FUO9&4@/2 B0TU92R(-"B @("!E;'-E. at T*"7)A:7-E(%-Y;G1A
M>$5R<F]R+" B8V%N;F]T(&AA;F1L92 E9"UL87EE<B!I;6%G97,B("4@<V5L
M9BYL87EE<G,-"@T*(" @(&EF(&UA<FME<B!I;B!;,'A&1D,R+" P>$9&0S8L
M(#!X1D9#02P@,'A&1D-%73H-"@ES96QF+FEN9F];(G!R;V=R97-S:6]N(ET@
M/2 Q#0H-"B @("!F;W(@:2!I;B!R86YG92 at V+"!L96XH<RDL(#,I. at T*"70@
M/2!S6VDZ:2LS70T*"2, at -"UT=7!L97,Z(&ED+"!V<V%M<"P@:'-A;7 L('%T
M86)L90T*"7-E;&8N;&%Y97(N87!P96YD*'1;,%TL(&]R9"AT6S%=*2\Q-BP@
M;W)D*'1;,5TI)C$U+"!O<F0H=%LR72DI#0H-"F1E9B!$450H<V5L9BP@;6%R
M:V5R*3H-"B @("!S(#T@<V5L9BYF<"YR96%D*&DQ-FIP9RAS96QF+F9P+G)E
M860H,BDI+3(I#0H-"@T*(PT*(R!*4$5'(&UA<FME<B!T86)L90T*#0I-05)+
M15(@/2![#0H@(" @,'A&1D,P.B H(E-/1C B+" B0F%S96QI;F4 at 1$-4(BP@
M4T]&*2P-"B @(" P>$9&0S$Z("@B4T]&,2(L(")%>'1E;F1E9"!397%U96YT
M:6%L($1#5"(L(%-/1BDL#0H@(" @,'A&1D,R.B H(E-/1C(B+" B4')O9W)E
M<W-I=F4 at 1$-4(BP at 4T]&*2P-"B @(" P>$9&0S,Z("@B4T]&,R(L(")3<&%T
M:6%L(&QO<W-L97-S(BP at 4T]&*2P-"B @(" P>$9&0S0Z("@B1$A4(BP@(D1E
M9FEN92!(=69F;6%N('1A8FQE(BP at 4VMI<"DL#0H@(" @,'A&1D,U.B H(E-/
M1C4B+" B1&EF9F5R96YT:6%L('-E<75E;G1I86P at 1$-4(BP at 4T]&*2P-"B @
M(" P>$9&0S8Z("@B4T]&-B(L(")$:69F97)E;G1I86P@<')O9W)E<W-I=F4@
M1$-4(BP at 4T]&*2P-"B @(" P>$9&0S<Z("@B4T]&-R(L(")$:69F97)E;G1I
M86P@<W!A=&EA;"(L(%-/1BDL#0H@(" @,'A&1D,X.B H(DI01R(L(")%>'1E
M;G-I;VXB+"!.;VYE*2P-"B @(" P>$9&0SDZ("@B4T]&.2(L(")%>'1E;F1E
M9"!S97%U96YT:6%L($1#5" H04,I(BP at 4T]&*2P-"B @(" P>$9&0T$Z("@B
M4T]&,3 B+" B4')O9W)E<W-I=F4 at 1$-4("A!0RDB+"!33T8I+ T*(" @(#!X
M1D9#0CH@*")33T8Q,2(L(")3<&%T:6%L(&QO<W-L97-S($1#5" H04,I(BP@
M4T]&*2P-"B @(" P>$9&0T,Z("@B1$%#(BP@(D1E9FEN92!A<FET:&UE=&EC
M(&-O9&EN9R!C;VYD:71I;VYI;F<B+"!3:VEP*2P-"B @(" P>$9&0T0Z("@B
M4T]&,3,B+" B1&EF9F5R96YT:6%L('-E<75E;G1I86P at 1$-4("A!0RDB+"!3
M3T8I+ T*(" @(#!X1D9#13H@*")33T8Q-"(L(")$:69F97)E;G1I86P@<')O
M9W)E<W-I=F4 at 1$-4("A!0RDB+"!33T8I+ T*(" @(#!X1D9#1CH@*")33T8Q
M-2(L(")$:69F97)E;G1I86P@<W!A=&EA;" H04,I(BP at 4T]&*2P-"B @(" P
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M(&EM86=E(BP at 3F]N92DL#0H@(" @,'A&1D0Y.B H(D5/22(L(")%;F0@;V8@
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M(BP at 05!0*2P-"B @(" P>$9&13,Z("@B05!0,R(L(")!<'!L:6-A=&EO;B!S
M96=M96YT(#,B+"!!4% I+ T*(" @(#!X1D9%-#H@*")!4% T(BP@(D%P<&QI
M8V%T:6]N('-E9VUE;G0 at -"(L($%04"DL#0H@(" @,'A&1D4U.B H(D%04#4B
M+" B07!P;&EC871I;VX@<V5G;65N=" U(BP at 05!0*2P-"B @(" P>$9&138Z
M("@B05!0-B(L(")!<'!L:6-A=&EO;B!S96=M96YT(#8B+"!!4% I+ T*(" @
M(#!X1D9%-SH@*")!4% W(BP@(D%P<&QI8V%T:6]N('-E9VUE;G0 at -R(L($%0
M4"DL#0H@(" @,'A&1D4X.B H(D%04#@B+" B07!P;&EC871I;VX@<V5G;65N
M=" X(BP at 05!0*2P-"B @(" P>$9&13DZ("@B05!0.2(L(")!<'!L:6-A=&EO
M;B!S96=M96YT(#DB+"!!4% I+ T*(" @(#!X1D9%03H@*")!4% Q,"(L(")!
M<'!L:6-A=&EO;B!S96=M96YT(#$P(BP at 05!0*2P-"B @(" P>$9&14(Z("@B
M05!0,3$B+" B07!P;&EC871I;VX@<V5G;65N=" Q,2(L($%04"DL#0H@(" @
M,'A&1D5#.B H(D%04#$R(BP@(D%P<&QI8V%T:6]N('-E9VUE;G0@,3(B+"!!
M4% I+ T*(" @(#!X1D9%1#H@*")!4% Q,R(L(")!<'!L:6-A=&EO;B!S96=M
M96YT(#$S(BP at 05!0*2P-"B @(" P>$9&144Z("@B05!0,30B+" B07!P;&EC
M871I;VX@<V5G;65N=" Q-"(L($%04"DL#0H@(" @,'A&1D5&.B H(D%04#$U
M(BP@(D%P<&QI8V%T:6]N('-E9VUE;G0@,34B+"!!4% I+ T*(" @(#!X1D9&
M,#H@*")*4$<P(BP@(D5X=&5N<VEO;B P(BP at 3F]N92DL#0H@(" @,'A&1D8Q
M.B H(DI01S$B+" B17AT96YS:6]N(#$B+"!.;VYE*2P-"B @(" P>$9&1C(Z
M("@B2E!',B(L(")%>'1E;G-I;VX@,B(L($YO;F4I+ T*(" @(#!X1D9&,SH@
M*")*4$<S(BP@(D5X=&5N<VEO;B S(BP at 3F]N92DL#0H@(" @,'A&1D8T.B H
M(DI01S0B+" B17AT96YS:6]N(#0B+"!.;VYE*2P-"B @(" P>$9&1C4Z("@B
M2E!'-2(L(")%>'1E;G-I;VX at -2(L($YO;F4I+ T*(" @(#!X1D9&-CH@*")*
M4$<V(BP@(D5X=&5N<VEO;B V(BP at 3F]N92DL#0H@(" @,'A&1D8W.B H(DI0
M1S<B+" B17AT96YS:6]N(#<B+"!.;VYE*2P-"B @(" P>$9&1C at Z("@B2E!'
M."(L(")%>'1E;G-I;VX at ."(L($YO;F4I+ T*(" @(#!X1D9&.3H@*")*4$<Y
M(BP@(D5X=&5N<VEO;B Y(BP at 3F]N92DL#0H@(" @,'A&1D9!.B H(DI01S$P
M(BP@(D5X=&5N<VEO;B Q,"(L($YO;F4I+ T*(" @(#!X1D9&0CH@*")*4$<Q
M,2(L(")%>'1E;G-I;VX@,3$B+"!.;VYE*2P-"B @(" P>$9&1D,Z("@B2E!'
M,3(B+" B17AT96YS:6]N(#$R(BP at 3F]N92DL#0H@(" @,'A&1D9$.B H(DI0
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M(" @:68 at 34%22T52+FAA<U]K97DH:2DZ#0H@(" @(" @(" @("!N86UE+"!D
M97-C<FEP=&EO;BP@:&%N9&QE<B ]($U!4DM%4EMI70T*(" @(" @(" @(" @
M(R!P<FEN="!H97 at H:2DL(&YA;64L(&1E<V-R:7!T:6]N#0H@(" @(" @(" @
M("!I9B!H86YD;&5R(&ES(&YO="!.;VYE. at T*(" @(" @(" @(" @(" @(&-C
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M(" @(" @(" @<F%I<V4 at 4WEN=&%X17)R;W(L(")N;R!M87)K97(@9F]U;F0B
M#0H-"F-L87-S($1U;6UY. at T*(" @('!A<W,-"@T*9&5F(&EM86=E4VEZ92 at I
M. at T*(" @(&9P/6]P96XH<WES+F%R9W9;,5TL)W(K8B<I#0H@(" @8G5F/69P
M+G)E860H,3,I#0H@(" @:68 at 8G5F6SHV72!I;B!;(D=)1C at W82(L(")'248X
M.6$B73H-"B @(" @(" @<')I;G0 at 8G5F6SHV70T*(" @(" @("!S:7IE/2AI
M,38H8G5F6S8Z72DL:3$V*&)U9ELX.ETI*0T*(" @(" @("!R971U<FX@<VEZ
M90T*(" @(&5L<V4Z#0H@(" @(" @(&9P+G-E96LH,"D-"B @(" @(" @<V5L
M9CU$=6UM>2 at I#0H@(" @(" @('-E;&8N9G ]9G -"B @(" @(" @<R ](&9P
M+G)E860H,2D-"B @(" @(" @:68@;W)D*'-;,%TI(#T](#(U-3H-"B @(" @
M(" @(" @(',@/2!F<"YR96%D*#$I#0H@(" @(" @(" @("!S(#T at 9G N<F5A
M9"@Q*0T*(" @(" @(" @(" @<F5T=7)N(%]O<&5N*'-E;&8L(',I#0H@(" @
M(" @(" @("!S(#T at 9G N<F5A9"AI,38H9G N<F5A9"@R*2DM,BD-"B @(" @
M(" @(" @('-I>F4@/2!I,38H<ULS.ETI+"!I,38H<ULQ.ETI#0H@(" @(" @
M(" @("!R971U<FX@<VEZ90T*#0H-"FEF(%]?;F%M95]?(#T](")?7VUA:6Y?
<7R(Z#0H@(" @<')I;G0@:6UA9V53:7IE*"D-"@``
`
end



From backov at nospam.csolve.net  Thu Jan  6 15:30:08 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Thu, 06 Jan 2000 20:30:08 GMT
Subject: newbie embedding questions
Message-ID: <kX6d4.41059$Ke.26557@news21.bellglobal.com>

Well, I've finally gotten around to embedding Python in an app, after
thinking what a cool language it was for a couple years now. ;> However, I
have some stupid C++ programmer questions...

The main app is essentially a skeleton, that will be driven by Python
scripts.. It does however hold a bunch of C++ objects that it owns and
instantiates and that the Python scripts need access to. So basically I need
"glue" code to make sure the Python stuff has access to those instances'
methods, and to set callbacks in the python code for generated events. The
glue code itself is not too challenging to write (just going to have to read
the Python reference some more) but I'm a little confused as to how it
should be implemented. I can write a .c "spammodule" or whatever and have it
include the main programs .h and thus it's objects, but will that work? The
import command seems to want an .o file or something... Basically I'm a bit
confused at this point. Is there some way I can set up the glue code without
having a .o import module? The docs don't go too indepth on embedding.

Any help would be greatly appreciated.

Cheers,
Jason





From jimc at regnoc.com  Sun Jan 30 00:34:44 2000
From: jimc at regnoc.com (Jim Conger)
Date: Sun, 30 Jan 2000 05:34:44 GMT
Subject: Python MySQL.connect() question
Message-ID: <3893CD31.B17E8B84@regnoc.com>

I have mySQL running on a RH Linux 6.1 box, and have added the
MySQLmodule.so to my python library.  I can't figure out how to connect
within python.  My attempts look like:

>>> import MySQL;
>>> db = MySQL.connect();
Traceback (innermost last):
  File "<stdin>", line 1, in ?
MySQL.error: Can't connect to local MySQL server
>>>

I know the mySQL server is running in the background during this test,
so I'm noplussed.  Any examples of working code would be appreciated.

Jim C.



From marut at my-deja.com  Wed Jan 26 16:18:02 2000
From: marut at my-deja.com (marut at my-deja.com)
Date: Wed, 26 Jan 2000 21:18:02 GMT
Subject: parnassus? When will it be back up?
Message-ID: <86noa5$uh9$1@nnrp1.deja.com>

When I visit the parnassus page I see only one word: Conn

Anyone know its status?


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


From fredrik at pythonware.com  Thu Jan 13 03:08:07 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 13 Jan 2000 09:08:07 +0100
Subject: Ping in Python
References: <SKd8OPFYifsAUCG=19+7XNNTjqDx@4ax.com>
Message-ID: <007001bf5d9d$55e2d180$f29b12c2@secret.pythonware.com>

Anders M Eriksson wrote:
> Everyday about 1.30 pm my Intenet just dies. Everything takes forever
> and IE nearly almost will TimeOut on all pages. So I need to test and
> document this so I can tell my ISP that there is something wrong.
> 
> So I thought: Why don't I create an Python program that will PING a
> list of hosts and log the response! 
> 
> But how does a PING work?

it sends a low-level ICMP package; you have
to open the socket as SOCK_RAW, and build
the package yourself.

(and on a Unix box, your program needs to
run as root...)

it's probably easiest to run an external ping
utility and parse the output.  lots of people
have written ping wrappers; searching the
newsgroup [1] for "ping command" or "ping"
should help you find most of them:

and if you insist, Jeremy Hylton has ping and
traceroute code at:
    http://the-tech.mit.edu/~jeremy/python/

</F>

[1] www.python.org -> search -> newsgroup

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
-->




From tjg at avalongroup.net  Mon Jan 17 18:07:54 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Mon, 17 Jan 2000 15:07:54 -0800
Subject: DBI spec and parameter passing
Message-ID: <3883A0CA.AB0D4854@exceptionalminds.com>

Hi again,

I've been reading up on DBI and what it does, and I see that when I have
cursor for which I need to execute some SQL, I pass it some SQL and some
parameters, can someone please show me an example of how that is
supposed to look? Right now I build up a string using something like
this

	sql = "select * from table where id='%s'" % idvar
	cursor.execute(sql)

This seems to work quite well except in the in the instances where I
have *Empty* Data fields, then mxODBC reports errors back to me that
look like this:

('08S01', 1, "Error while executing the query;\012ERROR:  Bad date
external representation ''", 3205)

I'm not quite sure what to do to alleviate that problem.

Any help is greatfully accepted!
	
-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From hafhat at cs.rmit.edu.au  Mon Jan 31 08:37:00 2000
From: hafhat at cs.rmit.edu.au (Hafeez Bana)
Date: Mon, 31 Jan 2000 13:37:00 GMT
Subject: Python MySQL.connect() question
References: <3893CD31.B17E8B84@regnoc.com> <870tjf$514$1@news1.xs4all.nl> <3894737C.A676BC16@regnoc.com>
Message-ID: <381BAA74.863A8048@cs.rmit.edu.au>


Try using "127.0.0.1" instead of localhost.

Hafeez

Jim Conger wrote:
> 
> Boudewijn,
> 
> Thanks you for the quick response.  I'm still missing something.  Here is my
> attempt to follow you example.  First I check that mySQL is running using
> mysqladmin, and then attempt to connect in python:
> 
> $ mysqladmin -ppassword processlist
> 
> | Id | User | Host     | db | Command  | Time | State | Info
> | 12 | jimc | localhost |    | Processes | 0 |    |     |
> 
> $ python
> Python 1.5.1...........
> >>> import MySQL;
> >>> self.dbConnection = MySQL.connect("localhost","jimc","password")
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> MySQL.error: Can't connect to local MySQL server
> >>>
> 
> Does mySQL need to be already connected to a DB?  If so how?  I'm sure it is
> obvious, but I can't figure it out!
> 
> Regards,
> Jim Conger
> www.regnoc.com
> 
> Boudewijn Rempt wrote:
> 
> > Jim Conger <jimc at regnoc.com> wrote:
> > > I have mySQL running on a RH Linux 6.1 box, and have added the
> > > MySQLmodule.so to my python library.  I can't figure out how to connect
> > > within python.  My attempts look like:
> >
> > >>>> import MySQL;
> > >>>> db = MySQL.connect();
> > > Traceback (innermost last):
> > >   File "<stdin>", line 1, in ?
> > > MySQL.error: Can't connect to local MySQL server
> > >>>>
> >
> > I'm not really surprised - I think you need to supply
> > some parameters:
> >
> >     hostname="essle"
> >                 username="boud"
> >                 password=""
> >                 databasename="languages"
> >     self.dbConnection = MySQL.connect(hostname,username,password)
> >           self.dbConnection.selectdb(databasename)
> >
> > Which is what I did when I used MySQLmodule; however, I've since
> > began using the DB-API II compliant MySQLdb, but I miss some of
> > the conveniences of MySQLmodule.
> >
> > --
> >
> > Boudewijn Rempt  | http://www.valdyas.org
> 
> --
> Jim Conger
> Project Manager
> Silicon Valley Oil Co.
> jimc at regnoc.com, 925-842-6729


From thamelry at vub.ac.be  Tue Jan 25 10:46:44 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 25 Jan 2000 15:46:44 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k78t$thg@newton.cc.rl.ac.uk>
Message-ID: <86kgh4$s2v$1@mach.vub.ac.be>

Richard Brodie <R.Brodie at rl.ac.uk> wrote:

[snip]

: C++ has to be the classic case of a language that evolved rather than
: was designed: PDP-11 assembler with OO bits tacked on. If we can
: learn something from its evolution, it should be not to do the same
: again.

C++ is a monster because of its origin (comptibility with C),not because
of the way it evolved. The evolution of python is quite similar as far as
I can see : people try out new stuff (e.g. Tismer's stackless python), some 
of it ends up in the official python distribution. 

Cheers,

---
Thomas Hamelryck    Institute of Molecular and Structural Biology
Aarhus University   Gustav Wieds Vej 10C
DK-8000 Aarhus C    Denmark                                                                        


From tim_one at email.msn.com  Wed Jan 19 23:54:04 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 19 Jan 2000 23:54:04 -0500
Subject: LISP Rocks! (was: Re: Python Rocks! :-)
In-Reply-To: <oqg0vtyj4y.fsf_-_@titan.progiciels-bpi.ca>
Message-ID: <000801bf6302$60316ec0$31a2143f@tim>

[Fran?ois Pinard]
> ...
> P.S. - SNOBOL was particularly uneasy to punch.  Multi-
> punches all the time.
> Do some remember? ;-)

Oh yes!  But typical SNOBOL4 programs were tiny compared to typical Pascal
programs, and at the time the latter was introduced, my school's keypunches
still lacked a semicolon key.  I have hated languages with semicolons ever
since <wink>.

semicolons-bad-whitespace-good-ly y'rs  - tim





From bparsia at email.unc.edu  Thu Jan 27 09:32:16 2000
From: bparsia at email.unc.edu (Bijan Parsia)
Date: Thu, 27 Jan 2000 09:32:16 -0500
Subject: Smalltalk blocks in Python
References: <388F8E03.389BC83A@acm.org>
Message-ID: <1e5243j.2z962642rjn6N%bparsia@email.unc.edu>

Thierry Thelliez <thierry at acm.org> wrote:

> I come from the Smalltalk world and I am trying to understand Python.
> 
> Is there a Smalltalk blocks equivalent in Python ?
> Java has a pale imitation with Inner classes.
> 

There are lambdas (which are a bit funky). But more importantly, you can
pass functions around (and functions don't need to be defined in a class
context). While there no general syntax (other than lambdas, which have
limitations) for specifying literal anonymous code blocks, you can
certainly use compile, exec, apply, eval, and friends (exec and eval
take code blocks as well as strings).

Now, the thing that *will* trip you up, IMHO, is the lack of class side
*methods* (though there are class attributes).

Cheers,
Bijan Parsia


From dfan at harmonixmusic.com  Wed Jan 26 17:36:56 2000
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 26 Jan 2000 17:36:56 -0500
Subject: To find a point inside a polygon
References: <388F0E36.6CE37135@ust.hk>
Message-ID: <wksnzk5uk7.fsf@turangalila.harmonixmusic.com>

David Yeung <dyeung at ust.hk> writes:

| Does anyone know if any python implementation for finding whether a
| point is inside a polygon? E.g., by giving a point(x,y) and an array
| of polygon's points (x0,y0,x1,y1,x2,y2,...), it returns true or
| false to tell if the given (x,y) is inside the polygon.

1) Draw a line from the point out to infinity.  The point is inside
   the polygon if it intersects an odd number of segments of the
   polygon.

2) (only works if the polygon is convex) Assuming the points go around
   the polygon counterclockwise, the point is in the polygon if it's
   to the left of (x0,y0)-(x1,y1), to the left of (x1,y1)-(x2,y2),
   etc.  Each of those tests is a simple operation involving a dot
   product or something.

-- 
Dan Schmidt | http://www.dfan.org


From matthewm at zebrasoft.co.nz  Mon Jan  3 05:44:38 2000
From: matthewm at zebrasoft.co.nz (Matthew Miller)
Date: Mon, 03 Jan 2000 23:44:38 +1300
Subject: Losing with spawmv() under win98
Message-ID: <38707D96.E2EA9EFE@zebrasoft.co.nz>

hi everyone,

i'm helping a friend with a y2k issue involving a dinosaur dos-based
program. here's the scoop:- a batch file executes, calling a main prog
which won't load if the date exceeds 1999. however the program works
fine if at startup time, the date is <2000. one can can change the date
after the program loads to any y2k date without problem. so, i've tried
a python script which calls the batch file using os.system('batchfile')
and then spawned a new process using os.spawnv(params) which includes a
time delay (many time values tried)  to give time for the main program
to load before trying to change the date. never works though. i've tried
all combos including spawing the batchfile with all P_variants and also
calling (via spawnv) another batch file aswell as a .py script to do the
date change thing. in all situations, the dinosaur batchfile executes
but no date change process is spawned/detached. all further processing
of date change just sits there quietly until the dinosaur dos-progarm
terminates. i've used fork in linux heaps of times with no problems.
what am i doing wrong in win98? sorry, can't copy my program to the
newsgroup as it's locked away in a factory until jan-13.thanks for your
help & suggestions.

--
Matthew Miller
Zebra Software Ltd.

E-mail: matthewm at zebrasoft.co.nz
Ph: 025-2920625 / FAX: 03-3842397




From bitbucket at isomedia.com  Mon Jan 10 18:12:53 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Mon, 10 Jan 2000 23:12:53 GMT
Subject: New Python Books
References: <Pine.LNX.4.10.10001090941020.26842-100000@crazypenguins.commandprompt.com>
Message-ID: <387a6490.779260998@news.isomedia.com>

On Sun, 9 Jan 2000 09:56:25 -0800 (PST), python at commandprompt.com
wrote:

>Hello,
>
>I recently posted this message to comp.os.python:
>
...
>Last but definitely not least, no matter what we do, if we publish a
>Python book, 20% to 40% of the Gross Profits would go back to the PSA.
>

I wonder if it wouldn't be better, should y'all make such a book, to
price it 20% to 40% below what one would expect.  I apologize for
appearing to volunteer away PSA's money (which I couldn't do, because
I don't even know the PSA), but when it comes to forking out dollars
for documentation, I always find myself worried if I'm getting "the
exact right book," because I can generally only justify buying one at
the typical prices.  

One of the unfortunate realities of documentation is that no single
source is everything you need (with a few 65-lb exceptions).  So life
might be better if the Python newbie could afford _two_* pieces of
dead-tree documentation.

     -Eugene

* Yeah, I know that 20%-40% off a single book doesn't double your
purchasing power.  So just assume that the average newbie can
typically afford 1.6 to 1.8 doc books, and ...
import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From news at dorb.com  Mon Jan 17 13:10:40 2000
From: news at dorb.com (Darrell)
Date: Mon, 17 Jan 2000 13:10:40 -0500
Subject: Read In Directory Contents
References: <388358F1.F1771D82@netins.net>
Message-ID: <WTIg4.6001$U5.216258@typhoon.nyroc.rr.com>

>>> import glob
>>> glob.glob('*.py')
['Mscomm32.py', 'comct332.py', 'except.py', 'dicMap.py', 'serial.py']

--
--Darrell
"Scott Billings" <aerogems at netins.net> wrote in message
> For a program I'm working on, it would be very useful if I could use an
> expression such as:
>
> list = os.getcwd() + "*.txt"
>
> to have the names of all text files in the current dir placed into the
> list. However, when I try this method, I get an error about no such file
> existing.
>




From jbauer at rubic.com  Fri Jan  7 19:47:57 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 07 Jan 2000 18:47:57 -0600
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net> <slrn87cvv9.gcu.kc5tja@garnet.armored.net>
Message-ID: <3876893D.876A8949@rubic.com>

>>>"Samuel A. Falvo II" wrote:
>>> (Personally, I think the COM solution is a superior one
>>> anyway, but that's just my opinion.)

>> Sure it is -- but compatibility is even better.

> Then why aren't we all using CP/M?
>
> There is a point of diminishing returns when it comes to 
> this whole "backward compatibility" thing.  Sometimes you 
> just have to pick up your bags, and leave Hotel Compatibility 
> behind.  Time marches on, after all.

I'm with you, Samuel.  When we can leave this whole COM
thingy behind, all will be sweetness and light.

> IE 5 is 100% more stable than any Netscape version 
> I've ever used.

When I tried to run Explorer on Solaris, I discovered
why people called it "Exploder". ;-)

On a more serious note -- and as a matter of genuine 
interest -- are there any Python programmers using COM 
in mixed Unix and Windows environments?

Jeff Bauer
Rubicon Research



From drew.csillag at starmedia.net  Fri Jan 21 16:19:19 2000
From: drew.csillag at starmedia.net (Andrew Csillag)
Date: Fri, 21 Jan 2000 16:19:19 -0500
Subject: DBI Commentary
References: <1263695833-12432931@hypernet.com> <869v1t$a2m$1@nntp9.atl.mindspring.net>
Message-ID: <3888CD57.1BD59385@starmedia.net>

Aahz Maruch wrote:
> 
> In article <1263695833-12432931 at hypernet.com>,
> Gordon McMillan <gmcm at hypernet.com> wrote:
> >Timothy Grant writes:
> >>
> >> Aahz made the very valid point that specifying columns in your query
> >> *guarantees* tuple return order. I had not fully appreciated that, and I
> >> have put that into use in a number of places. However, I have to agree
> >> with Lance when he says that "SELECT * FROM table" is not as often a bad
> >> thing as Aahz claimed.
> >
> >That was me, not Aahz.
> 
> Actually, it was me, too.  Timothy made the boo-boo of cc'ing me in
> e-mail with no indicator, so he probably saw my response first.  He
> later responded to that e-mail publicly.
> 
> >I've used the trick myself - but it was for building HTML tables
> >to display the results from an ad-hoc query built from an HTML
> >form. I personally would never use "*" if I new what columns I
> >needed.
> 
> Yeah; as I told Timothy, if managing the fields is a problem, create a
> list and use string.join() to build the SQL string.  It's easy enough to
> extend that to dict-based access, which you can use to build your own
> SQL wrapper library/class.

There's also cursor.description which if you do a SELECT * FROM ... will
tell you the names of the columns so you can reorder them yourself (or
do as I do and often turn the result list into a list of dictionaries
keyed by the downcased column name).
-- 
print(lambda(q,p):'pmt:$%0.2f'%(q*p/(p-1)))((lambda(a,r,n),t:(a*r/
t,pow(1+r/t,n*12)))(map(input,('amt:','%rate:','years:')),1200.0))



From robin at jessikat.demon.co.uk  Wed Jan 26 07:19:39 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 26 Jan 2000 12:19:39 +0000
Subject: Binary Numerical module
References: <388EDE90.661499FF@obop.com.pl>
Message-ID: <JqoDdCAbZuj4EwCL@jessikat.demon.co.uk>

In article <388EDE90.661499FF at obop.com.pl>, =?iso-
8859-2?Q?Przemys=B3aw?= G. =?iso-8859-2?Q?Gawro=F1ski?=
<gawron at obop.com.pl> writes
>Is there a binary distribution of Numerical module ?
>I don't have VC++ installed and actually hope I can avoid doing so.
>
>Thanks Przemek
>--
>Przemyslaw G. Gawronski UIN:8358522
>mailto:gawronskip at usa.net
>mailto:gawron at obop.com.pl
>
>
>
I have a VC++ 6 compiled version; I would be happy to submit it to NumPy
at sourceforge.
-- 
Robin Becker


From chuck.meyers at lmco.com  Mon Jan 17 18:05:40 2000
From: chuck.meyers at lmco.com (Chuck Meyers)
Date: Mon, 17 Jan 2000 15:05:40 -0800
Subject: Can TkInter be frozen?
Message-ID: <3883A044.F60D145F@lmco.com>

Can TkInter be frozen?
I understand the answer used to be no, has it changed?
Is it expected to change?



-Thanx

chuck.meyers at LMCO.COM



From wtanksle at hawking.armored.net  Thu Jan 20 19:06:58 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 21 Jan 2000 00:06:58 GMT
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000120175408.21484B-100000@sunse>
Message-ID: <slrn88f9e2.4lu.wtanksle@hawking.armored.net>

On Thu, 20 Jan 2000 17:57:00 +0300 (MSK), Evguenii Smogailov wrote:
>On 20 Jan 2000, Mitchell Morris wrote:
>> Except that Perl considers both '==' and 'eq' to be operators, and they
>> do different things.

>Yes, i know, but i'm not sure yet that it's a good idea for 'eq' to be 
>different from '==' in Python

Good -- it's not.  Currently 'eq' has exactly the same behavior as '==' in
Python; it's even spelled exactly the same.

>evguenii

-- 
-William "Billy" Tanksley, in hoc signo hack


From dworkin at ccs.neu.edu  Mon Jan 10 13:49:04 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 10 Jan 2000 13:49:04 -0500
Subject: getservbyport() in Python?
In-Reply-To: jeffg@provenance.com.au's message of "Mon, 10 Jan 2000 07:54:56 GMT"
References: <85c38c$ksb$1@nnrp1.deja.com>
Message-ID: <vndogatiwyn.fsf@betelgeuse.ccs.neu.edu>

jeffg at provenance.com.au writes:

> Does anyone know of any Python implementation of the Linux standard call
> getservbyport()? This function is given a network socket port number and
> returns the service name. eg if passed 80, will return 'http'. This
> information is taken from the /etc/services file.

No, it seems as though getservbyname exists, but its sister calls are
unimplemented.  

And you would probably want to be more specific, saying that passing
80 and 'tcp' would return http.

> If nothing exists (& I've looked around pretty thoroughly), then I plan
> to scan the services file directly, using re to search for the port
> number I want.

You may be better off using `string' for that searching, not `re'.

Of course, the _real_ way to do this is to patch socketmodule.c to
call the underlying getservbyport function.  It should be pretty much
a cut-and-paste job from the getservbyname implementation.

-Justin

 



From adjih at crepuscule.com  Fri Jan 21 12:05:25 2000
From: adjih at crepuscule.com (Cedric Adjih)
Date: 21 Jan 2000 17:05:25 GMT
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com> <867t9d$aoi$1@mordred.cc.jyu.fi> <8680qp$e9t@news.or.intel.com>
Message-ID: <86a3kl$9cj$1@ites.inria.fr>

tye4 <tye4 at yahoo.com> wrote:
> Sami Hangaslammi <shang.remove_edu at st.jyu.fi.edu> wrote in message
> news:867t9d$aoi$1 at mordred.cc.jyu.fi...
>> > I intend to modify the language's syntax to add
>> > 'end' delimiters for the blocks.
>>
>> Wouldn't it be better (and easier) to write a script that converts
>> between .py and .pye (python with end delimeters) scripts. This would
>> allow you to both use existing Python libraries and distribute modules
>> you have made to others.
>>
>> --
>> Sami Hangaslammi
>> shang (at) st (dot) jyu (dot) fi
>>
>>
> Brilliant idea. Thanks. That's a lot less work than I had planned.
> -tye4


  There is already one in Tools/scripts/pident.py that does something
similar.
Alternatively you could do something like:
endFor=0 ; endfor=0 ; end=0 ; end_for=0

And write code such as follows to your heart content:

for i in range(0,100):
	xxxx
endfor

- or

for i in range(0,100):
	xxxx
endFor

- or

for i in range(0,100):
	xxxx
end_for

- or

for i in range(0,100):
	xxxx
end

-
same applies with endwhile, enddef, endclass, endtry, ...

-- Cedric


From hnezic at zamir.net  Sat Jan 22 11:45:24 2000
From: hnezic at zamir.net (Hrvoje Nezic)
Date: Sat, 22 Jan 2000 17:45:24 +0100
Subject: [Newbie] PythonWin - keyboard problems, please help!
Message-ID: <86cn53$432$1@as102.tel.hr>

Hi all,

I am new to Python, but I am convinced that it will help me a lot in my
projects :)

Command-line Python works very well, but I have a problem with PythonWin
IDE.

Namely, I don't use the English keyboard locale, but when I start PythonWin
it changes the keyboard locale to English, and this results in a mismatch
between program's keyboard arrangement, and real keys on my keyboard.

Can this be changed somehow? Because if not, the PythonWin program is
virtually unusable for me.

Any help would be highly appreciated.

Best regards,
Hrvoje Nezic





From tim_one at email.msn.com  Tue Jan 25 18:17:06 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jan 2000 18:17:06 -0500
Subject: using threads with python and winnt
In-Reply-To: <86k2rl$92k$1@nntp9.atl.mindspring.net>
Message-ID: <000a01bf678a$4bfd4320$862d153f@tim>

[posted & mailed]

[Tim]
> There's an exceedingly rare timing hole in the released 1.5.2
> Python (fixed in the CVS version) ... showing up in rare reports
> under extremely high rates of thread death and only on top-end
> >multiprocessor systems.

[Aahz Maruch]
> We're still using 1.5.1 (no comments, please! we're supposed
> to finally upgrade this week),

The problem also exists in 1.5.2.

> and I was wondering if you could explain what the symptom of this
> bug is, because we've seen a couple of recent cases where a
> heavily threaded app that creates hundreds of single-shot
> threads dies with a thread error.

The rare msg:

    Fatal Python error: PyThreadState_Delete: invalid tstate

or, even more rarely, a memory fault.  See

http://www.python.org/python-bugs/resolved?id=12;user=guest

for a bit more.

> (We're *also* fixing it to use a thread pool,

That would be an effective workaround; the error is triggered by a race
among threads that are competing to die.  It's like the door to the funeral
home is only wide enough for one, but sometimes two can sneak thru, and then
the casket gets overloaded and everyone at the funeral freaks out when it
bursts open <wink>.

> but I'd still like to know what our chances are of hitting
> this with the stock 1.5.2)

Inevitable, if you repeatedly kill off enough threads over a long enough
time, although the mean-time to failure in non-extreme programs (a dozen
threads spawned every minute or so) would likely work out to no less than
years.  It's much easier to trigger on a true multiprocessor (on a
uniprocessor, a dying thread has to lose its (OS) timeslice in a window of
just a few machine instructions, at the same time another dying thread gets
its timeslice in another window of a few, and they both need to have reached
exactly the same spot in a traversal of a long linked list).

threads-are-unforgiving-ly y'rs  - tim





From parkw at better.net  Fri Jan 28 11:34:01 2000
From: parkw at better.net (William Park)
Date: Fri, 28 Jan 2000 11:34:01 -0500
Subject: vim and tabs in Python code
In-Reply-To: <86sc2v$8nk$1@nnrp1.deja.com>
References: <86sc2v$8nk$1@nnrp1.deja.com>
Message-ID: <20000128113401.A3085@better.net>

On Fri, Jan 28, 2000 at 03:20:06PM +0000, super_banane at yahoo.com wrote:
> <blank><blank><blank><blank>FirstLineOfCode
> <one 4-blank tabulation>SecondLineOfCode

You can convert your file using, in this case,
    :retab! 4
You can automates using 
    :au BufRead *.py retab! 4

See ':help retab'.

William



From jgreve at directobjects.com  Wed Jan 19 00:58:28 2000
From: jgreve at directobjects.com (John Greve)
Date: Wed, 19 Jan 2000 05:58:28 GMT
Subject: COTUG (Cleveland Object Technology Users Group) meeting - Feb 3rd, 2000 (6:30pm-8:30pm)
Message-ID: <388b524c.22890388@news.oh.verio.com>

             Overview
     Object Oriented Technology's** rapid evolution makes
it hard to keep on top of emerging trends, tools, and best
practices. OO Professionals and Beginners alike are welcome
to attend COTUG (Cleveland Object Technology Users Group).
     Each meeting has a Main Presentation on contemporary
issues of interest to all OO practitioners.  Then we split
into independent Focus Tracks (think of them as sigs) to
address language / product specific issues.


             Meeting Data
main topic: Applied Patterns
date: Thursday, Feb. 3rd (first* Thursday of each month)
time: 6:30pm - 8:30pm...
dress: casual
location: WoodHawk Club Apts (in the ClubHouse)
   Fox Hollow Drive
   Mayfield Heights, OH  44124.
   (see bottom for directions)
*pushed back to the 2nd Thr. if 1st Thr is a holiday.

contacts:
	John Greve jgreve at directobjects.com
              (440-646-9558 evenings)

	David Mitchell dmmitchell at mindspring.com
	


Agenda
   group business (officers, schedules, etc.) - 15 min
   member presentations (book review, project report)  - 15 min
   main topic - 60 min
   focus tracks (C++ track, Java track, Smalltalk track, etc) - 30 min
        
---------------------------------------------
         CALL FOR...
We're always looking for people to share at the meeting.
Please send email about your ideas, or better yet come
to the meeting.

call for Book Reviews - if you've come across a good book
for OO practitioners recently, we want to hear about it.  

call for Product Reviews - if you've had interesting results
(positive or otherwise) with an OO product, we want to
hear about it.

call for Main Topic - if you have an idea for a presentation
of general interest to the OO community, we want to hear
about it.

call for Focus Track - if you have an idea for a Focus Track,
please come.
   note: the label SIGS is widely publicized in Cleveland's
   computer arena for non-OO purposes, so we use F.T.

call for Vendor Presentations - if you have an OO technology
or application (preferably language independent) you'd like
to present to the group, please email us.
---------------------------------------------

Directions:  Take 271 to the Mayfield Rd exit.
Go EAST (0.2 miles) to 91 (Som Ctr Rd).
Turn RIGHT onto Som Center Rd/91 (0.8 miles).
Enter the roundabout and take 1st exit onto WoodHawk Dr (0.6 miles).
The WoodHawk complex is located on the right near the end of
WoodHawk Dr.  Ask at the gatehouse for the COTUG meeting.

**yes, we talk about Components, too.


From pbible at home.com  Fri Jan  7 14:26:22 2000
From: pbible at home.com (Paul E. Bible)
Date: Fri, 07 Jan 2000 19:26:22 GMT
Subject: URL Domain Name
Message-ID: <y5rd4.710$hJ1.10236@news.rdc1.tx.home.com>

Anyone know how to get the Domain Name for a given IP Address??  (In Python,
of course)

Thanks in advance,
Paul




From helge at mdlink.de  Thu Jan  6 16:02:29 2000
From: helge at mdlink.de (Helge Hess)
Date: Thu, 06 Jan 2000 22:02:29 +0100
Subject: no return values for __init__ ??
References: <1264936306-6137156@hypernet.com>
Message-ID: <387502E5.F37B5FF5@mdlink.de>

Gordon McMillan wrote:
> Helge Hess writes:
> 
> What happens when a derived class chains to the base class
> __init__ and the base class __init__ changes "self"?

Good question. Depends on what one tries to accomplish with different
return values, usually it would look like:

  def __init__(self...
     s = super.__init__(self...
     if (s is not None) & (s is not self): return s
     self.a = 5
     return self

Note that I don't want to make this feature common usage. It's useful in
a relativly small problem domain, like clusters where the additional
code as shown above is acceptable.

The one real 'inconsistency' with my approach is that None has the
special meaning of self in __init__. This is no technical requirement
but would break compatibility if done otherwise (any plain __init__
would be required to return self instead of None or the default return
value would need to be changed to 'self').

BTW: I also always wondered why Python has None as the default return
value. 'self' seems better to me as one could make compound method calls
without additional work while None has no benefit I can see (maybe
better RC behaviour ?).

> How about MI?

This is related to the above.
There is no conceptual difference between __init__ returning a value and
any other method returning a value. Eg if you have 

  class A(B,C)

you need to define order explicitly. Same as above:

  s = B.__init__(self..)
  if s is not self: return s
  s = C.__init__(self..)
  if s is not self: return s

This is exactly the same like with any other method and MI.

> Make it obvious: rename the factory function to "getC()" and
> rename "_C" to "C".

Didn't get my complaint with your func. I would like to have a
consistent & simple environment. Eg in the cluster example the developer
doesn't care and doesn't need to know about the fact that he get's
returned different instances.

Eg the user just says:

  s = String("sfhakasfhad", encoding='utf8')
  -> s instanceof Utf8String(String)
  s = String("adfhasfdj", encoding='utf16')
  -> s instanceof Utf16String(String)
  s = String("asfhjasd")
  -> s instanceof CString(String)
  s = String("symbol", intern=1)
  -> s instanceof InternedString(String)

he doesn't care about private subclasses of String and he doesn't care
about special factory functions. He just want's to create an object
compliant to the String class.

> If really you want to customize the class, you can assign to
> the instance's __class__ attribute.

Ok, this is getting much nearer to my goal although it is much less
convenient. Eg this would probably break abstraction since the state of
the object needs to be converted. Returning a new object seems much
easier and more consistent to me.

> > This is called a 'class-cluster' in Objective-C and has quite
> > some advantages. Another useful application for clusters are
> > string-objects (different classes for different char widths, eg
> > Py8bitString, PyUnicodeString, ..)
> 
> And Fred's pattern is called a "factory" in a wide variety of
> languages.

Goes in the right direction. A cluster is a special kind of factory
where the produced objects are subclasses of the factory therefore
making it look very convenient for the user, see the String example
above. The discussed feature has the advantage that the user isn't
required to know about the factory, that is, it is abstract.

> Hackishness is in the eye of the beholder.

Ok ;-)

> Having "self" be a complete object when __init__ runs  means
> the language doesn't have to come up with a bunch of arcane
> rules about default constructors, order in which base class
> constructors are called etc.

I don't understand what you are trying to say. We are talking about
return values. In every invocation of __init__ the object is as complete
as it can be.

> It's what makes MI useful in Python (vs a feature that is nearly
> impossible to use, as in some other languages).

My major point is that __init__ is a method like any other and should be
treated as such. The issues related to MI and return values is *exactly*
the same as for any other MI method.
There isn't even a difference between the state as it is now since no
__init__ methods are called explicitly, so the order needs to be defined
manually in any case.

best regards
  Helge



From mal at lemburg.com  Tue Jan 18 17:33:58 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 18 Jan 2000 23:33:58 +0100
Subject: procmail-like spamfilter ?
Message-ID: <3884EA56.4A3C29C8@lemburg.com>

Does anyone know of a procmail like spam filter that I could
use to prevent all these spam mails arriving at my mail box ?

Ideally it should be written in Python, of course :-) Snakes
eating spam... you know.

Thanks,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net  Sun Jan 23 15:45:13 2000
From: 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net (24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net)
Date: Sunday, 23 Jan 2000 14:45:13 -0600
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
Message-ID: <23010014.4513@adsl-63-197-143-124.dsl.scrm01.pacbell.net>

My name is Darren Mulligan and I am an aspiring photographer from California.
I will get right to the meat and gravy of this post, so as not to waste your time!

+++++WHAT AM I SEEKING?+++++

I am looking for young models (prefer early teen 14-16 year old female) for nude and semi-nude photography. This is a great way for me to update my portfolio, and also a great way for you to get a free portfolio, too! This could lead to employment in the Adult XXX entertainment business. What a great way to make money!

Please respond to one of the following e-mail addresses. I run my very own mail server on an ADSL connection so you can be assured to complete discretion and privacy. If one e-mail address does not work, simply try the next! 

Sorry for the long list! My ADSL connection goes down from time to time, so I posted all the e-mail addresses I can be reached at. If one does not work, just move on to the next!

nanae at adsl-63-197-143-124.dsl.scrm01.pacbell.net
usenet at adsl-63-197-143-124.dsl.scrm01.pacbell.net
24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
news.admin.net-abuse.email at adsl-63-197-143-124.dsl.scrm01.pacbell.net
spamlinkpage at adsl-63-197-143-124.dsl.scrm01.pacbell.net
alt.make.money.fast at nntp.8m.com  nanas at nntp.8m.com   
  hostmaster at FREESERVERS.COM

THIS IS NOT A SCAM! IT IS NOT CHILD PORN!!
There is nothing wrong or illegal about the artistic representation of a young body. If you have a problem with this post, please e-mail me and I will not post to your newsgroup again!

Sincerely,
Darren Mulligan



From Oliphant.Travis at mayo.edu  Wed Jan 12 20:55:17 2000
From: Oliphant.Travis at mayo.edu (Travis Oliphant)
Date: Wed, 12 Jan 2000 19:55:17 -0600 (CST)
Subject: New i386 and SRC RPM's for Numerical Python Version 14
Message-ID: <Pine.LNX.4.10.10001121952220.5694-100000@us2.mayo.edu>

This is just a quick note to announce the availability of RPM's for
Numerical Python Version 14 at http://oliphant.netpedia.net

In this release Gist and RNG have been separated out.

I have applied some patches to the LLNL Version 14 Release which alleviate
some of the single-precision difficulties and fix a segmentation fault
bug.  The patch is also available here:

http://oliphant.netpedia.net/packages/Numerical-14.patch



--Travis





From just at letterror.com  Sun Jan 23 07:51:13 2000
From: just at letterror.com (Just van Rossum)
Date: Sun, 23 Jan 2000 13:51:13 +0100
Subject: Win32Installer + Numeric, anyone succeeded?
In-Reply-To: <CC59868A0A76D311B1F50090277B321C1F3096@VRTMAIL>
Message-ID: <l03102801b4b0a83fdeb3@[193.78.237.179]>

At 7:12 AM -0500 23-01-2000, Mike Fletcher wrote:
>I've been trying to get a Win32Installer standalone binary created for an
>application I've been working on.  However, the application dies with a
>memory read failure on what looks like a null pointer when trying to import
>multiarray.pyd (which the win32installer has placed in the target
>directory).  This is the standard multiarray.pyd from the LLNL distribution
>and the "f" release of win32installer.  Has anyone successfully created a
>standalone programme which includes Numeric?

Maybe this is related to a problem I had doing a similar thing on the Mac:

Numeric relies on _numpy.pyd, but _numpy was not imported in any Python
code, just C code, so freeze's modulefinder (which the Win installer also
uses I think) didn't notice it. And Numeric indeed crashes horribly when
the _numpy shared lib isn't available. I think this was later fixed by a
dummy "import _numpy" somewhere. Maybe you can check whether _numpy.pyd
gets included at all, and if your version of Numeric includes the dummy
import (in Numeric.py I believe).

Just





From mwh21 at cam.ac.uk  Sat Jan 15 05:11:50 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 15 Jan 2000 10:11:50 +0000
Subject: Help me improve my Python Skills, please
References: <m2g0vzsrte.fsf@excalibur.peernet>
Message-ID: <m3u2kfabkp.fsf@atrus.jesus.cam.ac.uk>

tarude at whitestar.localdomain (Trebor A. Rude) writes:

> I just started picking up Python a few days ago, and am rather
> enjoying it.  To help me learn it, I converted one of my old TCL
> scripts into Python, which I've attached below. However, this being my 
> first serious python script, I'm pretty sure I'm doing some things
> less efficiently than necessary, through lack of familiarity with the
> library and language in general. So I'd like to enlist the group's
> help in improving the code. Logic, style, better ways of doing things, 
> whatever.  Thanks in advance.
> 
> A little explanation before the code.  The script recursively descends 
> a directory structure, building a .jdb (Jukebox database) file for
> each directory based on the mp3 tag info of the mp3s in that
> directory and all its subdirectories. It also repairs damaged comment
> fields caused by a slightly buggy encoder program.

First imporessions: you're reinventing the wheel a lot. This is fair
enough, as you don't yet know what kind of wheels come with Python,
but after a while you should get the idea of the sort of things that
come built in to the standard libraries.

Also, you're using re where string would do. This is also pretty
common... remember the Tom Christiansen (I think) quote:

"When faced with a problem, soem people think 'I'll use regular
expressions.' Now they have two problems."

Also, if I were writing this, I'd break the task up more and have
fewer smaller functions. (or methods. see below).

> #!/usr/bin/env python
> 
> # Does this look strange? My basic philosophy on
> # using "import" vs. "from <foo> import <bar>" is
> # based on my conception of how easy it might be
> # to misinterpret the meaning of the function. Also,
> # if I need a bunch of names from a package, I'll
> # tend toward "import".
> 
> import os
> import re
> import select
> 
> from os.path import join, basename, isdir, splitext

There's string.join as well, which is one reason to leave
out-of-module references qualified.

> from glob import glob
> from fnmatch import fnmatch
> from operator import concat
> from commands import getoutput
> 
> # Only reason this is a function is because it gets called recursively.
> 
> def Make_Playlist ():
>     pwd = os.getcwd()
>     filename = join(pwd, basename(pwd) + '.jbd')
> 
>     try:
>         listfile = open(filename, 'w')
>     except IOError:
>         sys.exit("Cannot open " + filename + " for write: " + IOError)
> 
>     # Since sort() returns None, need to go through these extra
>     # steps instead of just putting all on the for line.

Hmm; if you don't care about the order records turn up in the file,
why sort the list? OTOH, you process directories mixed up with files,
so the order's going to be a little odd anyway.

>     files = glob(join(pwd, '*'))

There's os.listdir(os.curdir), which is probably easier...

>     files.sort()
>     
>     for file in files:
>         if isdir(file):   # Recursively descend directory.
>             os.chdir(file)
>             Make_Playlist()
>             os.chdir(pwd)

This use of chdir is yucky (IMHO). I'd make the directory a parameter
to Make_Playlist.
 
>             # Concatente the just-created .jbd onto the current one.
>             
>             infile = open(join(file, basename(file) + '.jbd'))
>             listfile.writelines(infile.readlines())
>             infile.close()

This makes me cringe a bit too.

>         elif fnmatch(file, '*.mp3'):
>             newfilename = re.sub(' ', '_', basename(file))

newfilename = string.replace(basename(file),' ', '_')

>             if (newfilename != basename(file)):
>                 os.rename(file, newfile)
>                 file = join(pwd, newfile)
> 
>             # Don't really need this since getting a better encoder,
>             # but still might want to make it optional.
>             
>             #artist = re.sub('_', ' ', basename(pwd))
>             #title = re.sub('_', ' ', splitext(basename(file))[0])
>             #os.system("mp3info -t " + title + " -a " + artist \
>             #+ " -g 1 " + file)
> 
>             # Yes, I really do have filenames which use all the
>             # "bad for the shell" characters below.
>             
>             escaped_filename = re.sub('(["\'()& ])', r'\\\1', file)

In these circumstances I'd consider using one of the
os.exec{l,v}{p,}{e,} functions and not letting a shell get near the
filenames. This is more work (more mucking with filehandles than it's
worth, maybe).

You've forgotten '{}$' btw. Or are these not a problem?

You can make string.translate do this, but it's a bit unwieldy.

>             # Deleted the actual format because it was too long.
>             # If anyone's interested, just ask.

This would be why the next line is a syntax error, yes?

>             mp3info = getoutput('mp3info -f "..." + escaped_filename)
> 
>             # The encoder program I use is slightly buggy and tends to 
>             # leave spaces followed by one random control character at 
>             # the end of the comment field of the mp3 tag. Generally,
>             # this isn't a problem, but it IS if that control
>             # character is ^J, so I fix it here.
> 
>             # Notice the (\B|\b) construct.  What I really wanted to
>             # do was make the whole of \1 optional, but then it causes 
>             # an error if \1 doesn't match, which is bad because the
>             # point is to delete the stuff after it.  Is there a
>             # better way to do this?
> 
>             fixedinfo = re.sub(\
>             r'\|((\B|\b)([A-Za-z \'/.]*[A-Za-z])?) +[\1-\26]\|', \
>             r'|\1|', \
>             reduce(concat, mp3info))

Hello? mp3info at this point is a string. So reduce(concat,mp3info) is
a no-op. If by something you've omitted mp3info is a list of strings,
"string.join(mp3info)" is a better bet (clearer & faster - what more
could you want?).

Also string.rstrip *might* do what you want.

> 
>             # Hmm, come to think of it, I might want to repair the
>             # actual mp3 file here.  I'll have to look into the
>             # "Matcher" objects. (Would that be the right way to save
>             # \1 from above?)
> 
>             listfile.write(fixedinfo)
>             listfile.write('\n')
>             
>     listfile.close()
> 
> # os.path.walk() might simplfy things mightily, but would require some
> # information to be saved outside the function to keep track of
> # just-constructed .jbd files for inclusion in the current one.

That's what classes and bound methods are for...

> Make_Playlist()
> 

You're aware of the 

if __name__ = '__main__':
    main_func()

idiom? You never know, you might want to use this script as a module
one day...

Let's have a go at this.

A little top-down design:

The top function:

def descend_directory(dir): ...

I think I'll have this return a list of info strings.

We want a function to extract the info (fixed if necessary) from an
.mp3:

def get_info(mp3file): ...

I'll let you write that one.

Then descend_directory looks somewhat like this:

def descend_directory(dir):
    mp3s = glob.glob(os.path.join(dir,"*.mp3"))
    
    info = []
    
    listfile = open(os.path.join(dir, 
                                 os.path.basename(dir) + '.jbd'),"w")

    for mp3 in mp3s:
        info.append(get_info(mp3file))

    dirs = filter(os.path.isdir,os.listdir(dir))

    for ndir in dirs:
        info = info + descend_directory(ndir)

    listfile.writelines(info)

    return info

Top your module with:

import os,glob

and tail it with:

if __name__ = '__main__':
    descend_directory(os.curdir)

and you should be sorted!

I know I've written a lot, but I think you're doing pretty well!

HTH,
Michael


From junkster at nospam.rochester.rr.com  Thu Jan 27 23:02:08 2000
From: junkster at nospam.rochester.rr.com (Benjamin Schollnick)
Date: Fri, 28 Jan 2000 04:02:08 GMT
Subject: Win NT Workstation (v4, sp #5) PWS Unable to use Python?
Message-ID: <rNHvjEdhm5Pp-pn2-UxVidzkzz7KW@d185d186f.rochester.rr.com>

Folks,

	I just changed my "Python" development machine, and even
though it's configured "identically" (as far as I can tell), PWS 
doesn't want
to allow me to deal with python scripts.

	I've tried Aaron Watters instructions for installing python for
IIS, via REGEDIT....But even after adding the reg keys it doesn't 
work.

	I finally uninstalled PWS, reinstalled it with the IIS manager, and
added .py extension info via IIS Manager.  That seems to work, but 
(!!)
if I have just "c:\progra~1\python\python.exe" the task just "sits" 
there until
it times out.

	If it's "c:\progra~1\python\python.exe -u %s %s" , the result is 
always
"python.exe: unable to open <filename>".

	Anyone have any suggestions?

			- Benjamin

====================================================
          (Remove "NoSpam" to Email me)
====================================================
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 abaitam at cyberia.net.lb  Sun Jan 16 15:11:03 2000
From: abaitam at cyberia.net.lb (Ali Baitam)
Date: Sun, 16 Jan 2000 20:11:03 GMT
Subject: IDLE for Python on Linux
Message-ID: <948053871.299655@opti.cyberia.net.lb>

hi,
I couldn't find the IDLE dev. env. in the Python RPM for Linux. Can any one
please tell me where I can download it.

thanks




From gvwilson at my-deja.com  Sat Jan 29 09:33:03 2000
From: gvwilson at my-deja.com (gvwilson at my-deja.com)
Date: Sat, 29 Jan 2000 14:33:03 GMT
Subject: concurrency in Python
Message-ID: <86utmv$1va$1@nnrp1.deja.com>

Hi.  I'm interested in hearing from anyone who's ever built a
concurrency toolkit for Python that used a model other than
threads + locks --- futures, tuple spaces, active expressions,
concurrent aggregates, actors, or anything else.

Thanks,
Greg (gvwilson at nevex.com)


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


From J.Knuth at ipro.de  Mon Jan 10 08:55:48 2000
From: J.Knuth at ipro.de (Jochen Knuth)
Date: Mon, 10 Jan 2000 14:55:48 +0100
Subject: Zope and NT databases
References: <387956BD.CFDF17DD@bby.com.au> <3879CE9B.BE77787C@bam.com>
Message-ID: <3879E4E4.B4BD96FD@ipro.de>

Hi,

Bill Scherer wrote:
> 
> You want ZODBCDA:
> 
> http://www.zope.org/Products/DA/ZODBCDA
> 
> Sarah Burke wrote:
> 
> > Hello...
> > I want to connnect up to a NT Access 97 Database through Zope running on
> > Unix. Is the mxODBC module in an External Method the way to go?
> > Thanks in advance
> >

if i remember correctly ZODBCDA only works on Win32 platform. 

also Access 97 is not a client/server database, so you cant connect to
it from unix directly to it.

perhaps the ODBC Socket Server from http://odbc.linuxbox.com/index.html
will work (haven't tried it). 

You have to write a python client for the ODBC Socket Server (if noone
else has written it, don't know).

Ciao,
Jochen
-- 
--------------------------------------------------
Jochen Knuth          WebMaster http://www.ipro.de
IPRO GmbH             Phone ++49-7152-93330
Steinbeisstr. 6       Fax ++49-7152-933340
71229 Leonberg        EMail: J.Knuth at ipro.de


From mikael at isy.liu.se  Fri Jan 21 03:46:02 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Fri, 21 Jan 2000 09:46:02 +0100 (MET)
Subject: Now it is getting wacky was( Re: A modest Error (I mean Prop
In-Reply-To: <p8Kh4.2347$L4.74646@newsread2.prod.itd.earthlink.net>
Message-ID: <XFMail.000121094602.mikael@isy.liu.se>

On 20-Jan-00 Arinte wrote:
 >  Just to test len.  And now my program works???? If I remove this line or
 >  even just remove the print it doesn't work???
 >  Now what ??? If anything.  Another not here that critical print statement
 >  doesn't print the len of the string either...

It sure is wacky! I think I'm getting lost here. What does it print? 
Just some other arbitrary integer or somthing completely different?
Both len and type seems redefined. What does len and type do before 
and after you call the function the first time and before and after 
you call your function the second time? 

Just to be sure: Are you doing something like

  from foo import *

somewhere?

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    21-Jan-00
Time:    09:38:58

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From wweston at worldnet.att.net  Fri Jan 14 12:14:51 2000
From: wweston at worldnet.att.net (wes weston)
Date: Fri, 14 Jan 2000 17:14:51 +0000
Subject: Tcl/Tk Path problems in Python 1.5.2 installation
References: <387F80E5.A0AF82A5@sourcesignal.com>
Message-ID: <387F598B.9F448DC6@worldnet.att.net>

David,
	I just had a heck of a time doing something similar. On NT, I had an
old version of tcl - which I removed. After that, Tk quit working. I
found the problem by doing a find *.dll in my tcl tree. That yielded
about six dll's. About four of these were also in winnt/system32. The
one causing the problem was tk80.dll.
		wes

David Nichols wrote:
> 
> I just installed python 1.5.2 (running py152.exe and answering 'yes - I
> want to install Tcl/Tk') and have been unable to get the path set
> up correctly for graphical applications.  For instance, if I go to the
> idle directory (//E/Programs/Py152/Tools/idle) and try to launch it I
> get the following result.
> 
> ============================================================
> BASH.EXE-2.02$ python idle.py
> Traceback (innermost last):
>   File "idle.py", line 3, in ?
>     PyShell.main()
>   File "E:\Programs\Py152\Tools\idle\PyShell.py", line 611, in main
>     root = Tk()
>   File "E:\Programs\Py152\Lib\lib-tk\Tkinter.py", line 886, in __init__
>     self.tk = _tkinter.create(screenName, baseName, className)
> TclError: Can't find a usable init.tcl in the following directories:
>     {} ./lib/tcl8.0 . E:/Programs/Py152/tcl8.0/library
> E:/Programs/Py152/Tools/library
> 
> This probably means that Tcl wasn't installed properly.
> ============================================================
> 
> The directories in sys.path are as follows ...
> E:\PROGRAMS\PY152
> E:\PROGRAMS\PY152\DLLs
> E:\PROGRAMS\PY152\PIL
> E:\PROGRAMS\PY152\lib
> E:\PROGRAMS\PY152\lib\lib-tk
> E:\PROGRAMS\PY152\lib\plat-win
> E:\Programs\Py152\DLLs
> E:\Programs\Py152\Lib
> E:\Programs\Py152\Lib\lib-tk
> E:\Programs\Py152\Lib\plat-win
> 
> Tcl in installed in E:\PROGRAMS\Py152\\Tcl, and init.tcl exists in
> E:\PROGRAMS\Py152\Tcl\lib\tcl8.0 ... but this directory does not
> seem to be searched.
> 
> Do I need to set additional environmental variables or make changes
> to the registry beyond those made by the installer?
> 
> Thanks in advance for any help
> 
> David Nichols
> Source Signal Imaging
> dnichols at sourcesignal.com


From pinard at IRO.UMontreal.CA  Tue Jan  4 21:16:15 2000
From: pinard at IRO.UMontreal.CA (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 04 Jan 2000 21:16:15 -0500
Subject: Progress at internationalising the TP robot
Message-ID: <vr3dsd2rgw.fsf@trex.IRO.UMontreal.CA>

Hello, my friends.  Please accept my best wishes!

I made some progress at internationalising the Python robot, which is
used to process submissions in the Translation Project.  Let me share a
quick status report with you, by repeating part of a message I sent to
the national team leaders, earlier today.  Comments, if any, are welcome.

----->
Oh, by the way.  All translatable strings in the robot tools have been
marked as such, and for the time being, merely translate to themselves.
If you are curious, a very crude POT file (containing many duplicate strings)
is available as:

   http://www.iro.umontreal.ca/contrib/po/tprobot.pot

This file should be remade using a more proper message extractor, before
it becomes available to translators for real.  That will occur only later,
as the Python extractor is not ready.  For the time being, I intend to use:

   http://www.iro.umontreal.ca/contrib/po/gettext/scan/xpot/*

as a base.  I originally made this program to explore how big would be the
problem of marking all strings of Emacs sources.  A bit sadly, I now need
to make it MULE aware, because many Emacs files (in the /lisp/language/ and
/lisp/international/ subdirectories) are saved on disk using the internal
MULE code of Emacs.  I rather recently switched the hashing module in
`xpot', from the `gettext' version to the `recode' version: Jim Meyering
and I made that one, hopefully better, but using a fairly different API.
Remains to make `xpot' more usable, and to add full Python support.

The `tprobot.pot' file has been produced using PO mode for Emacs.  I adapted
it so it works with Python strings in file, in all 12 flavours (either
raw or not, for '...', "...", '''...''', ''"...", ""'...' and """...""").
It properly concatenates sequences of strings (and allows embedded comments).
An unexpected difficulty came from PO mode not processing nested (or
included) TAGS file, so I rewrote that part as well, even if a bit uneasy.
In any case, I think PO mode is now quite functional for marking Python:

   http://www.iro.umontreal.ca/contrib/po/po-mode/po-mode.el

Audacious PO mode users should be kind enough to pretest this version.
-----<

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



From hanche at math.ntnu.no  Fri Jan 28 13:50:17 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 28 Jan 2000 13:50:17 -0500
Subject: Charts
References: <3890C74D.786411BC@acm.org> <87n1pr6m0i.fsf@iname.com> <389118DB.4B7997AE@wjk.mv.com>
Message-ID: <pcopuumyqs6.fsf@math.ntnu.no>

+ "Wm. King" <wjk at wjk.mv.com>:

| Other would be to use Tkinter and create a canvas ( have to do math
| as well) and another would be to use Tcl/TK and a create a
| canvas.. you would have to have plug-in for browser...

Or maybe you could create a postscript file from the canvas (the
Canvas class has a postscript method) and then convert that to a gif,
which is easily done using imagemagick.
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From gmcm at hypernet.com  Tue Jan 18 11:39:31 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 18 Jan 2000 11:39:31 -0500
Subject: DBI spec and parameter passing
In-Reply-To: <14468.37811.321605.587902@beluga.mojam.com>
References: <1263967432-11066499@hypernet.com>
Message-ID: <1263908923-14585671@hypernet.com>

Skip wrote:
> 
>     >> Instead of doing the parameter interpolation yourself, try letting
>     >> your database module do it for you, e.g.:
>     >> 
>     >> c = db.cursor()
>     >> c.execute("""select id from city where"""
>     >> """ city = %s and"""
>     >> """ state = %s and"""
>     >> """ country = %s""", (city, state, country))
> 
>     Gordon> but Skip _meant_:
> 
>     Gordon> c.execute(""" select id from city where city = ? and state = ?
>     Gordon> and country = ?""", (city, state, country))
> 
> I suppose so, in the correct context.  Not being an ODBC person I didn't
> know what was required in the context of the original question, so all I
> could do was cut/paste from my MySQLdb code, which wants the %s thingies...

Sorry. DBI is flexible about this, but the only dynamic 
parameter marker mentioned in the SQL 92 (aka SQL II) spec 
is "?". Or to put it another way, since ODBC is part of SQL II, 
the ODBC way *is* the standard way ;-).

- Gordon



From mal at lemburg.com  Wed Jan 19 12:24:11 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 19 Jan 2000 18:24:11 +0100
Subject: Python and MS SQL (Microsoft SQL server)
References: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in> <3884C60B.BE820782@psychosis.com> <3884ECE8.DFB83340@lemburg.com> <3885DF0C.819CB0FB@psychosis.com>
Message-ID: <3885F33B.93EF007C@lemburg.com>

Dave Cinege wrote:
> 
> "M.-A. Lemburg" wrote:
> >
> > Dave Cinege wrote:
> > >
> > > "George Thom at s" wrote:
> > > >
> > > > Hi Dave,
> > > > The odbc module that comes as part of the win32 extensions bundled with
> > > > Python 1.5.2 for the various flavours (!) of Windows should do just fine.
> > >
> > > So you're saying python has to be runnign on NT for this to work?
> > >
> > > > Additional functionality (with some tweaking ... ymmv ;-) can be obtained
> > > > with marc Lemburg's mxODBC module.
> > >
> > > What about this? we're looking to run on solaris or linux.
> >
> > Then you should probably try out the Sybase interface together
> > with the Sybase dev libs.
> 
> So you're saying the sybase interface works with

... MS SQL Server ?! Yes, it does from what I've heard. They both
use the same protocol (MS probably bought Sybase and renamed
it).
 
> mxODBC will only connect you to
> > local databases unless you let it talk to the DB through
> > an ODBC-ODBC bridge.
> 
> That sounds like a better solution. Can I do that using the unix ODBC thing
> that I've seen listing in the python modules page?

You could try the OpenLink or Intersolv driver sets or
the EasySoft ODBC-ODBC bridge. All are reported to work
together with mxODBC.

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





From fredrik at pythonware.com  Thu Jan 13 07:28:16 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 13 Jan 2000 13:28:16 +0100
Subject: inverse of __repr__ ?
References: <yqtn1qa8aga.fsf@toad.scripps>
Message-ID: <001801bf5dc1$ac43a990$f29b12c2@secret.pythonware.com>

Thien-Thi Nguyen wrote:
> i've looked through _programming python_ for the inverse of __repr__, but
> could not find anything.  perhaps i'm missing some clue?

> or is there no way to extend the reader in a python repl?

if "reader" means the python source parser, as exposed
by eval/exec/compile, it's limited to python syntax.  or in
other words:

    x == eval(repr(x))

only holds for simple types.

dunno what a "repl" is, so I cannot answer the rest
of your question.

</F>




From aahz at netcom.com  Sat Jan  8 10:05:19 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jan 2000 15:05:19 GMT
Subject: is variable a list?
References: <38729120.7616054D@simplypets.com> <86hfgp6o7m.fsf@ruby.somewhere.washington.us>
Message-ID: <857jnf$ne4$1@nntp5.atl.mindspring.net>

In article <86hfgp6o7m.fsf at ruby.somewhere.washington.us>,
Evgeny Roubinchtein  <eroubinc at u.washington.edu> wrote:
>
>Hmm, I guess TIMTOWDY _sometimes_ applies to Python :-).

Well, sure, that's going to be true of pretty much any powerful
programming language, but TMTOWTDI (note spelling) is not a design goal
for Python and it's usually the case that there are only one or two
obvious ways of accomplishing a task.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From jbauer at rubic.com  Fri Jan  7 14:48:41 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 07 Jan 2000 13:48:41 -0600
Subject: UDPserver bug / documentation error?
References: <8557h8$1536@newton.cc.rl.ac.uk>
Message-ID: <38764319.2291C75B@rubic.com>

> I was attempting to kludge together a quick UDP server 
> program in Python (1.5.2).

Richard,

Here's a simple UDP server that I use for debugging
CGI scripts:

http://starship.python.net/crew/jbauer/creosote/

Regards,

Jeff Bauer
Rubicon Research



From fdrake at acm.org  Mon Jan 10 10:05:15 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 10 Jan 2000 10:05:15 -0500 (EST)
Subject: Error using string.strip
In-Reply-To: <38775BED.7D4595B3@worldnet.att.net>
References: <38775227.2CC0D685@worldnet.att.net>
	<38775BED.7D4595B3@worldnet.att.net>
Message-ID: <14457.62763.288030.432621@weyr.cnri.reston.va.us>

Eric Hausig writes:
 > I'm going through the online tutorial at www.python.org.  Unfortunately,
 > they didn't make mention of import prior to the String section of the
 > tutorial.  Hehe....this is a really neat language so far.

  Please be assured that the next release of the documentation adds
the required import; you're not the first to be tripped up by this
omission!


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From pinard at iro.umontreal.ca  Wed Jan 19 16:03:36 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 19 Jan 2000 16:03:36 -0500
Subject: Python BUG?
In-Reply-To: "Tim Peters"'s message of "Thu, 13 Jan 2000 02:43:52 -0500"
References: <000401bf5d99$f00804a0$d62d153f@tim>
Message-ID: <oq1z7dydsn.fsf@titan.progiciels-bpi.ca>

"Tim Peters" <tim_one at email.msn.com> writes:

> Conversion to decimal is a quadratic-time (in the number of bits) process.

This is why `bc' and some other tools work in base-(10**N), for some suitable
N, when it is known that decimal printing will often be needed.  Such a
choice does not imply much slowdown in long integer computations.  Even if
unlikely to occur, Python was offering some way for the user to specify
the preferred base for printing, and choosing super-digits accordingly.
This might avoid quadratic behaviour in practice, but speaking practice,
I wonder if people handling huge numbers really need to print them :-).

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





From tim_one at email.msn.com  Sun Jan 23 22:55:28 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 23 Jan 2000 22:55:28 -0500
Subject: Python compiler?
In-Reply-To: <20000123092040.A2010@stopcontact.palga.uucp>
Message-ID: <000401bf661e$da1c4100$902d153f@tim>

[J Dubal]
> Does [Python] have a compiler for Linux and/or windows that can
> produce native executables?

[Gerrit Holl]
> Yes, it has one. It's included in the Python distribution,
> Tools/freeze.  It's called freeze, and with it, you won't
> need any Python interpreter.

Just to prevent any possible misunderstanding, freeze works by sticking a
copy of the interpreter into the executable.  This is probably not what J
Dubal had in mind, but that's life.  A Python VM is needed to execute Python
bytecodes just as a Java VM is needed to execute Java bytecodes, and that's
what "the interpreter" is.

the-natives-are-restless-tonight-ly y'rs  - tim





From akhar at videotron.ca  Fri Jan  7 01:36:58 2000
From: akhar at videotron.ca (Akhar)
Date: Thu, 6 Jan 2000 22:36:58 -0800
Subject: How do I use the pop module??
Message-ID: <lcdd4.567$KI6.19133@weber.videotron.net>

I am trying to use python for my web site but i need to give e-mail access
but it seems the the password doesn't work!! why is that?? I use OpenBSD 2.5
and plan to use 2.6 can you give me a hint ?? or code snippets.
thank you in advance
Akhar




From gerrit.holl at pobox.com  Thu Jan 13 09:36:12 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 13 Jan 2000 15:36:12 +0100
Subject: a small problem
In-Reply-To: <20000113130514.A21652@xs4all.nl>; from thomas@xs4all.net on Thu, Jan 13, 2000 at 01:05:14PM +0100
References: <20000112015923.A8450@xs4all.nl> <000701bf5cb4$c72320a0$da2d153f@tim> <20000112145107.B8450@xs4all.nl> <20000112165932.A2758@stopcontact.palga.uucp> <20000113130514.A21652@xs4all.nl>
Message-ID: <20000113153612.A1808@stopcontact.palga.uucp>

Thomas Wouters wrote on 947765114:
> On Wed, Jan 12, 2000 at 04:59:32PM +0100, Gerrit Holl wrote:
> 
> > People should hang around in comp.lang.c or even comp.lang.asm.i386 if
> > they care about speed.
> 
> No, they should be there if they _only_ care about speed. 

OK.

> Another amusing thing was that I wrote:
> 
> for x, in mylist:
> 
> to iter over the testlist (a list of single-item tuples.) I didn't
> consciously think about it, but when i reread it, i decided it couldn't be
> right. It would just be too obvious. There had to be a reason for it to be a
> special case of some kind, like perl has with the magic ++ operator on
> strings. Contrary to the string-++ operator in perl, though, it worked just
> as expected ;)

> <snip>
> > How would I do that without lambda's, if this code is in a method? Should
> > I nest two functions?
> 
> For a straight translation without lambda's you can _always_ use functions.
> nested, if you wish, but not necessarily. lambda's work _exactly_ like
> nested functions, only you cant use statements in them, only expressions.
> 
> But doing it without lambda is not the point -- if lambda wasn't there,
> there would possibly be a more convenient way to do such things, or even
> more speedy (like a module of functions, like operator, for use with map.
> basic things like adding the first element of each argument. Or something.)

Well, if Guido doesn't like them, and they occur almost nowhere in the
library, he might have another thing to do it.

> <snip>
> > lambda's are good, because otherwise, it wouldn't be possible to write
> > obfuscated code in Python... Can someone please rewrite the one-liner
> 
> You're missing the point. Guido does not believe obfuscated code is good.
> Why would Python need obfuscated code ? It goes through great pains(*), in
> fact, to make obfuscated code very hard.

I was making a joke ;-)

> (*)Admittedly mostly other peoples' pains, the ones who dislike forcefed
> indentation, for instance :)

Revenge... all those Perl scripts....... REVENGE!!
The Amsterdam Perl Mongers even have '@ $ %' in there logo, as if they
are proud of it... I don't know any Perl, but I can't understand how to
write a function that takes any kind of argument if you need to type
such a prefix...

> > > Hi! I'm a .signature virus! copy me into your .signature file to help me spread!
> > Oh, help! My computer is infected! What should I do now?!
> 
> The usual cure is to install a less gullible mind, or indeed, any at all.
> But that is what the signature tries to do, so perhaps you can consider
> yourself cured ;)

:-)

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my mail!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From fredrik at pythonware.com  Mon Jan  3 22:38:00 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: 3 Jan 2000 21:38:00 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jan  3)
Message-ID: <525447BA99B1B464.8832B2C25AFE47BB.AB44C329D970F070@lp.airnews.net>

So at last, the last year of this millenium arrives, and not only
Python, but also most other things in the computing universe turn
out to be fully Y2K compliant.

    Will Ware starts the new year by posting code to down-
    load and process historical stock data:
        http://www.deja.com/=dnc/getdoc.xp?AN=567038652

    Are you planning to go to the conference, but haven't
    gotten around to register yet?  It's time to do some-
    thing about that:
        http://www.deja.com/=dnc/getdoc.xp?AN=567818406
        http://www.python.org/workshops/2000-01/

    Is Python a VHLL?  What exactly *is* a VHLL?  Pythonistas
    discuss this interesting topic in this thread:
        http://www.deja.com/=dnc/viewthread.xp?AN=566833227

    More books: Tim Altom's "Programming With Python" has
    arrived, and technical editor Mitch Chapman explains what
    it is all about:
        http://www.python.org/psa/bookstore/
        http://www.deja.com/=dnc/getdoc.xp?AN=567442652

    O'Reilly's "Learning Python" is now available as eMatter:
        http://www1.fatbrain.com/asp/bookinfo/bookinfo.asp?theisbn=EB00004715

    And another eMatter Python book climbs the bestseller
    lists...
        http://www1.fatbrain.com/ematter/bestsellers.asp?vm=e


========================================================================

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

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      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 <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From jbauer at rubic.com  Fri Jan 21 08:35:30 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 21 Jan 2000 07:35:30 -0600
Subject: WAP with Python
References: <20000121120558.50068.qmail@hotmail.com>
Message-ID: <388860A2.4A5EADFF@rubic.com>

Ulf Engstr?m wrote:
> Has anyone tried to do any WAP-things with Python?

The following site demonstrates a WAP application. It
does an online search for new and used cars in the UK.
If you don't have a WAP browser, you can view a static
demo.

  http://www.rcp.co.uk/distributed/wapdemo

RCP Distributed Systems web site is powered by Digital 
Creation's Zope product.  You might search the Zope 
archives for more WAP info.  http://www.zope.org

Jeff Bauer
Rubicon Research



From kc5tja at armored.net  Fri Jan 21 18:13:47 2000
From: kc5tja at armored.net (Samuel A. Falvo II)
Date: Fri, 21 Jan 2000 15:13:47 -0800
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net> <slrn87cvv9.gcu.kc5tja@garnet.armored.net> <3876893D.876A8949@rubic.com> <wkn1q0erh0.fsf@sccsi.com>
Message-ID: <3888e2c4@fermi.armored.net>

> I'm currently looking at doing this for one of my consulting clients.
> I'm looking at using Nouveau from RougeWave to help with the CORBA and
> COM bridging.

People interested in helping complete the GCOM project (a generic version of
COM for Linux) is free to contact me at any time.





From jblazi at netsurf.de  Sat Jan 29 09:07:15 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Sat, 29 Jan 2000 15:07:15 +0100
Subject: Cutting off trailing zeroes?
References: <3892ad81$1_2@goliath.newsfeeds.com> <20000129142714.A2146@stopcontact.palga.uucp>
Message-ID: <3892f230_6@goliath.newsfeeds.com>

Thx. I am such a stupid person! I *saw* that python put it right when I
wrote "print x" and I *knew* i was able to do the same but I could not
figure out how. Oh God!.

J.B.



Gerrit Holl <gerrit.holl at pobox.com> schrieb in im Newsbeitrag:
20000129142714.A2146 at stopcontact.palga.uucp...
> Janos Blazi wrote on 949137250:
> > Id like to format a string like this:
> >   s = "(%f)" % x
> >
> > Now if x=1.5 then s="1.50000" or something like this. Is there a
tirivial
> > way of cutting of the zeroes, i.e. getting s="1.5" instead of
s="1.50000"?
> >
> > Of course it can be programmed very simply. My question is: can I do it
with
> > specifiers somehow?
>
> s = "(%s)" % str(x)
>
> regards,
> Gerrit.
>
> --
> Please correct any bad English you encounter in my email message!
> -----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
> Version: 3.12
> GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V
PS+ PE?
> Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
> -----END GEEK CODE BLOCK-----
>




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From tseaver at aftershock.neosoft.com  Tue Jan  4 13:09:28 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 4 Jan 2000 12:09:28 -0600
Subject: CPAN for Python?
References: <38720F70.A5DC6F33@ttm.com> <84t52c$1gg$1@nntp9.atl.mindspring.net> <C28D1A44F2165E48.B5C4E34C4CD957A5.44F995461BA2F6ED@lp.airnews.net> <slrn874buo.2do.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>
Message-ID: <A99BAC4F8CFFB5C8.FB3BECE999060AC1.EAF3C286CA4FB724@lp.airnews.net>

In article <slrn874buo.2do.coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com>,
Stephen Coursen <coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com> wrote:
>On 4 Jan 2000 11:30:50 -0600, Tres Seaver <tseaver at aftershock.neosoft.com>
>wrote:
>>In article <84t52c$1gg$1 at nntp9.atl.mindspring.net>,
>>Aahz Maruch <aahz at netcom.com> wrote:
>>>In article <38720F70.A5DC6F33 at ttm.com>,
>>>Scott Anderson  <sanderson at ttm.com> wrote:
>>>>
>>>>Is there an equivalent of Perl's CPAN for Python?
>>>
>>>Well, there's Parnassus (http://www.vex.net/parnassus/), but there's not
>>>yet an overarching authority.
>>
>>www.python.org blesses it, which is probably about as close as we'll see.
>>I may misrecall (blessedly little Perl experience) but doesn't CPAN allow
>>for browsing from the interpreter prompt, and downloading?  That would be
>>a cool extension of a wonderful resource.
>>
>>>OTOH, Python's "built-in" libraries are
>>>more extensive than Perl's; see the Library Reference for details.
>>
>>parnassus.py,-anyone?'ly
>>
>That would be a useful module to have, but for it to work, there would have
>to be a standard installation mechanism.  Perl uses:
>     perl Makefile.pl
>     make
>     make install
>     
>If python had a standard way of installing modules, writing a parnassus.py
>script would be easy.  Unfortunately, not every module installs the same way
>(although there seems to be only a few different methodologies).
>
>Steve

The distutils SIG will be reporting at IPC8/DevDay: here's hoping they are
thinking along theses lines.

Clapping-coz-I-believe'ly

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From herzog at online.de  Thu Jan 20 12:03:04 2000
From: herzog at online.de (Bernhard Herzog)
Date: 20 Jan 2000 18:03:04 +0100
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p>
Message-ID: <m3n1q0fzg7.fsf@greebo.nodomain.de>

Edward Welbourne <eddyw at lsl.co.uk> writes:

> Markus Stenberg and Bernhard Herzog consider the two-namespace rule to
> be a deficiency in python, and think it would be `simpler' to use
> `lexical scoping'.

I think I should clarify my position. I don't think that the
two-namespace rule is a deficiency. I have never really needed more than
that in all the Python programs I wrote. That may be simply be because I
have more of an OOP background (mostly C++ before I discovered Python)
than functional, so I tend to use class instances or bound methods where
some other people want to use curried functions, etc.

However, several people have proposed lexical scoping extensions and
some of these proposals have even been implemented. Lexical scoping is
somewaht controversial and I don't know what Guido's opinion is,
although if the technical reasons (cyclic references) Tim mentioned in
his post were the only reason Guido dropped it, it seems likely that
lexical scoping might be implemented if the technical problems can be
solved.

Lexical scoping would make your tunneling feature unnecessary.
Therefore, IMO, if there's a reasonable chance that Python 2 has lexical
scoping there's no real need to have tunneling at all.

>  The Tim has given an excellent and erudite summary
> of what that could be asked to mean, since Evan asked and its advocates
> didn't answer.  Personally, I consider the two-namespace rule to be a
> major *strength* of python: it is a major ingredient in pythonicness.
> 
> Define a function, foo, which returns a function it has defined, as bar,
> during its suite; have bar refer to some of the local variables of foo.
> With lexical scoping, the interpreter is obliged to keep the namespace
> foo used (on that invocation) for as long as anyone retains a reference
> to what it returned.
> 
> I contend that the namespace in which a function executes its suite
> *should* be discardable at the time the function returns.

That is indeed a useful feature because it makes it easy to predict when
objects are collected in most cases (assuming refcounting).

>  Lexical
> scoping would mean that we can't discard foo's namespace when foo
> finishes execution.  The implications of this for garbage collection are
> singularly unpleasant - give them some thought.

It seems to me that in most cases there are only a few names of f's
namespace that have to be presevered and the compiler can easily tell
which. Of course they can still introduce cyclic references especially
if you want bar to be recursive.


> The only way out is for the values bar accesses from foo to be
> transcribed, before foo finishes executing, into some namespace whose
> lifespan is at least as long as that of bar (and, ideally, coeval with
> it).

If lexical scoping is not adopted this might be useful.

>  Since bar carries around a namespace, for its defaults, with
> exactly the right life-span, that's the right place to which to
> transcribe such values (promoting them to globals being, I trust, to
> no-one's liking).  Now, some folk might prefer to have that done
> implicitly: personally, I'm all in favour of each suite of code needing
> to be explicit about its dependencies on others.

Yes, if such a mechanism is added it should be explicit and not implicit.

>  Like Evan,
> 
> > I rather like the idea of having to explicitly bring names into an
> > inner scope, and with semantics identical to parameter defaults.

I think what this is about is not bringing names into the inner scope
but values.

And to Evan's next two questions:

> Why is this so objectionable? Is it just the syntax?

IMO, it's the syntax, mostly. The argument list is the wrong place.
Arguments are used when the funtion is called, not when it's defined.

To reuse one of Edward's examples:

  def filename(*fragments, **, root=os.getcwd()):
      """Resolves relative paths."""
      # Assume os.path.join joins arbitrarily many fragments.
      return apply(os.path.join, (root,) + fragments)

IMO, the root=os.getcwd() should be moved outside the argument list.
It's assigned at definition time and can't be changed in any way when
the function is called, so it's not an argument. Something like this
would be better:

  def filename(*fragments), root=os.getcwd():
      """Resolves relative paths."""
      # Assume os.path.join joins arbitrarily many fragments.
      return apply(os.path.join, (root,) + fragments)


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


From pierre at saiph.com  Wed Jan  5 18:49:08 2000
From: pierre at saiph.com (Pierre Imbaud)
Date: Thu, 06 Jan 2000 00:49:08 +0100
Subject: module path:funny bug, funny solution
Message-ID: <3873D874.62A35E7B@saiph.com>

I run into a funny bug, working on a python script, importing modules
laying in the same directory. The script performed some chdir. I debug
under xemacs, executing the script with py-execute-buffer. At some
point, I had to modify an imported module: I put a reload(module) in
the code: surprise! raises a "ImportError: No module named xx"!
what happens is quite simple: sys.path[0] is '', that means: first
search for modules in the current directory. First time, modules are
found ok. when trying to reload, the current directory has changed!
Now, I think its funny. I didnt when it happened.

The solution I found? between import os and import custom modules:

try:
    os.chdir(startDirectory)            # succeeds at subsequent runs
except:
    startDirectory = os.getcwd()        # run at first run.

well, it works, but may I let such a puzzle in my code? The
unadvertized reader will think Im plain mad...

-- 
Pierre Imbaud <pierre at saiph.com>
12 Rue des Bosquets 91480 Quincy Sous S?nart France
Tel:  01 69 00 94 57 Fax 79 65


From wtanksle at hawking.armored.net  Fri Jan 21 22:19:18 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 22 Jan 2000 03:19:18 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net> <8631bt$1nu@news.or.intel.com>  <slrn88d63f.13f.wtanksle@hawking.armored.net> <slrn88fs4o.2d0.warlock@raven.myth>
Message-ID: <slrn88i92u.bbu.wtanksle@hawking.armored.net>

On Thu, 20 Jan 2000 21:37:28 -0800, Jim Richardson wrote:
>On 20 Jan 2000 04:58:00 GMT, 
> William Tanksley, in the persona of <wtanksle at hawking.armored.net>,
> brought forth the following words...:

>>indentation.  So what?  Python uses assignment.  ML doesn't.  Does that
>>mean ML is bad and Python is good, or is it the other way around?

>what does ML do then? this has me puzzled...

In ML (and in many other programming languages) all object are immutable
-- once you create them you can't change them.  This includes variables
and function parameters.

Instead of changing variables, though, you can easily create new ones with
the same name as the old one.

>Jim Richardson

-- 
-William "Billy" Tanksley, in hoc signo hack


From bww00 at amdahl.com  Tue Jan  4 14:22:53 2000
From: bww00 at amdahl.com (Bryan Webb)
Date: Tue, 4 Jan 2000 13:22:53 -0600
Subject: profiling help
Message-ID: <84thor$211@masters0.InterNex.Net>

Hi,
i WOULD LIKE TO PROFILE A script and capture the output to a file.
I would like to use the profile.py script from the cmd line like
profile.py bww.py

How do I capture the profile output

Thanks
Bryan




From fdrake at acm.org  Tue Jan  4 13:57:01 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Tue, 4 Jan 2000 13:57:01 -0500 (EST)
Subject: Grail
In-Reply-To: <387237F3.C35B7FA1@callware.com>
References: <387237F3.C35B7FA1@callware.com>
Message-ID: <14450.17021.111722.710187@weyr.cnri.reston.va.us>

Ivan Van Laningham writes:
 > Any clues?  I'm running on WinNT 4.0, SP5.  160MB ram, PII 400Mhz.

  That's a bug.  If you want to fund Grail development, I'll see if I
can come up with more clues.  ;)

 > If this can't be made to run, are there any HTML widgets out there?

  Doesn't wxWindows include one?  Or are you specifically interested
in Tk?


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From cpatti at black-racer.atg.com  Fri Jan 28 12:21:05 2000
From: cpatti at black-racer.atg.com (chris patti)
Date: 28 Jan 2000 12:21:05 -0500
Subject: lib reference in Unix man format
References: <200001271548.KAA04215@trm.bmo.com> <slrn891ljb.5tk.cjc26@bridget.mindriot.net>
Message-ID: <yu1emb2qfi6.fsf@black-racer.atg.com>

cjc26 at nospam.cornell.edu (Cliff Crawford) writes:
[snip question about doc in man page format]
> There isn't any documentation in man format.  I know this because Tom
> Christiansen came over here once and VERY LOUDLY complained about the
> lack of this. :)  Actually I find this annoying too, so maybe I'll
> work on converting the on-line docs to man pages (when I have some
> free time...:).  I'd prefer to have each built-in function have its
> own page though, rather than glomming them all together into one
> large "pythonfunc" page.
> 

Wait a second, Tom Christensen says things in a way OTHER than VERY
LOUDLY? :)

The man has some serious UNIX/Perl clue though, he's just blustery as
all get out in person or live online.

-Chris


From calishar at *NOSPAM*home.com  Sun Jan 16 17:43:12 2000
From: calishar at *NOSPAM*home.com (Calishar)
Date: Sun, 16 Jan 2000 22:43:12 GMT
Subject: problem with makepy output
Message-ID: <4vg48skfrigntmuslpfm40ngea5pdl4r2m@4ax.com>

Hello again folks,

  Just tried using makepy (from win32all-127) to create a python
module wrapping MS-Excel 2000, it worked fine, I moved the module to a
directory in the python path, and imported it. This was the result.

------------
PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> import excel5
>>> app=excel5.Application()
Traceback (innermost last):
  File "<interactive input>", line 1, in ?
  File "c:\Program Files\Python\kadabra\excel5.py", line 804, in
__init__
    oobj = pythoncom.new(self.CLSID)
com_error: (-2147221164, 'Class not registered', None, None)
------------

However, the same process worked on MS-Word 2000. Just wondering how I
would start debugging this.

Calishar


From otis at my-deja.com  Thu Jan 13 17:32:33 2000
From: otis at my-deja.com (Otis Gospodnetic)
Date: Thu, 13 Jan 2000 22:32:33 GMT
Subject: best Template module
Message-ID: <85ljq1$mms$1@nnrp1.deja.com>

Hello,

Could anyone recommend a good 'Template module'?

Template module - a module that will let me create a plan text/HTML
file with some special markup/tags that I can then replace by some text
values from within my Python code

good - fast execution and not too complex :)

Sorry for this newbie question...

Thanks,

Otis


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


From mmillikan at vfa.com  Mon Jan 31 10:09:03 2000
From: mmillikan at vfa.com (mmillikan at vfa.com)
Date: Mon, 31 Jan 2000 10:09:03 -0500 (EST)
Subject: Stackless Python 1.0 + Continuations 0.6
Message-ID: <200001311509.KAA18274@python.org>

Congratulations on 'stackless'!

There appears to be a problem in handling control flow in the
following class of constructs:

def stacklesstest():
    try:
        return 45
	print 'still here' #not printed (correct)
    finally:
        print 'in finally' #printed (correct)
    return 'x = 99'

standard 'CPython' execution:

C:\WebSite102\bin>python -i I:\_members\mark\p5\experimental\stacktest.py
>>> x = stacklesstest()
in finally
>>> x
45
>>>

JPython execution:

>>> ## working on region in file i:\_members\mark\p5\Experimental\python-BDAqVJ...
>>> x = stacklesstest()
in finally
>>> x
45
>>> 

'stackless' execution:

>>> ## working on region in file i:\_members\mark\p5\Experimental\python-BDAEqV...
>>> x = stacklesstest()
in finally
>>> x
'x = 99'
>>> 

The 'return' inside the 'try:' clause exits the 'try:' clause and
executes the 'finally:' clause as expected. However, the 'return' and
associated value is forgotten after the 'finally:' clause has
executed: the unnested 'return' is then executed before the function
returns.

The behaviour is the same in the following function with the
expicatory 'noise' removed.

def stacklesstest2():
    try:
        return 45
    finally:
        pass

This impacts some code in wxPython, possibly it also may be the cause
of the difficulties in Zope(?)





From annis at biostat.wisc.edu  Thu Jan 13 11:52:33 2000
From: annis at biostat.wisc.edu (William Annis)
Date: 13 Jan 2000 10:52:33 -0600
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com> <002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com>
Message-ID: <jtu2kh52y6.fsf@wazor.biostat.wisc.edu>

"Fredrik Lundh" <fredrik at pythonware.com> writes:

> # -*-python-*-
> 
> or
> 
> #  -*- Mode: Python; tab-width: 4 -*-

        Those all work, but as I recall those work only on the first
line of the file.  A quick test seems to confirm this.

        I use this in my .emacs, which looks for the word "python" in
the shebang line:

------------------------------
(setq interpreter-mode-alist
      (cons '("python" . python-mode) interpreter-mode-alist))
------------------------------

-- 
William Annis - System Administrator - Biomedical Computing Group
annis at biostat.wisc.edu                       PGP ID:1024/FBF64031
Mi parolas Esperanton - La Internacia Lingvo    www.esperanto.org


From JoeSmith at bogusaddress.com  Thu Jan 20 20:58:44 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Fri, 21 Jan 2000 01:58:44 GMT
Subject: Python Rocks! - ASP problem?
References: <860ftk$bas@news.or.intel.com> <38858F25.84946F4B@bogusaddress.com> <010101bf6320$7cdb3fb0$f29b12c2@secret.pythonware.com>
Message-ID: <3887BE1F.EEDD9954@bogusaddress.com>

No.  I doubt Microsoft even had a version of IIS (internet info server) on the
design table.  This is the only known problem (other than people not liking it) that
I know of with not having an end delimiter that I know of.

Fredrik Lundh wrote:

> Joe Smith <JoeSmith at bogusaddress.com> wrote:
> > This conditionally put the html code into the page.  If I remember right,
> > python can't do this in ASP because an end marker is needed.  I would say that
> > python probably was not designed with ASP in mind.
>
> did ASP really exist back in the late eighties?
>
> </F>



From Vadim.zeitlin at dptmaths.ens-cachan.fr  Wed Jan  5 15:02:17 2000
From: Vadim.zeitlin at dptmaths.ens-cachan.fr (Vadim Zeitlin)
Date: Wed, 05 Jan 2000 21:02:17 +0100 (CET)
Subject: Python and GUI options
In-Reply-To: <20000105141519.A13656@vislab.epa.gov>
Message-ID: <XFMail.000105210217.Vadim.zeitlin@dptmaths.ens-cachan.fr>

On 05-Jan-00 Randall Hopper wrote:
> I wasn't even thinking of the make/gmake part.  Using gmake was the first
> thing I did 

 Good :-)

> From a user standpoint, it would be useful if the build system were changed
> to a GNU configure sense-and-testwhat-you-have scheme rather than a
> configure-this-file, put-overrides-in-this-file, and see-what-breaks
> scheme.

 Sorry, I don't understand what are you referring to. wxWindows uses
configure exclusively (i.e. we don't support manually editing the
makefiles any more) since more than a year and a half and it had
it along with the makefiles long time before it... Under Unix
(and even Windows if you're lucky enough to have bash) you don't
have to edit anything - just run configure with necessary options.
 
> Re the other problems.  If build system changes don't do away with it (I
> hope they do), more docs on creating the build.local file are a must.

 Sorry again, but I don't even know what is it?

> Also, I had to copy this around when I was building wxPython and then its
> assorted modules because it didn't look for it except in $PWD.  I had to
> make it compile in an RPATH so wxPython could find wxWindows.  This is
> probably there for Solaris (-R) but it's not there for SGI (-rpath) since
> you're not testing on that OS.  

 It would be nice if someone knowing how shared libraries work on SGI
(I don't) could patch configure.in to take this into account. I believe
that people did succeed in building libwx_gtk.so under Irix though...

 Anyhow, patches are always gladly accepted :-)

> Also, wxGTK and wxPython sense different
> compilers (SGI vs. GCC).  I had to coerce them to use the same ones.  

 Hmm... How did this happen? Did the value of your CXX env var changed
in between somehow?

> configured with --with-opengl and it didn't pull in the GL canvas.

 I believe that this problem will disappear in future wxPython releases
because it won't need it any more at all.

 But please do tell us if there is something wrong with configure under
Irix (or elsewhere) - obviously, we can't fix it if we don't know about
it.

 Regards,
VZ



From neilh at hare.net.au  Fri Jan 14 08:16:52 2000
From: neilh at hare.net.au (Neil Hodgson)
Date: Sat, 15 Jan 2000 00:16:52 +1100
Subject: Perl speed vs. Python speed
References: <85hjfa$mlp$1@nnrp1.deja.com> <pd0f4.26480$W2.358425@iad-read.news.verio.net> <00f801bf5d0e$ae601140$f29b12c2@secret.pythonware.com>
Message-ID: <008901bf5e91$a040e6a0$01646464@computer>

> (searching for the given patterns in strings padded
> with blanks on both sides.  sre8 is the new python
> 1.6 engine on 8-bit strings, sre16 is the same engine
> using unicode)

   Any hope of a UCS-4 version?

   Combining Python's preeminence for Mayan calendrical work with support
for hieroglyphics will allow us to conquer the whole of the ancient world.

   Neil





From mgushee at havenrock.com  Mon Jan 17 15:10:53 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 17 Jan 2000 15:10:53 -0500
Subject: Tkinter: wm_protocol() error on RedHat Linux 6.0
Message-ID: <m3snzwtq5u.fsf@havenrock.com>

Hello, all--

The other day I released a small Tkinter application which I developed 
on a Red Hat Linux 5.2 system with Python 1.5.2 (my own build). A Red
Hat 6.0 user has reported the following error:

 >  ./surfnotes.py
 > Traceback (innermost last):
 >   File "./surfnotes.py", line 56, in ?
 >     snapp = SurfNotes()
 >   File "./sngui.py", line 164, in __init__
 >     self.wm_protocol('WM_DELETE_WINDOW', self.quit)
 > AttributeError: wm_protocol

He has tkinter-1.5.1-10.i386.rpm.

Anyone have an idea what might cause this?

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From moshez at math.huji.ac.il  Mon Jan 31 15:35:36 2000
From: moshez at math.huji.ac.il (Moshe Zadka)
Date: Mon, 31 Jan 2000 22:35:36 +0200 (IST)
Subject: nesting class definitions?
In-Reply-To: <20000127155551.A2069@stopcontact.palga.uucp>
Message-ID: <Pine.SOL.3.96.1000131223443.20329C-100000@sundial>

On Thu, 27 Jan 2000, Gerrit Holl wrote:

> There aren't functions in the default library that take a class as
> argument, are there?

BaseHTTP.test takes two of those.
(Very convinient design pattern, which should be used more in Python
IMHO)
--
Moshe Zadka <mzadka at geocities.com>. 
INTERNET: Learn what you know.
Share what you don't.




From soundhack at hotmail.com  Thu Jan 20 23:28:05 2000
From: soundhack at hotmail.com (Robert)
Date: Thu, 20 Jan 2000 23:28:05 -0500
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu>
Message-ID: <3887dff7$0$22466@senator-bedfellow.mit.edu>

Roy Smith wrote in message ...
>When I was first learning programming (a quarter of a century ago!),
>there was the idea of the "telephone test", i.e. your code should be
>simple enough that if you read it to somebody on the phone they would
>understand it.  Python not only fails the phone test, it fails the


If I understand you correctly, you mean python will fail it  because (for
example)
there are no end statements to if or for statements, and a line that should
be outside
the block might be interpreted by the listener to be inside. Putting aside
questions of
the usefulness/relevance of this test (I can not imagine why anyone would
want to read
source code over the phone in real life) I think python surpasses other
common languages
in this test by the fact that a  python program will be shorter and easier
to understand
than an equivalent C/++, Perl or Java program. I would rather read a few
lines of code in python,
 specifically denoting  to the listener where the blocks should be, than
read C code of an
equivalent program that is many times longer, or a Perl program with lots of
!@#$ punctuation
marks. (Case in point, when i was working with sockets what took me a
handful of python lines took
 about two pages of C code)

>printed page test; I can print two programs on a piece of paper, which
>differ only in tabs and spaces, and they will appear identical yet which
>do different things.  Barf.


I am not sure what you mean here. Yes, if you mix and match tabs and spaces
you will
experience this (and there are many things you can do to avoid this),
but if you have one program that has only tabs and one program that
has only spaces, they should not only look the same but run the same. I'm
not really a python or programming expert, tho, so you might be referring to
something else.

>I'm in love with python, but this idea of white space as significant
>syntax is the biggest wart on the language by far.

to me the biggest wart is speed, since I am usually too lazy to port
stuff from python to C to make things faster :)

Robert




From scherbi at bam.com  Mon Jan 31 14:12:46 2000
From: scherbi at bam.com (Bill Scherer)
Date: Mon, 31 Jan 2000 14:12:46 -0500
Subject: list -> dict help
References: <3895BE15.A7CF569B@bam.com> <slrn89bidu.p04.scarblac-spamtrap@flits104-37.flits.rug.nl> <slrn89bipe.p04.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <3895DEAE.574232D2@bam.com>

Thanks! This works almost perfect.  My only problem now is that of course I'm
dealing with many lines of input, and I want to put them all in the same
dictionary. (sorry for not mentioning this originally).  eg:

I'll have many lists that are functionally equivalent to the following:

['a', 'b', 'c', 1]
['a', 'b', 'r', 21]
['a', 'g', 'e', 'z', 13]
etc...

my earlier statement, that l[:-2] are netsed keys and l[-2:] is the terminal
key/value pair, still holds.

My first inclination was to use the dictionary update method, but that
overwrites existing keys.  I need something to add dictionaries together.  I'll
hack this together on my own unless someone has somthing ready to go...

Thanks again.

Bill

Remco Gerlich wrote:

> Remco Gerlich wrote in comp.lang.python:
> > def convert(l):
> >     if len(l) == 1:
> >          return l[0]
> >       else:
> >          return { l[0] : convert(l[1:]) }
> >
> > But this copies the list every call, and recursion can use a lot of stack.
> > Since it's tail recursion, there must be an iterative version.
>
> Actually, this being Python, it's all done by reference and the lists
> aren't copied all of the time. My brain is still thinking in terms of
> other languages now and then...
>
> That means the recursive solution is quite good already, I think.
> Can't benchmark them at the moment though.
>
> Also, my editor treats .py files and Usenet posts differently space/tab-wise,
> and copying the functions from my testfile mangled the indentation :(.
>
> --
> Remco Gerlich,  scarblac at pino.selwerd.nl
> Hi! I'm a .sig virus! Join the fun and copy me into yours!
> --
> http://www.python.org/mailman/listinfo/python-list




From bitbucket at isomedia.com  Thu Jan  6 20:32:56 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Fri, 07 Jan 2000 01:32:56 GMT
Subject: Newbie Tkinter / Tk Question
Message-ID: <3875408e.442362303@news.isomedia.com>

I'm finding that on Windows NT with Python 1.5.2 and Tcl 8.0.5,
calling Toplevel.lift () with no arguments doesn't do what it's
supposed to.  According to some doc I have:

 "Lift
lift(), tkraise(), lift(above), tkraise(above). Move self to the top
of the window stack. If self is a child window, it is moved to the top
of it's[sic] toplevel window. If self is a toplevel window (the root
or a Toplevel window), it is moved in front of all other windows on
the display."

However, I'm finding that the window doesn't really go anywhere.  By
my understanding, the following code should have the second window on
top, but it doesn't.  I have found that if I call lift () and give it
the first window as the arg, it works.  Is this a bug, or am I
misunderstanding something about handling these windows?  

And, finally, should I be searching comp.lang.tcl or somewhere else
for these sorts of answers, or is comp.lang.python OK?

->>>
import Tkinter

def cleanup ():
	owSecond.destroy ()
	owFirst.destroy ()
	
owFirst = Tkinter.Tk ()
Tkinter.Label (owFirst, text='This is the first window').pack ()
owSecond = Tkinter.Toplevel (owFirst)
Tkinter.Label (owSecond, text='This is the second window').pack ()
owFirst.protocol ('WM_DELETE_WINDOW', cleanup)
owSecond.protocol ('WM_DELETE_WINDOW', cleanup)

owSecond.lift ()
owFirst.wait_window (owSecond)
-<<<
   -Eugene
import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From erno at iki.fi  Wed Jan 26 12:31:15 2000
From: erno at iki.fi (Erno Kuusela)
Date: 26 Jan 2000 19:31:15 +0200
Subject: re.LOCALE | re.IGNORECASE doesn't work as you'd expect
Message-ID: <87wvow4u58.fsf@fun77.kivikko.hoas.net>

hello,

i was surprised to notice that re.LOCALE doesn't work with re.IGNORE
like you'd think, ie re.search('K??', 'k??rme', re.I | re.L) returns None.

is this a bug or a feature, and is pcre to blame, or python?

  -- erno


From dheise at debitel.net  Thu Jan 27 22:31:57 2000
From: dheise at debitel.net (Dirk Heise)
Date: 27 Jan 2000 22:31:57 MEST
Subject: memory
References: <86q8ee$2g7k$1@noao.edu>
Message-ID: <01bf690d$62199bc0$LocalHost@dreadzone>


Matt Cheselka <cheselka at tucana.noao.edu> schrieb im Beitrag <86q8ee$2g7k$1 at noao.edu>...
> Hi,
> 
> Are there any issues with freeing memory in python?  Say for example I do a
> 
> 	d = array.array('i')
> 	for i in range (2000):
> 	    d.append(0)
> 
> Do I need to free 'd' before I quit the program?
> 
> Cheers,
> 
> Matt Cheselka
> 

As "bjorn" said, it will be freed automatically when the
last reference to it is gone. This means there is a
trap:

a = [] # an empty list
b = [] # another one
a.append(b) # The first element of a will be the list 
  # referred to by b
print a # should be [[]]
b.append(a) # do the reverse thing to b
print b # at least PythonWin is cute enough to not run into
  # an endless recursion here and prints "[[[...]]]"
  
Now each list refers to the other; a less cute print could
run into an endless recursion. 

del a
del b

will help you nothing here. The lists form a cycle,
so the refcounts will never go down to 0 if you don't
do something about it. So they will not be disposed,
although you deleted all references to access them.

Dirk




From Navtej.Riyait at ukgateway.net  Mon Jan 24 18:51:08 2000
From: Navtej.Riyait at ukgateway.net (Navtej Riyait)
Date: Mon, 24 Jan 2000 23:51:08 -0000
Subject: Python CGI and Accessing MS QL Server Database
Message-ID: <86ioek$iek$1@lure.pipex.net>

Hello,

I am trying to create a little test web application in python/HTML using
IIS. I have setup IIS so that .cgi scripts are sent to the python
interpreter. I am trying to access a SQL Server database, 'pubs' (using
Python DB APIs). I am importing odbc, dbi in my script. Database access
works fine when I run a dbtest script at the command line, but the same bit
of code gives an error when run in the .cgi script. It is not finding the
data source name (DSN). I have set up the DSN in the ODBC Driver Manager
correctly.

I can't see what environment parameters to set to get the .cgi python script
to access the 'pubs' database. Can anyone help ?


Thanks





From tino at wildenhain.de  Mon Jan 24 04:14:17 2000
From: tino at wildenhain.de (Tino Wildenhain)
Date: Mon, 24 Jan 2000 10:14:17 +0100
Subject: python and www
In-Reply-To: <003501bf65c2$d2fe3880$918ffea9@desktop>
Message-ID: <3816052520.948708857@localhost>


--On Sonntag, 23. Januar 2000 08:56 -0800 Sameer Chowdhury 
<sameerc1 at yahoo.com> wrote:

>
> I want to invoke the python interpreter, pass it my script, and direct
> the script's output as a response to a request made by a client browser.
> Without using Python with ASP, does anyone know how I would go about
> doing that (say using IIS or Apache).  Thanks in advance.
>

all you need is usual:

#!/path/to/python/executable/python

at the top of the file.
(please dont type it literaly ;)

This works for Apache on Unix (naturally)
and as well on Windows.
(with Windows you need the drive-letter c: or something
at the start)

If the script resides in /cgi-bin/script.py
you need only refer to the script. The python
interpreter will be located automatically.
(note: witrh unix you must also set the flags
on the file for "executable")

hth
Tino Wildenhain





From themantis at trojanslair.zzn.com  Tue Jan  4 23:17:47 2000
From: themantis at trojanslair.zzn.com (Black Mantis)
Date: Tue, 04 Jan 2000 20:17:47 -0800
Subject: using python to press keys
Message-ID: <3872C5EA.D004E77D@trojanslair.zzn.com>

Can you use python to "press down" keys? If so, how would you do that?

thanks



From gerrit.holl at pobox.com  Fri Jan 21 07:13:35 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 13:13:35 +0100
Subject: help with rfc822
In-Reply-To: <868pns$ain$1@nnrp1.deja.com>; from drmental@my-deja.com on Fri, Jan 21, 2000 at 05:10:30AM +0000
References: <868pns$ain$1@nnrp1.deja.com>
Message-ID: <20000121131335.A2265@stopcontact.palga.uucp>

drmental at my-deja.com wrote on 948427830:
> Please help a newbie,
> 
> I'm trying to parse an email that is received with poplib, what i need
> out of it is the email address of the sender and the email body.
> 
> I've tried to store the email as a temp file and then use the
> rfc822.Message(tmpfile) and that does not return a Message instance.
> I've tried to use the StringIO and that give me an error.  Can someone
> give me a brief example of how get the email addres out of an email that
> is stored as a string?

I don't know about poplib, but suppose you have a string returned like:
'''From: foo <foo at somewhere.com>
To: bar <bar at somewher.com>
Subject: This is the subject.
X-Mailer: Mutt 1.0i

Hi,

This is a test.

regards,
Gerrit.
'''

Then, you can create a StringIO instance in it using:
sio = StringIO.StringIO(s)

where s is the string returned.
You can pass 'sio' to rfc822.Message now!

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From bwarsaw at cnri.reston.va.us  Wed Jan  5 17:52:05 2000
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Wed, 5 Jan 2000 17:52:05 -0500 (EST)
Subject: Schedule changes for IPC8
Message-ID: <14451.51989.872547.768485@anthem.cnri.reston.va.us>

There have been a few minor scheduling changes for the Tuesday and
Wednesday sessions of IPC8.  Please see the web site

    http://www.python.org/workshops/2000-01/

for details.
-Barry



From emmanuel.pietriga at xrce.xerox.com  Tue Jan 11 02:07:46 2000
From: emmanuel.pietriga at xrce.xerox.com (Emmanuel Pietriga)
Date: Tue, 11 Jan 2000 08:07:46 +0100
Subject: JPython serialization problem
References: <387989FF.658A8F68@xrce.xerox.com> <iac66x1r2jp.fsf@vier.idi.ntnu.no>
Message-ID: <387AD6C1.C6C634A9@xrce.xerox.com>

"Magnus L. Hetland" wrote:

> Emmanuel Pietriga <emmanuel.pietriga at xrce.xerox.com> writes:
>
> >
> > At first I thought the paths were defined incorrectly, but if I create
> > an instance of the same class directly on the python side using its
> > constructor, it works!  Therefore JPython knows where to find the class
> > definition.
>
> Have you tried importing the class before deserializing the object?
> (Just a hunch - may be useless...)

Yes I did. And I even instanciated it  (v=V_struct()) before reading any
object from the socket.

Doesn't work.

Thanks anyway.







>
>
> >
> > Anyone can help?
> >
> > Thanks.
> >
>
> --
>
>   Magnus
>   Lie
>   Hetland                                mailto:<anything>@hetland.org

--
emmanuel.pietriga at xrce.xerox.com               |  Xerox Research Centre Europe
Document Model and Transformation Technologies |  6, Chemin de Maupertuis
+33 4 76 61 50 32  (direct)                    |  38240 Meylan, France
+33 4 76 61 50 50  (switchboard)               |  http://www.xrce.xerox.com





From cfelling at iae.nl  Sat Jan 29 18:28:26 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 30 Jan 2000 00:28:26 +0100
Subject: Case Sensitivity and Learnability
References: <000501bf6a9f$8f501e00$78a0143f@tim>
Message-ID: <86vt2q$2fl$1@vvs.superst.iae.nl>

Tim Peters <tim_one at email.msn.com> wrote:
> [Carel Fellinger]
>> this would make it impossible to import library modules that do depend
>> on case distinction. Or would it be a per module switch, settable in
>> the source?  Now that will improve things:)

> Relying on case-sensitive file names is already broken in *portable* Python,
> as some filesystems don't support the notion (e.g., while NT's NTFS
> *preserves* filename case, system functions ignore it).  And Python had to
> grow the PYTHONCASEOK envar switch for filesystems that don't even preserve
> case.

us oldtimers know that to be portable means to be extremely carefull with
file names, sometimes the length is severely restricted, allowed chars
differ widely and so on. A dreadfull mess, so please don't invent another!

>> distinct classes from variables:), but then in natural language
>> we use this same trick to distinguish persons from objects like
>> in Bill and bill.

> As in Bill of Materials, or

>> groetjes, carel

> <wink>? ...

the lowercase in my name is a remnant of the days that people thought
of me as a hippy, I lowercased everybody to show that respect didn't
come for free, or something equally vague:)

>...  The problem seems more that in programming we use a great many
> compound "made up" words, and it can be a real bitch to remember e.g.
> whether some function is named moveUp or MoveUp or moveup.  Some moron API

True, but likely you forgot the exact order or names of the parameters
as well, so you better look it all up in one go then:) Besides, this is not
particulairy a newbie problem. Newbies don't get it that case has meaning.
I noticed this very often with 'fresh' informatics students, but then they
didn't appreciate the indentatian thing either:) I guess it takes reading
a lot of others code to appreciate indentation, proper documentation and
above all reasonable naming schemes that allow us to follow math practice
like "for n in N" (winky).

> ever-use-a-case-sensitive-web-search?-ly y'rs  - tim

yes
-- 
groetjes, carel


From glen at electricorb.com  Fri Jan 28 07:18:40 2000
From: glen at electricorb.com (Glen Starchman)
Date: 28 Jan 2000 06:18:40 -0600
Subject: Lexing in Python 2
References: <001501bf63f3$3c018ac0$ec2d153f@tim> <388B855A.78D6E46B@prescod.net>
Message-ID: <38918815.E21BC7C2@electricorb.com>

even though the code is a bit hairy (blatant understatement!), the tokenize
module works very well with a little bit of tweaking. I am currently using it for
a handcrafted language that is translated into python. I had previously tried
kjBuckets (good but SLOW) and pyBison (I think that is the name == never got it
to work quite the way I wanted).

However, I would stand on my feet and cheer for a nice, open (meaning not tied
directly into python lexing rules) lexer to be part of python2.


Paul Prescod wrote:

> Tim Peters wrote:
> >
> > Yes, but that's really a different topic.  The Python world has no good
> > approach to that now, paying attention to the "fast" part, and where "good"
> > means "enough like Flex and Bison that you don't feel you've been stranded
> > on some strange alien planet" <wink>.
>
> At the last XML conference I told someone that the reason that re
> doesn't take a stream instead of string parameter was because anyone
> sane working on a large file would use a proper tokenizer. Shouldn't
> such a tokenizer come with Python? With all due respect, what the hell
> is shlex and how did it get into the standard distribution?
>
> I mean the standard distribution alone must contain half a dozen
> hand-coded lexers and in a few places, the weirdness you need to apply
> regular expressions to streams. Surely we can do better for Python 2?
>
> It is my unconsidered, uneducated opinion that lexers do not vary as
> widely as parsers (LL(1), LR(1), LR(N) etc.) so we could just choose one
> at random and start building modules around it.
>
> All in favor? Opposed? Carried.
> --
>  Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
> Earth will soon support only survivor species -- dandelions, roaches,
> lizards, thistles, crows, rats. Not to mention 10 billion humans.
>         - Planet of the Weeds, Harper's Magazine, October 1998



From aahz at netcom.com  Tue Jan 25 06:53:25 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jan 2000 11:53:25 GMT
Subject: using threads with python and winnt
References: <000801bf670c$dab7a480$632d153f@tim>
Message-ID: <86k2rl$92k$1@nntp9.atl.mindspring.net>

In article <000801bf670c$dab7a480$632d153f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>There's an exceedingly rare timing hole in the released 1.5.2 Python (fixed
>in the CVS version) that had been there for years, showing up in rare
>reports under extremely high rates of thread death and only on top-end
>multiprocessor systems.

We're still using 1.5.1 (no comments, please! we're supposed to finally
upgrade this week), and I was wondering if you could explain what the
symptom of this bug is, because we've seen a couple of recent cases
where a heavily threaded app that creates hundreds of single-shot
threads dies with a thread error.

(We're *also* fixing it to use a thread pool, but I'd still like to know
what our chances are of hitting this with the stock 1.5.2)
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From tismer at appliedbiometrics.com  Sat Jan  1 13:44:30 2000
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Sat, 01 Jan 2000 19:44:30 +0100
Subject: pickle problem:cant reload object saved with binary option
References: <386e1de1.998928692@news-server> <YHpb4.914$Sr1.104202@typhoon.nyroc.rr.com> <386e31c2.1004019140@news-server>
Message-ID: <386E4B0E.C9DD0BF@appliedbiometrics.com>


Marshall wrote:
> 
> Thanks, now I have a new problem.  It still works without the binary
> option.  I think I have the organization of my data screwed up
> somehow, but I don't understand why it works without the binary, and
> breaks if I turn it on.  It doesn't matter if I use pickle or cPickle,
> I still get errors.  The object I'm pickling is self.info=crccvars()
> Thanks in advance for any insight.

No-no-no! :-)
Please understand the following:
Whenever you use the binary (c)Pickle format under Windows,
you *always* must use a binary mode for your file, *both*
for reading and writing.

The reason is: Binary pickle makes use of all 256 characters.
Since Dos/Windows has this carriage return/linefeed (CR/LF)
convention, reading a binary file as a text file turns a CR/LF
into one LF char for Python. And writing binary data in text mode
acts similar: Every single LF char is turned into a CR/LF sequence.

Here the simple rule:
Whenever you use binary pickles, open the file for writing with
mode "wb" (or w+b if you want to append). Open it for reading
with mode "rb". That's all about it.

Any combination of text and binary mode should be strictly
avoided. It can work in few rare cases but will fail quickly.
Since emission/recognition of LF is dependent from your
actual data, things which work today will break tomorrow.

>         def save(self,file='crcc.sav'):
>                 blah=open(file,'w')
@@@ look this is the bad line :-)
>                 cPickle.dump(self.info,blah,1)
>                 blah.close()
>         def load(self,file='crcc.sav'):
>                 print 'print self     :',self
>                 print 'print __name__ :',__name__
>                 print 'print self.info:',self.info
>                 print ''
>                 blah=open(file,'r+b')
@@@ better use "rb" since you don't want to append.
>                 self.info=cPickle.load(blah)
>                 blah.close()
>                 self.buildlist()

A last addition: On Unix, it is not necessary to distinguish
between text and binary mode, since no conversions take place.
But in order to write system independant applications, it is
a good idea to always use those explicit filemodes. Under
Unix, they just have no effect. With Windows or the Mac,
they prevend your above trouble.

Wishing you a trouble-free new millennium - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From nnever50 at yahoo.com  Tue Jan 25 05:24:30 2000
From: nnever50 at yahoo.com (nathan)
Date: Tue, 25 Jan 2000 11:24:30 +0100
Subject: delete files?
Message-ID: <86jtm2$7cs$1@news1.sunrise.ch>

Hello
somebody knows how can I quickly delete files (without using os.system..) ?
Working under WinNT, v.1.5.2.
Thank you

nat




From slinkp23 at yahoo.com  Fri Jan 21 00:23:21 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Fri, 21 Jan 2000 00:23:21 -0500
Subject: Realtime scheduling??
References: <38858064.CDA4866C@yahoo.com> <oqbt6hyic5.fsf@titan.progiciels-bpi.ca>
Message-ID: <3887ED49.1099AF05@yahoo.com>

Fran?ois Pinard wrote:
> So, I just set the scheduler so it runs after real-time, instead of using
> the scheduler to insert theoretical delays.  That is, you maintain a logical
> real time in your program,....

I'm afraid I don't understand this term. What is "logical" real
time?

>... and when about to ask for a delay, you delay
> the logical real time plus the wanted delay minus the actual real time.
> The effect is that you automatically catchup on lags, and very surprisingly,
> it works well.  The musical behaviour is often acceptable even under heavy
> machine load.  Of course, it is perfect (or not far from) with light load.

I'd be very interested to see your code! Is it publically available?

I browsed around your website, but couldn't find anything that
looked relevant...

thanks very much,

PW 

................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
A member of ARMS    ----->    http://www.reacharms.com
or http://www.mp3.com/arms or http://www.amp3.net/arms
personal page   ---->    http://www.ulster.net/~abigoo


From roy at popmail.med.nyu.edu  Sat Jan  8 21:25:59 2000
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Sat, 08 Jan 2000 21:25:59 -0500
Subject: fastest way to merge lists.
Message-ID: <roy-D565B0.21255908012000@netnews.nyu.edu>

I've got 5 (unsorted) lists of strings.  Each list has O(10^4) strings 
in it.  Within each list, each string appears exactly once.  Also, there 
is a large overlap between the lists.

I want to form a single list which contains exactly one copy of each 
string from the combined lists.  In unix, this would be:

cat list[12345] | sort | uniq

I'm not yet sure if care if the final list is sorted or not.  For now, I 
think it doesn't matter.  What's the fastest way to do that?  It seems 
to me:

d = {}
for s in list1:
   d[s] = 1
for s in list2:
   d[s] = 1
for s in list3:
   d[s] = 1
for s in list4:
   d[s] = 1
for s in list5:
   d[s] = 1
list = d.keys()
del d   # if I care about memory

would be pretty good.  Anything better?

Now, if I want it sorted, I could just sort the above list.

I suspect that if I knew ahead of time I wanted the final list sorted, 
sorting each individual list then doing a linear merge might make more 
sense, if there was little overlap between the lists.  But, since I know 
they overlap greatly (i.e. len(list) will be approximately the same as 
len(listN)), I'm guessing doing it as described above and sorting the 
result might just work out faster.


From markk at paritech.com.au  Mon Jan 24 05:47:15 2000
From: markk at paritech.com.au (Mark Klink)
Date: Mon, 24 Jan 2000 21:17:15 +1030
Subject: License Question: Embedding Python
Message-ID: <388c274e@news.access.net.au>

Hello.

I want to embed Python in a commercial application (binary only) with
extensions and a scintilla based debugger. The Python License requires that
the copyright and permission notices in the original code appears in
supporting documentation.  Thus, the documentation will include the
following text:
"Permission to use, copy, modify, and distribute this software and its
documentation for any purpose and without fee is hereby granted, provided
that the above copyright notice appear in all copies and that both that
copyright notice and this permission notice appear in supporting
documentation ..."

Could you help me understand how I can satisfy the spirit of the Python
License requirements without comprising the commercial nature of the product
or confusing the copyright of the overall application where we do not give
permission to copy, modified and distribute the overall application?

Thankyou,
Mark.





From bsb at winnegan.de  Tue Jan 11 16:15:19 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 11 Jan 2000 22:15:19 +0100
Subject: Please, someone tell me what's going wrong here. ;>
In-Reply-To: "Jason Maskell"'s message of "Tue, 11 Jan 2000 19:32:54 GMT"
References: <GzLe4.54805$Ke.42098@news21.bellglobal.com>
Message-ID: <87hfgkl388.fsf@baal.winnegan.de>

Hi Jason,

your code looks like Borland C++ on WinXX to me, you'd better tell us which
environment you're using :)

"Jason Maskell" <backov at nospam.csolve.net> writes:

> I posted a while back on embedding, and after scouring deja and reading the
> demos, I can't even get this simple example to work. It's essentially a C++
> Builder'ized version of the simple embedding demo.. PyRun_SimpleString
> returns -1 and foo is never executed.
> 

[...]

extern "C" {

> void initxyzzy(); /* Forward */

[...]

> void initxyzzy()
> {
>  PyImport_AddModule("xyzzy");
>  Py_InitModule("xyzzy", xyzzy_methods);
> }

}

[...]

HIH
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
                  bsb at north.de - http://www.north.de/~bsb/
****** ceterum censeo javascriptum esse restrictam *******



From orlov at diasoft.ru  Wed Jan 12 04:52:25 2000
From: orlov at diasoft.ru (Oleg Orlov)
Date: Wed, 12 Jan 2000 12:52:25 +0300
Subject: Python BUG?
References: <85hg1q$2nli$1@gavrilo.mtu.ru>
Message-ID: <85hiou$2ohr$1@gavrilo.mtu.ru>

Thanks. It helps.
My right way was:

d = [1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L,
    10L, 9L, 8L, 7L, 6L, 5L, 4L, 3L, 2L, 1L]
d = map(lambda x: x*x, d)
print reduce(lambda x, y: x + y, d)

or even
print reduce(lambda x, y: x + y*y, d, 0)

 --
 ====================
 Oleg Orlov
 orlov at diasoft.ru







From felixt at dicksonstreet.com  Sat Jan 22 17:22:57 2000
From: felixt at dicksonstreet.com (Felix Thibault)
Date: Sat, 22 Jan 2000 16:22:57 -0600
Subject: Suggestions for python 2
In-Reply-To: <863ka6$7ej$1@nntp9.atl.mindspring.net>
References: <000501bf623e$18b56a20$9a2d153f@tim>
Message-ID: <3.0.5.32.20000122162257.009af7f0@mail.dicksonstreet.com>

>In article <000501bf623e$18b56a20$9a2d153f at tim>,
>Tim Peters <tim_one at email.msn.com> wrote:
>>[Evan Simpson]
>>>
>>> I rather like the idea of having to explicitly bring names
>>> into an inner scope,
>>
>>Me too -- except in the cases that bite most often, where a trivial little
>>lambda wants to refer to one or two measly little names from its context.
>>People stumble on that so often it's a FAQ.
>

How hard would it be to implement something like:

def like_a_list(*now_args):
    def return_func(!now_args, *later_args, **show):
        if show:
            return list(now_args+later_args)
        else:
            return apply(like_a_list, now_args + later_args)
    return return_func

where the ! (or whatever) means something like this happens:

def like_a_list(*now_args):
    """like_a_list(value0, value1,...) store some values in a function"""
    fun_str = """def return_func(*later_args, **show):
	now_args = %s
	if show:
		return list(now_args+later_args)
	else:
		return apply(like_a_list, now_args + later_args)\n"""\
		 % (now_args,)
    exec(fun_str)
    return return_func

when the inner function is defined ? (Except that it works when
eval(repr(object)) doesn't and it's not as ugly). 



Felix




From bernhard at intevation.de  Tue Jan 11 08:13:57 2000
From: bernhard at intevation.de (Bernhard Reiter)
Date: 11 Jan 2000 13:13:57 GMT
Subject: XML-RPC Server
Message-ID: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de>

Please enlight me again:

Does anybody have a simple XML-RPC server example flying around?
(Yes I know that Zope speaks XML-RPC. :) )

I know it should be hard to do, someone must have done it.
Just a couple of simple objects answering towards XML-RPC requests.

And I also appreciate and update on SOAP client and server python
implementation efforts.

ADVthanksANCE-ly, yours
	Bernhard

-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From jeremy at cnri.reston.va.us  Thu Jan 20 14:37:46 2000
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 20 Jan 2000 14:37:46 -0500 (EST)
Subject: [PSA MEMBERS] getpass .. bug ?
In-Reply-To: <1000120110301.ZM48480@noah.scripps.edu>
References: <1000120110301.ZM48480@noah.scripps.edu>
Message-ID: <14471.25610.767260.153081@bitdiddle.cnri.reston.va.us>

>>>>> "MFS" == Michel Sanner <sanner at scripps.edu> writes:

  MFS> Hi, just waned to mention the when the optional argument
  MFS> |prompt| is used in getpass from the getpass module, the
  MFS> passowrd is echo'ed to the terminal.

  MFS> When the function is called without any argument it works as
  MFS> expected.

  MFS> This was observed with Python1.5.2 on sgi and sun. The PC under
  MFS> windows works as expected.

This is strange.  I do not see the reported behavior on my Sparc.  And
a glance at the code tells me that the prompt argument has no
significant control dependency; if you don't supply a prompt, the
default 'Password:' is used.

Jeremy





From neelk at brick.cswv.com  Sun Jan  2 10:30:00 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 2 Jan 2000 15:30:00 GMT
Subject: Pseudo-SQL: searching for Parser/Rewriter
References: <16600378@oberon.noris.de>
Message-ID: <slrn86uroa.fph.neelk@brick.cswv.com>

hartmut Goebel <hartmut at oberon.noris.de> wrote:
> Hello everyone,
> 
> I'm looking for a general purpose tokenizer/parser/rewriter for
> Python. If a parser is available, it best should return a simple
> parse tree.

Try John Aycock's SPARK, the Scanning, Parsing, and Rewriting Kit:

   <URI:http://www.csr.UVic.CA/~aycock/python/content.html>
 
It implements an Early parser in pure Python, so it's very powerful,
very elegant, and not very fast. But if your texts are of moderate
size it should be fine.


Neel


From martinp at mincom.com  Wed Jan 12 02:21:17 2000
From: martinp at mincom.com (Martin Pool)
Date: Wed, 12 Jan 2000 17:21:17 +1000
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <3875099A.2CF9333C@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net> <3876B79F.30439283@mdlink.de> <856k4d$7e9$1@nntp5.atl.mindspring.net> <3879074F.1A4EDED3@mdlink.de>
Message-ID: <387C2B6D.3FFECBD1@mincom.com>

Helge Hess wrote:

> I can't see on what you base this statement on ?! This capability
> generalizes, not specializes, the environment - which leads me to the
> assumption that the *current* state will make programmers wonder.
> 
> Until now I haven't heard a reason against this 'feature' which I can
> follow. I guess that it is just a matter of taste.

Here's my argument: one of Python's strengths is that despite being
quite flexible, it also generally produces code which is transparent to
the reader.  Factory methods seem to me like they would solve your
problem and are clearer than constructors that don't really construct.  

If I write

  import mangler
  o = mangler.NetworkMangler(22)

and know that NetworkMangler is a class, then I can feel pretty sure
that I've just created a new instance.  Of course that could be a method
call, or indeed any other kind of callable thing, but in general it's
pretty clear.  

If that class wants to offer a singleton or an instance from a pool then
the interface ought to be like this:

  o = mangler.mangler(22)

which makes it clearer that I'm not necessarily getting a new instance. 
This idiom seems pretty well established.

It seems to me that one point in favour of your proposal is that it
hides from the caller the details of whether they're getting a new
instance or not.  I don't remember ever caring in Python about whether
`a is b' rather than `a == b' (though it often happens in Java), but
still it seems like something the client ought to be allowed to rely
on.  I guess it also prevents callers from accidentally creating new
instances when they're not supposed to, though the flavour of Python
seems to be that the documentation should just say "don't do that."

My main criticism is that it seems to lead off down the C++ path of
adding individually harmless features that in sum mean it's hard to tell
what to tell what any single statement does.

Aahz wrote:
>  __init__ gets called by the constructor to
> ensure that the returned object is in a valid state, and it seems
> reasonable to me that __init__ cannot substitute a different object for
> the constructor to return.

What's supposed to happen if somebody calls super.__init__(self)?  Does
all the code that does that have to check if the superclass returned
something else?  What if the subclass already started initializing
things before it called the superconstructor.  What if in a
multiply-inherited class the two constructors _both_ wanted to return
different objects?  (Python seems happy to arbitrarily resolve conflicts
like this, but still it seems bad to introduce them when they're not
necessary.)

Helge wrote:

> a = MyString(data="sadfjsadf", encoding='utf8')

Why not 

  a = mystring(data='sadlksad", encoding="utf8")

and that can return any instance of MyString that it likes.  (Hmm, maybe
I've been in Java for too long and am too used to using factories to
work around limitations.)

As a humble programmer I'm very happy that Guido adds features
cautiously.

-- 
Martin

`illegal copying' as opposed to `piracy'. ie, changing ones and
zeroes as opposed to pillaging, rape and murder on the high seas.
				-- aj


From skip at mojam.com  Wed Jan 26 16:27:17 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 26 Jan 2000 15:27:17 -0600 (CST)
Subject: CGI: keeping mySQL awake between hits?
In-Reply-To: <86nn6m$tkg$1@nnrp1.deja.com>
References: <86nn6m$tkg$1@nnrp1.deja.com>
Message-ID: <14479.26293.570865.55435@beluga.mojam.com>

    Jim> I have CGI scripts that want to write to a database.  If I have
    Jim> that every hit causes Python to wake up, and then mySQL to wake up,
    Jim> so that the script can write to the database (and then they both go
    Jim> back to sleep) then the response is very slow.

Your performance hit comes mostly from two sources: Starting the CGI script
and connecting to your mySQL server.  Best way to get around that is to
dispense with CGI.  You already are aware of at least one Python/Apache
module.  You might check the Vaults of Parnassus (there's a pointer on the
python.org search page) for others.  I think there are two or three.

Also, check out both medusa and ZServer.  ZServer is built on top of medusa
and is part of Zope (http://www.zope.org/).

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From cpatti at black-racer.atg.com  Wed Jan  5 17:09:56 2000
From: cpatti at black-racer.atg.com (chris patti)
Date: 05 Jan 2000 17:09:56 -0500
Subject: htmllib question - Is this the best tool for my task?
References: <84ecvh$mh2$1@nnrp1.deja.com> <84efnr$rgl$1@nntp9.atl.mindspring.net> <84g50u$rtf$1@nnrp1.deja.com>
Message-ID: <yu1so0cw4p7.fsf@black-racer.atg.com>

feoh at cosmic.com writes:

> Folks;
> 
> I have a task where I need to figure out if certain HTML tags (Rather
> Java Server Pages tags, but they're still HTML, kinda :) contain certain
> things in an HTML document I'd feed to the script.
> 
> I'm thinking htmllib is the right tool for this task, am I wrong?
> 
> So as far as I understand things thus far, I should subclass HTMLParser,
> define start_java and end_java tag functions.. But where from there?
> 
> How do I get a hold of the text within the tag I'm processing? None of
> this is clear from the htmllib doc.
> 
> Thanks!
> -Chris P.

To answer my own question - I ended up using regexps for this.

htmllib was overkill.

-Chris


From eddyw at lsl.co.uk  Fri Jan 21 12:14:58 2000
From: eddyw at lsl.co.uk (Edward Welbourne)
Date: Fri, 21 Jan 2000 17:14:58 +0000
Subject: In defence of the two-namespace rule
Message-ID: <E12BheE-0005Zo-00@lsls4p>

Haskell B. Currie was a mathematician who made the simple observation
that a function of many arguments may be described systematically as a
function taking one argument, the return from which is in turn a
function taking the next argument, and so on until you get a function
which takes the last argument and returns the answer you wanted.
This is a great simplifier for mathematicians discussing algebra.

In computing, his first name graces an excellent `functional
programming' language - which, unlike some of its kin is also a
functional `programming language' - and his second name is used to
describe the operation (e.g. binding a method to an instance) of
combining a function and a value to form a function which takes one
argument fewer than the other, using the value just given as the first
argument, so as to make up the missing one.

And while I grant the os.getcwd() one may be dopey, there are places
where a user interface specifies that `if the user supplies a relative
path for this input, it will be interpreted as a path relative to
/blah/stuff/' in which one may wish to supply a tool-function to the
dialog through which the user provides their input - the tool-function
has the job of resolving this spec - so use '/blah/stuff' in place of
os.getcwd(); more likely, use os.environ.get('blahrc', '/blah/stuff').
Since the dialog is going to call this function quite a lot (e.g. it's a
file selection dialog, so it's going to allow directory-selection on a
widget and build lists of things the user might call, etc.), we don't
want to recompute blahrc, so we do it once.

As to passing the default through globals, the problem then gets
complicated by the fact that, of course, that dialog gets used by loads
of bits of code, so someone writes this helper:

def relativeto(root):
    def result(*fragments, r=root):
        return apply(os.path.join, (r,) + fragments)
    return result

which doesn't have the option of passing root through globals.  Either
you need to transcribe it, as shown, or you need to remember the
namespace relativeto() used in each of its runs for as long as you kept
its result.  Not a big problem with relativeto, but then if I go giving
an example in which you would really have a huge mess on your hands, the
example is going to be culled from some general-purpose context and folk
are going to complain that they can't follow what's going on.  Examples
depend on the reader imagining how much more complex the situation might
be in real-world code instead of nice easy-to-follow code.

	Eddy.
--
Uh-oh - pervect alert ;^>



From cwr at crash.cts.com  Sun Jan 23 09:03:49 2000
From: cwr at crash.cts.com (Will Rose)
Date: 23 Jan 2000 14:03:49 GMT
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim>
Message-ID: <86f1o4$am8$1@thoth.cts.com>

Tim Peters <tim_one at email.msn.com> wrote:
[...]
: As Michael Hudson said, the connection between colons and blocks was adopted
: from Python's closest predecessor, ABC.  Formal usability studies showed
: that while indentation was less confusing than assorted bracketing
: constructs to newbies, the combination of indentation and an introductory
: colon was even better.  You're right that the grammar has little use for it;
: although, pragmatically, it's a great aid to semi-stupid tools (like Python
: editor modes -- not to mention all us Python programmers <wink>).

Cool.  But if Python is designed for newbies, does that mean we all have
to give up using it after a month (or three months, in my case) when we
actually have some experience with it?

There's certainly a niche for learning languages: but is that really where
Python is headed?  From the libraries and tools available, I hadn't
thought so.


Will
cwr at cts.com




From emmanuel.pietriga at xrce.xerox.com  Mon Jan 10 02:27:59 2000
From: emmanuel.pietriga at xrce.xerox.com (Emmanuel Pietriga)
Date: Mon, 10 Jan 2000 08:27:59 +0100
Subject: JPython serialization problem
Message-ID: <387989FF.658A8F68@xrce.xerox.com>

My problem is rather "hard" to explain, so I'll take an example:


I have two applications which communicate using a socket:
One is written in Java and sends serialized objects to the other one,
written in python, which uses java classes (therefore it runs under
JPython).
The python application receives serialized objects from the java
application.

I use java.io.ObjectOutputStream.writeObject() on the java side, and
ObjectInputStream.readObject() on the python side.

Of course, the objects I send all implement the Serializable interface.

Here's the problem:
    if I send an instance of a class which is part of the standard Java
API  (like Vector, Hashtable, or even awt.Label,.....) everything is
fine.
    but if I send objects of a class created by me, no matter how simple
the class is, I get a NoClassDefFound exception on the python side when
reading the serialized object from the objectinputstream.

At first I thought the paths were defined incorrectly, but if I create
an instance of the same class directly on the python side using its
constructor, it works!  Therefore JPython knows where to find the class
definition.



Anyone can help?

Thanks.



From eleisonme at juno.com  Tue Jan 11 10:17:15 2000
From: eleisonme at juno.com (eleisonme at juno.com)
Date: Tue, 11 Jan 2000 15:17:15 GMT
Subject: Newbie simple question - can't find answer
Message-ID: <85fhha$6js$1@nnrp1.deja.com>

Hello all,
I am using Python 1.5.2-7 on intel Red Hat Linux 6.1.
I do not seem to be able to use any of the math or cmath functions
either in a script or on the python command line.

For example if I type:
>>>import math
>>>exp (4)

OR

>>>exp(4)

I get a traceback error:
NameError: exp (or any other math function that I try to use).

I looked in the lib-dynload directory and I do have cmathmodule.so and
mathmodule.so present.  However they do not come up when I do a command
to see what is being dynamically loaded.  In any case, I would expect a
module missing error rather than a name error.  My suspicion is that the
lib-dynload files are not linked.  Any help on how to go about fixing
this would be appreciated.  Can't seem to find anything in the
documention - so much for rtfm!

Thanks,
Rachel
Educational Computing
Harvard Medical School


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


From akuchlin at mems-exchange.org  Wed Jan  5 14:26:52 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 05 Jan 2000 14:26:52 -0500
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk>
Message-ID: <3dzouke2v7.fsf@amarok.cnri.reston.va.us>

Bala <engbpa at liquidinformation.com> writes:
>         Python Essential Reference (OTHER NEW RIDERS)
>         by David M. Beazley, Guido Van Rossum
> or
>         Learning Python (Help for Programmers)
>         by Mark Lutz, David Ascher, Frank Willison (Editor)

The Essential Reference is just what its title says: a reference
guide.  You'd have to be *really* motivated to learn Python from it.
I wrote a review of it, located at
http://starship.python.net/crew/amk/python/writing/books.html .)

The true competition to _Learning Python_ is the _Quick Python Book_;
check other threads in comp.lang.python for a discussion of it.  I
think both LP and QPB are pretty good; I'll review QPB as soon as I
finish reading the appendix.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
He spoke to the embryonic silicon dreams who clustered in a far ballroom, and
whispered to them, briefly, about the other machines that had dreamed in the
distant past.
    -- From a week in Dream's life, in SANDMAN #64: "The Kindly Ones:8"




From robin at jessikat.demon.co.uk  Sun Jan  9 02:16:22 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 9 Jan 2000 07:16:22 +0000
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu>
 <slrn87fol9.15t.kc5tja@garnet.armored.net>
Message-ID: <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk>

In article <slrn87fol9.15t.kc5tja at garnet.armored.net>, Samuel A. Falvo
II <kc5tja at garnet.armored.net> writes
...
I tried com for many years without coming close to understanding it. If
M$ continues to have huge leverage it will continue to be arcane and for
wizards only! Python at least allowed me in without a lot of exotic
verbiage. I'm not a whizz at COM, but Bill Tutt, Hark Hammond and the
other COMerati regularly throw up their hands at the complexity and
opacity of it all.
-- 
Robin Becker


From hei at adtranzsig.de  Wed Jan 26 05:48:12 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Wed, 26 Jan 2000 11:48:12 +0100
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <948807929.472741@news.adtranzsig.de> <Thjj4.6690$S64.195053@nnrp1.uunet.ca>
Message-ID: <948887438.236049@news.adtranzsig.de>

Warren Postma schrieb in Nachricht ...
>Rather bloody minded of me to point this out, but hey "to each his own"...
>
>#define BEGIN {
>#define END     }

Of course; now convince people to *use* this throughout their programs.
Until a committee can declare "{" and "}" as obsolete.

>
>Can I do the same somehow in Python?  ;-)
>

No. If you need optical block markers, i suggest this:

def blah():
  #begin
  ...
  #end

Makes the code look nearly like Pascal; except for you don't get
the "semicolon"-errors. Ain't that cute.
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net





From cjensen at be-research.ucsd.edu  Thu Jan 27 18:45:13 2000
From: cjensen at be-research.ucsd.edu (Curtis Jensen)
Date: Thu, 27 Jan 2000 15:45:13 -0800
Subject: NumPy
References: <388DC152.9624D8F5@obop.com.pl>
Message-ID: <3890D889.85FD92E3@be-research.ucsd.edu>

"Przemys?aw G. Gawro?ski" wrote:
> 
> [ WinNT 4.0 SP 5, Python 1.5.2]
> 
> I've just downloaded Numerical-15.2 and tryied to run the setup.py
> script but I get:
> Tracebak (innermost last):
>     File "setup.py", line 16, in ?
>         from distutils.command import install
> ImportError: No module named distutils.command
> 
> am'I missing something ?
> 
> --
> Przemyslaw G. Gawronski UIN:8358522
> mailto:gawronskip at usa.net
> mailto:gawron at obop.com.pl

With the newer versions of Numeric, you have to install the Distutils. 
It is not bundled with the Python instalation.  It can be found at:
http://www.python.org/sigs/distutils-sig

-- 
Curtis Jensen
cjensen at be-research.ucsd.edu
http://www-bioeng.ucsd.edu/~cjensen/
FAX (425) 740-1451


From tye4 at yahoo.com  Tue Jan 18 19:41:06 2000
From: tye4 at yahoo.com (tye4)
Date: Tue, 18 Jan 2000 16:41:06 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net>
Message-ID: <8631bt$1nu@news.or.intel.com>

> >Tom Culliton <culliton at clark.net> wrote in message
> >news:YZRg4.1292$Ve7.28796 at iad-read.news.verio.net...
> >>tye4 <tye4 at yahoo.com> wrote:
> >Are you insinuating that this is not a problem since I'm not as
> >experienced as Guido?
>
> First if had I intended insinuation it would have been more subitle
> than that.  Second your experience relates to how much weight we're
> going to assign to your _opinion_.  Indentation isn't a problem for me
> or for most of other people around here, whose opinions and experience
> I and the other folks here hold in much higher regard.
>
> >I don't need to design a language to detect flaws in it.
>
> Your opinion or prejudice doesn't make it a flaw.  Having designed a
> widely used language would add weight to your opinion.  BTW - If you
> want to wave Nick Wirth at us, I think he tends toward baby talk in
> his designs.

So you don't think this is a problem?? Hello!, read all the messages in this
thread. Seems to me a lot of programmers have a problem with this.
Thanks to this quirk, I now have to worry how an editor saves tabs to files.
I have offered you many reasons why we need 'end' block support..
You have not offered a single reason why this is not a problem..

-tye4




From ivanlan at callware.com  Fri Jan 21 14:28:32 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 21 Jan 2000 12:28:32 -0700
Subject: ConfigParser and .ini files
References: <86a1s2$7ha$1@nnrp1.deja.com>
		<14472.39533.223900.313932@weyr.cnri.reston.va.us>
		<38889FB1.F692A45D@callware.com> <14472.44829.865615.286152@weyr.cnri.reston.va.us>
Message-ID: <3888B360.EAE0A804@callware.com>

Hi All--

"Fred L. Drake, Jr." wrote:
> 
> Ivan Van Laningham writes:
>  > The spec is how Windows treats .ini files. ...
> 
>   That's a huge problem for those of us who don't often use Windows.
> If anyone has tried to document the format with real details, I'd love
> to get a pointer to the description.
> 

I'll take a look at the windocs later today.

>  > 1)  Section headers of the form "[Button States]" come up as
>  > ParsingError because of the whitespace.  I read the regex for the
>  > headers as allowing whitespace in the header, so what's going on here?
>  > Here's the regex:
>  >
>  >  __SECTCRE = re.compile(
>  >         r'\['                                 # [
>  >         r'(?P<header>[-\w]+)'                 # `-', `_' or any alphanum
>  >         r'\]'                                 # ]
>  >         )
>  >
>  > Not being a regex guru (not even an acolyte), I'm probably unqualified
>  > to evaluate it.  What should this be to allow whitespace in section
>  > names?  [*lots* of .ini files have whitespace in the section names. ...]
> 
>   Add a space before or after the \w in the regex.

Works beautifully.  Thanks.

>   Are there any characters *not* allowed in the section names?  Are
> section names supposed to be case-sensitive?
> 

An informal survey (grepping my winnt/*.ini files) reveals: 
alphanumeric, tabs and spaces, -, _, ., (, ), *, commas, { and }. 
Pretty much any character that is legal in a dos/win filename appears to
be legal as a section name.
 
>  > 2)  The ParsingError lists the lines it thinks are bad at the end of the
>  > run, but they come up like this on the listing:
>  >
>  >         [line  5]: '[Button States]\012'
>  >         [line 23]: '[CallWare Archive]\012'
>  >
>  > Now, is this because ConfigParser thinks it's on UNIX, because it treats
>  > everything as a UNIX file, because it is not expecting the '\r\n'
>  > combination, or because it's broken?
> 
>   This is because the files are opened in text mode, and CR/LF is
> represented as LF internally.  This is normal.  The only error was
> that the space wasn't allowed in the section name, as you pointed
> out.
>

OK, thanks.

<is-it-an-ini-or-an-outi?>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From rob at hooft.net  Mon Jan 31 08:08:01 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: 31 Jan 2000 14:08:01 +0100
Subject: IOError: (0,"Error") when closing "os.popen"ed file?
Message-ID: <m3n1pmmlse.fsf@devel.no.rwd.dinet>

In my project tools, I have a routine "myopen" with the following
code:

compresstable={'.Z':'compress','.gz':'gzip','.bz2':'bzip2'}

def myopen(filename,mode):
    """Open a file. Based on the filename extension, open it through different
       compression/decompression program pipes.

       parameters:
           filename : name of file to open.
           mode : 'r' or 'w' for read-only or write-only open.
    """
    from interface import compress
    if mode=='r':
        for ext,prog in compresstable.items():
            if filename[-len(ext):]==ext:
                return os.popen(prog+' -dc '+filename+' 2> /dev/null' ,'r')
        else:
            return open(filename,'r')
    elif mode=='w':
        for ext,prog in compresstable.items():
            if filename[-len(ext):]==ext:
                return os.popen(prog+' > '+filename,'w')
        else:
            return open(filename,'w')
    raise IOError("unknown filemode?")

Now I can use 
 
   f=projtls.myopen("file.Z","w")

to write compressed data to disk: myopen will open the file as:
 
   f=os.popen('compress > file.Z','w')

This approach has worked until a few weeks ago..... Now we have a new
fast computer running RedHat linux 6.1, and a 450MHz PIII processor.
Suddenly, I started getting strange exceptions from "f.close()" statements
for these "popen"ed files:

   IOError: (0,"Error")

When such an error on close occurs, most of the time the last few kB of
data are missing from the compressed output file. 

After hacking for a while, I managed to work around the problem by
doing:

    time.sleep(0.1)
    f.close()

But this is highly unsatisfactory..... Is there anyone that can give me 
a better suggestion?

Regards,

Rob Hooft
-- 
=====   rob at hooft.net          http://www.xs4all.nl/~hooft/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========


From seanb at home.com  Mon Jan 24 19:36:59 2000
From: seanb at home.com (Sean Blakey)
Date: Tue, 25 Jan 2000 00:36:59 GMT
Subject: Multidimensional Dictionaries: Possible?
In-Reply-To: <slrn88ppsg.49t.w6uv@voyager.home.com>
References: <slrn88ppsg.49t.w6uv@voyager.home.com>
Message-ID: <Pine.LNX.4.21.0001241621140.3546-100000@seanb.sttln1.wa.home.com>

On Tue, 25 Jan 2000, Jerry Gardner wrote:

> 
> Are multidimensional dictionaries possible in Python? I'm sure they're
> possible, but I can't figure out the syntax.
> 
> Here's how I do it in Perl:
> 
> 
> %dict = ();  # create empty dictionary
> 
> 
> $dict{$key1}{$key2} = $string;       # make an entry
> $newstring = ${dict{$key1}{$key2}};  # access an entry
> 
> 
> 
You might want to try something like this:
>>my_dict = {}	# Create the dictionary
>>>try:						# Make an entry
...	my_dict[key1][key2] = my_string	
...except KeyError:
...	my_dict[key1] = {}			# Create the dictionary in key1
...	my_dict[key1][key2] = my_string
...
>>>string_reference = my_dict[key1][key2]	# Access the string

Alternately, if this usage of try/except makes you uncomfortable:
>>>my_dict = {}			# Create the dictionary
>>>if not my_dict.has_key(key1):
...	my_dict[key1] = {}	
...my_dict[key1][key2] = my_string
>>>string_reference = my_dict[key1][key2]

-- 
Sean Blakey
(206)297-7123
quine = ['print "quine =",quine,"; exec(quine[0])"']; exec(quine[0])



From tye4 at yahoo.com  Wed Jan 19 14:48:20 2000
From: tye4 at yahoo.com (tye4)
Date: Wed, 19 Jan 2000 11:48:20 -0800
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com>
Message-ID: <8654iq$fcl@news.or.intel.com>




Hans-Joachim Widmaier <hjwidmai at foxboro.com> wrote in message > Accidently
add or delete a space on a line? To me, it looks like you're looking
> at it the wrong way: The whitespace in front of the line isn't "spaces"
and "tabs",
> it's *indentation*. And it's added und deleted not by hitting <BackSpace>
or <Del>
>
> Every sane programmer indents his or her programs anyway. And all the C
programmers
> I know were bitten by this old example:
>
>   if (a > 3)
>     if (b == 4)
>        c = 2;
>   else
>     c = 3;
>

Perfect example.
   if (a > 3)
     if (b == 4)
        c = 2;
    else
     c = 3;

If you add a space, the 'else' will belong to the inner 'if'. If you delete
one, the 'else' will belong to the outer one. Highly error prone.

> Indentation is a way to show the structure of a program to the human
reader.
> Why do we need a second way to show it to the compiler/interpreter? This
only
> makes the code ambiguos. What was the programmer's intention in the
example
> above?
> I see the structuring by indentation as one of python's strongest points.
>
Don't get me wrong. Indentation tremendously improves readability. However,
it should not be part of the language. That way, we don't have to put an
immense amount of effort just trying get everything into the right block.
There is not single programming language that forces such a bizarre
indentation rule.
If thousands of programmers have a problem with this, why isn't it fixed.
The solution is backward compatible:
if x = 5:
    print 'x is 5'
end if

-tye4




From mfletch at tpresence.com  Sun Jan 23 23:31:40 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Sun, 23 Jan 2000 23:31:40 -0500
Subject: Lexing in Python 2
Message-ID: <CC59868A0A76D311B1F50090277B321C1F309A@VRTMAIL>

What would the requirements for a lexer be?

	Be able to write token definitions with simple, possibly regex-like
or BNF-like definitions?
	Be able to drop in more complex state-table definitions for
harder-to-define constructs.
	Fast operation on "normal" computer-and-human-language tokenisation.
	<Compatibility should be in here somewhere>
	Output as something like (tokentype, startindex, stopindex )
(optional children)

I would guess you could create such a beast with simpleparse + mxTextTools
without breaking too much of a sweat.  Of course, as Tim has pointed out
before, it would be an alien sweat, so lots of ammonia and formaldehyde, but
that's always a risk.  You should be able to save the parser-definition
tables Simpleparse generates so that you only need mxTextTools at run-time
(I believe the tables are now pickle-able).  Note: mxTextTools doesn't
current support token-start-mapping (I think this is called the fastmap
optimisation in regex land).  Using mxTextTools's splitting features might
work better for whitespace determination.

Of course, there's no really useful error handling in these systems, so
you'd have to add that.  Likely also want a line-counting feature (should be
a trivial task, but I've never got around to it).  For real utility, you'd
want to do some hacking on mxTextTools to let it parse streams instead of
strings.

Looking at Chris Tismer's RTF loader, of course, is a good place to start
for designing really fast streaming tokenisers.

Ah well, stuff I don't have time to work on.  Suppose I'll stop rambling
about it. Enjoy yourselves,
Mike

-----Original Message-----
From: Tim Peters [mailto:tim_one at email.msn.com]
Sent: Sunday, January 23, 2000 10:55 PM
To: python-list at python.org
Subject: RE: Lexing in Python 2
...
If nobody was motivated enough to write the code for Python 1, I don't know
why that would change for Python 3000 (that's what Guido insists on calling
it now <wink>).  If you want a *fast* Python lexer today, mxTextTools is
your best hope.

> It is my unconsidered, uneducated opinion that lexers do not
> vary as widely as parsers (LL(1), LR(1), LR(N) etc.) so we
> could just choose one at random and start building modules
> around it.

Curiously, mxTextTools is nothing like lex/flex.  Flex does such a good job
it's hard to get motivated to duplicate all that effort (it's not easy)
solely to get something releasable under a more Python-like license.  I
don't know how Marc-Andre would feel about folding mxTextTools into the
distribution.
...



From duncan at rcp.co.uk  Wed Jan  5 09:31:10 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 5 Jan 2000 14:31:10 GMT
Subject: http POSTing
References: <8EB19A9E7newshoesmailcom@205.147.63.94>
Message-ID: <8EB29A735duncanrcpcouk@194.238.50.13>

newshoes at mail.com (NewShoes) wrote in 
<8EB19A9E7newshoesmailcom at 205.147.63.94>:

>i'm trying to create an offline form i can send to a website
>through "POST". being new to python and minimal knowledge of
>the http protocol, this has stumped me rather quick.
>
>obviously the urllib is out, since it seems i can only
>perform "GET" requests. but i can't figure out if the
>httplib is lowlevel enough, or if i need to go down to sockets
>
>anyways, if someone has an HTML form like this...
>
>
><FORM ACTION="http://www.test.com/form.html" METHOD="POST">
><INPUT TYPE="hidden" NAME="hidd" VAL="en">
><INPUT TYPE="text" NAME="email" VALUE="you at yourmail.com">
><INPUT TYPE="image" SRC=post.gif NAME="POST">
></FORM>
>
>
>how would one go about sending this off if filled out
>from a simple TKinter form??
>(quick example code being what i seek here) :]
>

What is your objection to urllib? It does POST requests just fine.
Try the following:

import string,urllib
def format(items):
    list = []
    for k, v in items:
       list.append("%s=%s" % (k, urllib.quote_plus(v)))
    return string.join(list, '&')

data = { 'hidd':'en', 'email':'you at yourmail.com',
         'POST':'post.gif' }
d = format(data.items())
try:
    sock = urlopen(YourURL, d)
    for l in sock.readlines():
        print l
    sock.close()
    finally:
        urlcleanup()


From robin at alldunn.com  Thu Jan  6 23:51:18 2000
From: robin at alldunn.com (Robin Dunn)
Date: Thu, 6 Jan 2000 20:51:18 -0800
Subject: Mac version of wxPython?
References: <200001061700.JAA15706@smtp2.jps.net> <3874E1E7.1CA2B8A5@jps.net> <01cc01bf587b$defa71c0$1a25d2d1@jenkondev.com> <387530EE.456591C0@jps.net>
Message-ID: <007701bf58ca$d7a29de0$0301a8c0@easystreet.com>

>
> By the way, Robin, since you brought up the Mac, and have been talking
> about money, what do you suppose it would cost for you to produce (and
> support) a Mac version of wxPython?

Okay, Let me open up this "race" to everybody.  The first person or company
to deliver to my doorstep a PowerMac (used is okay) powerful enough for
software development, and appropriate compiler and other software tools will
win the following prizes:  My undying gratitude and a 1 year full support
contract once the Mac version is in a usable state.  Drivers, start your
engines!

<grin>

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!






From mhammond at my-deja.com  Tue Jan 18 12:58:36 2000
From: mhammond at my-deja.com (mhammond at my-deja.com)
Date: Tue, 18 Jan 2000 17:58:36 GMT
Subject: win32pipe missing WaitNamedPiped ?
References: <gmvg4.6200$kv.435556@news.easynews.com>
Message-ID: <8629k2$ghs$1@nnrp1.deja.com>

In article <gmvg4.6200$kv.435556 at news.easynews.com>,
  "Roger Upole" <rupole at compaq.net> wrote:
> I can't seem to find the WaitNamedPipe routine anywhere in the
> win32pipe module.  Am I misspelling it or just looking in all the
> wrong places ?  Any help greatly appreciated.
>

Nope - it should be there, and isnt.  Ive added it here.  If you have
win32all-128 (released, but not announced :-) then I could send you the
new win32pipe - mail me - soon :-)  (Im off to DC in a few days where I
dont expect to have internet access...)

Mark.


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


From gmcm at hypernet.com  Wed Jan 12 12:11:37 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 12 Jan 2000 12:11:37 -0500
Subject: Perl speed vs. Python speed
In-Reply-To: <YZ1f4.26511$W2.359579@iad-read.news.verio.net>
Message-ID: <1264425395-448550@hypernet.com>

Tom Culliton wrote:

> Fredrik Lundh <fredrik at pythonware.com> wrote:

> >if you can reorganize the
> >code to use a single regular expression on a larger string,
> >'re' beats the hell out of 'regex'.  ...and if you make things
> >complicated enough, 'regex' stops working... (but as usual,
> >some people prefer to get the wrong answer quickly ;-)
> 
> Big complicated regular expressions are a recipe for insanity.  Ask
> Tim, and he'll recommend using small simple regular expressions 9
> times out of 10.  At least thats how he's responded every time I've
> asked about problems with some excessively clever regular expression
> here.  Heck, I think you've even told me the same thing on at least
> one occasion. ;-)

Fredrik is emphasizing size of the string, not complexity of the 
expression. While re is also better at backtracking and the 
hairy stuff, the main correlation is with the size of the string 
being searched - regex is better at short inputs and falls off 
quickly thereafter.

Also, a long pattern is not necessarily a complex pattern. A 
humongous set of alternatives is no problem. The problem is 
when people expect to be able to do more than one thing at a 
time with a pattern.

Such as match an entire Python triple quoted string.

Which Tim does.

bad-Tim-<spank>-<spank>-ly y'rs


- Gordon



From fdrake at acm.org  Mon Jan 10 15:44:48 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 10 Jan 2000 15:44:48 -0500 (EST)
Subject: Python is much faster than Java?
In-Reply-To: <387A43CA.250B52F4@prescod.net>
References: <387A492A.CBB58B86@iqsoft.hu>
	<387A43CA.250B52F4@prescod.net>
Message-ID: <14458.17600.204923.392925@weyr.cnri.reston.va.us>

Paul Prescod writes:
 > Java and Python implementations vary widely so it is probably not useful
 > to say that one is always faster than the other. Presuming you used
 > CPython, the big variable is probably the JVM you used for Java. I would
 > expect some to be much faster than CPython and some to be much slower.

  This and the fact that if the [].sort() method was used from
CPython, and only built-in data types affected the sort computation,
all the code was in C and is very highly optimized.
  Tim Peters recently described some aspects of the [].sort()
implementation if anyone really wants to dig in and understand the
details of that.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From meng.engineering at bluewin.ch  Mon Jan 17 15:44:26 2000
From: meng.engineering at bluewin.ch (Markus Meng)
Date: Mon, 17 Jan 2000 21:44:26 +0100
Subject: [Q] HowTo extract the system time within python
Message-ID: <85vv88$m18$2@bw107zhb.bluewin.ch>

Hi and Hello,

I need to know the System Time on a DOS-Only
system using the python interpreter for a script.
What is needed to get the time/date from DOS
into python?

Any help would be appreciated.

markus

--
********************************************************************
** Meng Engineering        Telefon    056 222 44 10               **
** Markus Meng             Natel      079 230 93 86               **
** Bruggerstr. 21          Telefax    056 222 44 10               **
** CH-5400 Baden           Email      meng.engineering at bluewin.ch **
********************************************************************
** Theory may inform, but Practice convinces. -- George Bain      **








From aahz at netcom.com  Wed Jan 26 16:05:39 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 26 Jan 2000 21:05:39 GMT
Subject: parnassus? When will it be back up?
References: <86nl2t$rr5$1@nnrp1.deja.com>
Message-ID: <86nnj3$p6l$1@nntp6.atl.mindspring.net>

In article <86nl2t$rr5$1 at nnrp1.deja.com>,  <marut at my-deja.com> wrote:
>
>When I visit the parnassus page I see only one word: Conn
>Anyone know its status?

It's working for me (using the link on www.python.org); either it was
down only temporarily or you have a network problem.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From gmcm at hypernet.com  Thu Jan  6 16:58:11 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 6 Jan 2000 16:58:11 -0500
Subject: no return values for __init__ ??
In-Reply-To: <387502E5.F37B5FF5@mdlink.de>
Message-ID: <1264926603-6720782@hypernet.com>

Helge Hess wrote:

> Gordon McMillan wrote:
> > Helge Hess writes:
> > 
> > What happens when a derived class chains to the base class
> > __init__ and the base class __init__ changes "self"?
> 
> Good question. Depends on what one tries to accomplish with
> different return values, usually it would look like:
> 
>   def __init__(self...
>      s = super.__init__(self...
>      if (s is not None) & (s is not self): return s
>      self.a = 5
>      return self

OK, but it introduces a "gotcha". As in, what happened to the 
stuff I set up before calling super.__init__, (which is not only 
valid, but often useful).

> BTW: I also always wondered why Python has None as the default
> return value. 'self' seems better to me as one could make
> compound method calls without additional work while None has no
> benefit I can see (maybe better RC behaviour ?).

It's consistent with function behavior (which has no self to 
return), and after all, a method (when viewed on the class 
object) is a function.
 
> > How about MI?
> 
> This is related to the above.

With yet more "gotchas".  The (one and only) base class 
constructor that could change "self" would have to be the first 
one executed.

> There is no conceptual difference between __init__ returning a
> value and any other method returning a value.  

Not if you don't accept the axiom that __init__ is "just another 
method". In fact, as an initializer, __init__ is a lot less special 
that constructors are in most languages.
 
> > If really you want to customize the class, you can assign to
> > the instance's __class__ attribute.
> 
> Ok, this is getting much nearer to my goal although it is much
> less convenient. Eg this would probably break abstraction since
> the state of the object needs to be converted. Returning a new
> object seems much easier and more consistent to me.

I believe it does more of what you want than what you asked 
for, since it doesn't discard the instance, it simply replaces a 
branch of the attribute lookup tree.

Abstraction is one of those notions that Python tosses out the 
window, yet expresses very well.
 
> > Having "self" be a complete object when __init__ runs  means
> > the language doesn't have to come up with a bunch of arcane
> > rules about default constructors, order in which base class
> > constructors are called etc.
> 
> I don't understand what you are trying to say. We are talking
> about return values. In every invocation of __init__ the object
> is as complete as it can be.

It just may get unexpectedly tossed ;-).
 


- Gordon



From arcege at shore.net  Sat Jan  8 12:39:46 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 08 Jan 2000 17:39:46 GMT
Subject: C Extension Questions (Binary Trees)
References: <84mobo$4d9$1@nnrp1.deja.com>
Message-ID: <CDKd4.2782$Ef6.704898@news.shore.net>

achatham at hotmail.com wrote:
: I have been trying to work on an extension module in C, binary
: (search) trees. I know that you can do most of that stuff with lists or
: dictionaries, and I'm sure that someone has already created a tree data
: type, but I'm doing it mostly for practice (I have trouble coming up
: with my own examples). It has certainly given me a greater appreciation
: of how things work in Python: I now know a whole lot more about
: reference counting, attribute lookup, etc. And I have much greater
: respect for the built-in dictionary type (no matter how contrived my
: test cases are, a dictionary is still faster than my search trees by
: 10-20%).

: But I have some questions that I haven't been able to clear up:

: Question 1. I keep getting warnings from GCC (Redhat 6.1) such as:
: "tree.h: 34: warning: initialization from incompatible pointer type"
: I am not very fond of this warning, though things seem to work fine. The
: relevant parts of tree.h are:

: --------------------
: typedef struct {
:   PyObject_HEAD
:   PyObject *left;
:   PyObject *right;
:   PyObject *parent;
:   PyObject *key;
:   PyObject *data;
:   PyObject *current;
:   PyObject *func;
:   int nodes_in_tree;
: } treeobject;

: staticforward PyTypeObject TreeType;

: static PyObject *tree_insert(treeobject *self, PyObject *args);
: static PyObject *tree_search(treeobject *self, PyObject *args);
: static PyObject *tree_issibling(treeobject *self, PyObject *args);
: static PyObject *tree_successor(treeobject *self, PyObject *args);
: static PyObject *tree_predecessor(treeobject *self, PyObject *args);
: static PyObject *tree_remove(treeobject *self, PyObject *args);
: static PyObject *tree_map(treeobject *self, PyObject *args);
: static PyObject *tree_as_tuple(treeobject *self, PyObject *args);

: static struct PyMethodDef tree_methods[] =
: {
:   {"insert", tree_insert, METH_VARARGS},
:   {"issibling", tree_issibling, METH_VARARGS},
:   {"search", tree_search, METH_VARARGS},
:   {"succ", tree_successor, METH_VARARGS},
:   {"pred", tree_predecessor, METH_VARARGS},
:   {"remove", tree_remove, METH_VARARGS},
:   {"map", tree_map, METH_VARARGS},
:   {"datatuple", tree_as_tuple, METH_VARARGS},
:   {NULL, NULL}
: };
: -------------------

: The warnings are from the assignments in tree_methods. Every one (except
: {NULL, NULL}) generates that warning. How can I get rid of it?

Change the prototypes so the "self" argument is (PyObject *), or cast
the function as "PyCFunction":
  {"map", (PyCFunction),tree_map, METH_VARARGS},

: Question 2: Is there a way to override "map"? I created a nice little
: __getitem__ function for the tree class which works by getting the
: leftmost node in the tree if the index is 0, or the successor to the
: last item if the index > 0, so I can do something like "for x in T:
: print x.data" to print out all the data elements in the tree using an
: inorder traversal. The only problem is that repeatedly calling successor
: is not a very efficient way to do an inorder traversal of a tree. For
: __getitem__ I really don't have a choice about doing it that way, since
: it's hard to store the call stack necessary to do the recursive
: traversal between calls of __getitem__. But for "map", it would make
: more sense to do that, since it's a single call. But from what I can
: tell, map just called __getitem__ on its own until it gets an
: IndexError. This isn't such a big deal, since I just made a member
: function "map", but it'd be nice to override the real one.

The only ways would be to create a method which works like map() and to
use that instead, or to write your own version of map.  The built-in
functions "map", "filter" and "reduce" all access the object using
"__getitem__".

: Question 3: A minor question. How do I get member variables to show up
: when I do "dir(T)"? All I get is a list of methods, when I know that
: T.left is a valid reference (it works fine). Or maybe I'm just too tired
: and "dir()" doesn't show variables normally.

The dir() function access the "__members__" and "__methods__" attributes
of a C extension object.  Each should return a list of strings that are
the names of the attributes.  Below is a code fragment to help.

  static PyObject *
  tree_getattr(PyObject *self, char *attrname)
    {
      ...
      else if (strcmp(attrname, "__members__") == 0)
        return Py_BuildValue("[sssssss]",
           "data", "current", "func", "key", "left", "parent", "right",
        );
      else  /* used for __methods__ too */
        return Py_FindMethod(tree_methods, self, attrname);
      ...
    }

: Again, it's not like I really _need_ this thing working, but I'd like to
: at least get these things straight. And so far the files for this data
: structure have turned out larger than any single Python program I've
: ever written! Yeah, I'm a newbie :)

: Thanks,
: Andrew


From georgethomas at pspl.co.in  Thu Jan 27 02:07:35 2000
From: georgethomas at pspl.co.in (George Thom@s)
Date: Thu, 27 Jan 2000 12:37:35 +0530
Subject: ARP and RARP
Message-ID: <000701bf6895$30014a70$cf02a8c0@intranet.pspl.co.in>

Hi everyone,
I'd like to know if there are any implementations/modules in python that
handle ARP/RARP ?
regards,
george
--------------------------------------------------------------------------
George Thomas
Resident,Planet Earth,Third Rock from the Sun
georgethomas at pspl.co.in


Flat No. FI-5, Fifth Floor,
Himagirinath Co-op. Hsg. Society,
486A Elphinstone Road,
Kirkee, Pune - 411003,
Maharashtra State,
India.
--------------------------------------------------------------------------
Faced with the choice between changing one's mind and proving
there is no need to do so, almost everybody gets busy on the proof.
                                                                 - John
Kenneth Galbraith
--------------------------------------------------------------------------





From coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com  Fri Jan  7 14:27:43 2000
From: coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com (Stephen Coursen)
Date: 7 Jan 2000 19:27:43 GMT
Subject: URL Domain Name
References: <y5rd4.710$hJ1.10236@news.rdc1.tx.home.com>
Message-ID: <slrn87cfb0.7sp.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>

On Fri, 07 Jan 2000 19:26:22 GMT, Paul E. Bible <pbible at home.com> wrote:
>Anyone know how to get the Domain Name for a given IP Address??  (In Python,
>of course)
>
>>> import socket
>>> socket.gethostbyaddr( '127.0.0.1' )
('localhost', ['localhost.localdomain'], ['127.0.0.1'])
>>>

Steve

>Thanks in advance,
>Paul
>
>


From wtanksle at hawking.armored.net  Fri Jan  7 20:46:12 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 8 Jan 2000 01:46:12 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu> <3874FC58.69F1D59D@rubic.com> <slrn87a45v.upr.kc5tja@garnet.armored.net> <slrn87avnr.d7v.wtanksle@hawking.armored.net> <slrn87cvv9.gcu.kc5tja@garnet.armored.net>
Message-ID: <slrn87d689.gi8.wtanksle@hawking.armored.net>

On 8 Jan 2000 00:01:31 GMT, Samuel A. Falvo II wrote:
>>>(Personally, I think the COM solution is a superior one anyway, but that's
>>>just my opinion.)

>>Sure it is -- but compatibility is even better.

>Then why aren't we all using CP/M?

We are -- MSDOS is a clone of CP/M.  We left CP/M behind because it didn't
have the capability of being compatible; it was a backward-looking design.
HTML has the same problems, and XML is trying to fix them.

>There is a point of diminishing returns when it comes to this whole
>"backward compatibility" thing.  Sometimes you just have to pick up your
>bags, and leave Hotel Compatibility behind.  Time marches on, after all.

When designing a web page, you want to let people read it.  I can't
believe that you, of all people, are denying this.

>>Mozilla is right now (M12) almost as stable than Netscape 4.7 or IE 5.0

>For the record, three people in this office, who have used Mozilla before,
>laughed when I mentioned what you said.  In three words, Not Even Close.

I suppose throwing fragments of animated GIFs all over the screen counts
as being stable.  I personally don't like it.  What made me stop (I
switched back to IE 4.0) was two things: first of all, a HUGE display bug
which would grey out my whole screen until reboot; and second, the huge
security holes.

>First, from personal experience at least, IE 5 is 100% more stable than any
>Netscape version I've ever used.

I've seen SO many more crashes and glitches in IE 5 than in Netscape that
this is not even funny.  I can only guess that you're comparing IE on
Windows to Netscape on Linux -- the Linux version of Netscape is REALLY
bad.

>As far as the current versions of Netscape
>are concerned, Mozilla doesn't even hold a candle to them.

I've *seen* it hold a candle to them.  They went up in flames, too!  At
any rate, I wasn't able to make M12 crash.  I didn't keep on using it,
though, because it doesn't have some of the features I use regularly, such
as trackpoint scrolling.

>Which is pretty
>suprising since its built from the same codebase (supposedly) as the much
>more stable NT version.

Um...  Sam?  Mozilla is a complete rewrite.  You were supposed to know
this.  We've talked about it before.

>People have been saying, "Mozilla is almost as
>stable as..." for the last three years. Without fail, every time I try it,
>Mozilla segfaults on something trivial on every Linux box I've used it on.

That may be our problem -- I've always tested it under Windows.

>(I'd be willing to bet that the bugs that destabilize Mozilla under Linux
>have to do with XPCOM, and their apparent disregard for existing COM
>specifications).

Probably not -- I'd bet that there are many other causes for crashes.

>Right now, my bets are on Opera, if and when they ever release it.  And yes,
>I'd be willing to pay for it.

I don't bet, but I like Opera too.  A graphical browser which lets me use
keyboard-only access -- how cool can you get?  Everybody try out Opera.

It'll be really nice to have a halfway decent graphical browser for Linux
again.

>Samuel A. Falvo II

-- 
-William "Billy" Tanksley, in hoc signo hack


From bjorn at roguewave.com  Wed Jan 19 13:17:14 2000
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Wed, 19 Jan 2000 11:17:14 -0700
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com> <slrn88bsde.4pp.grant@grante.comtrol.com
Message-ID: <3885FFA9.CF0F3BE5@roguewave.com>

> P.S. Don't tell anybody about this.  It would be just fine if
>      everybody continued to think I knew nothing about win32
>      programming.  ;)

We'll elect you official Python MAPI guru whether you want it or not <wink>

-bjorn




From kuncej at mail.conservation.state.mo.us  Wed Jan  5 10:54:55 2000
From: kuncej at mail.conservation.state.mo.us (Jeff Kunce)
Date: Wed, 5 Jan 2000 09:54:55 -0600
Subject: http POSTing
References: <8EB19A9E7newshoesmailcom@205.147.63.94>
Message-ID: <OOJc4.245$B5.4746@news.more.net>

> i'm trying to create an offline form i can send to a website
> through "POST". being new to python and minimal knowledge of

For a simple example, see
    urlpost.py
at
    http://starship.python.net/crew/jjkunce/

  --Jeff




From tanming_fong at phoenix.com  Tue Jan  4 21:00:55 2000
From: tanming_fong at phoenix.com (TM)
Date: 4 Jan 2000 20:00:55 -0600
Subject: Error in inserting blob into ORACLE using ODBC.Windows
References: <386c0577.107657473@news.phoenix.com> <386CD167.8525C821@roguewave.com> <84ijb0$u46$1@nntp8.atl.mindspring.net> <38722065.507767100@news.phoenix.com>
Message-ID: <3872a20d.7967757@news.phoenix.com>

It worked! So it was a configuration problem.  I fetched a newer ODBC
driver from Oracle web site and got both the python and ADO code to work,
BLOB in and BLOB out.  The newer driver has a "Enable LOBs" checkbox in the
ODBC Manager driver configuration window, which was probably the key to the
solution.  There's also a "Force Retrieval of Long Columns" checkbox in the
"WorkAround Options" , which i don't know what kind of work around it is
for.

Tanming Fong



From kc5tja at armored.net  Wed Jan 19 23:12:50 2000
From: kc5tja at armored.net (Samuel A. Falvo II)
Date: Wed, 19 Jan 2000 20:12:50 -0800
Subject: How to copy a binary file?
References: <3886876F.20FFEA27@erols.com>
Message-ID: <388685dc@fermi.armored.net>

> def file_copy(source, dest):
>     in_file = open(source, 'r')
>     out_file = open(dest, 'w')

Try:

  in_file = open( source, 'rb' )
  out_file = open( dest, 'wb' )

This way, you're always opening the file in _binary_ mode instead of the
default, text mode.  I suspect that your reads are stopping at end-of-line
characters.





From gerrit.holl at pobox.com  Mon Jan 17 14:35:59 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 17 Jan 2000 20:35:59 +0100
Subject: putenv() and os.environ
Message-ID: <20000117203559.A1246@stopcontact.palga.uucp>

Hello,

>>> os.putenv('aaa', 'bbb')
>>> os.environ.get('aaa')

I can't get any environment variables not set with the os.environ
dict. Shouldn't os.environ be rewritten in C then, so it can get
the REAL environment variable?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From scherbi at bam.com  Fri Jan 14 11:33:33 2000
From: scherbi at bam.com (Bill Scherer)
Date: Fri, 14 Jan 2000 11:33:33 -0500
Subject: __name__
References: <387F4F42.B9E1DB8E@kw.igs.net>
Message-ID: <387F4FDD.89305BB@bam.com>

I think you want:

    return self.__class__.__name__

-Bill

JB wrote:

> Could anyone tell me how to get a classes name:
>
> class MyClass:
>    def className( self ):
>        return self.__name__
>
> This doesn't fly. The docs say a class has the __name__ attribute but...
>
> thanks
> jb
>
> --
> There are bugs and then there are bugs.  And then there are bugs.
> -- Karl Lehenbauer
> --
> http://www.python.org/mailman/listinfo/python-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000114/315b87d1/attachment.vcf>

From tismer at tismer.com  Mon Jan 24 00:43:42 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 24 Jan 2000 00:43:42 -0500
Subject: Stackless 1.01 Homepage
Message-ID: <388BE68E.B26985CA@tismer.com>

"Why should I care about this, isn't my Python fine"?

If you know what Generators, Coroutines and Continuations
are about and want to have them at your fingertips,
this is the real thing for you.

If not, how about speed?
About 5 percent faster, binary compatible as a drop-in
replacement for your python15.dll, unlimited recursion,
why not give it a try? Just click on the installer :-)

Stackless Python 1.01 with Continuations 0.6
can now be downloaded from a web page:

http://www.tismer.com/research/stackless/

Fact sheet, links to documentation, source and
binaries can be found there.

Source and binary distribution have been split.
The Windows binaries come with a Wise installer.

SLP has been successfully compiled on Solaris (Thanks to Jeff
Collins). It will be used on PalmIII, to solve the 4k stack problem.

Version 1.01 is a small clean-up for non-Windows platforms.

<P><A HREF="http://www.tismer.com/research/stackless/">Stackless
Python 1.01 + Continuations 0.6</A> - a version of Python 1.5.2 that
does not need space on the C stack, and first-class callable
continuation objects for Python.  (20-Jan-2000)

Christian Tismer
Mission Impossible 5oftware Team



From aahz at netcom.com  Fri Jan 21 10:47:09 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 15:47:09 GMT
Subject: DBI Commentary
References: <1263695833-12432931@hypernet.com>
Message-ID: <869v1t$a2m$1@nntp9.atl.mindspring.net>

In article <1263695833-12432931 at hypernet.com>,
Gordon McMillan <gmcm at hypernet.com> wrote:
>Timothy Grant writes:
>>
>> Aahz made the very valid point that specifying columns in your query
>> *guarantees* tuple return order. I had not fully appreciated that, and I
>> have put that into use in a number of places. However, I have to agree
>> with Lance when he says that "SELECT * FROM table" is not as often a bad
>> thing as Aahz claimed. 
>
>That was me, not Aahz. 

Actually, it was me, too.  Timothy made the boo-boo of cc'ing me in
e-mail with no indicator, so he probably saw my response first.  He
later responded to that e-mail publicly.

>I've used the trick myself - but it was for building HTML tables 
>to display the results from an ad-hoc query built from an HTML 
>form. I personally would never use "*" if I new what columns I 
>needed.

Yeah; as I told Timothy, if managing the fields is a problem, create a
list and use string.join() to build the SQL string.  It's easy enough to
extend that to dict-based access, which you can use to build your own
SQL wrapper library/class.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From gerrit.holl at pobox.com  Thu Jan 13 09:52:45 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 13 Jan 2000 15:52:45 +0100
Subject: Wanneer is nu?
In-Reply-To: <200001112348.AAA00993@esperanto.hetnet.nl>; from Johan.Swenker@hetnet.NL on Wed, Jan 12, 2000 at 12:48:59AM +0100
References: <20000111104436.A4538@dnd.utwente.nl> <200001112348.AAA00993@esperanto.hetnet.nl>
Message-ID: <20000113155245.D1928@stopcontact.palga.uucp>

Johan.Swenker at hetnet.NL wrote on 947634539:
> werkdag: maandag t/m vrijdag

Heet dat in Onze Taal niet een heel klein beetje een "werkweek"?
                                                          ~~~~

groeten,
Gerrit.

-- 
Please correct any bad English you encounter in my mail!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From tim_one at email.msn.com  Fri Jan 21 04:38:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 21 Jan 2000 04:38:17 -0500
Subject: In defence of the two-namespace rule
In-Reply-To: <m3n1q0fzg7.fsf@greebo.nodomain.de>
Message-ID: <001701bf63f3$3f1c8660$ec2d153f@tim>

[Bernhard Herzog]
> ...
> However, several people have proposed lexical scoping
> extensions and some of these proposals have even been
> implemented. Lexical scoping is somewaht controversial
> and I don't know what Guido's opinion is, although if the
> technical reasons (cyclic references) Tim mentioned in
> his post were the only reason Guido dropped it, it seems
> likely that lexical scoping might be implemented if the
> technical problems can be solved.

I have a hard time distinguishing Guido's opinion from mine, but *some* bad
days he does too so it's OK <wink>.

FWIW, as a long-time Lisp and functional language fan, I harassed Guido
about the lack of lexical scoping before Python 1.0 was released.  He simply
wanted to *try* a simpler name resolution scheme, and prevailed upon me to
just give it a chance; in part, I believe he was unhappy with the extreme
abuse of deep nesting common in Pascal programs of that time (I know I was
<0.4 wink>), which often made large Pascal programs impossible to
understand.

So I tried it his stupid way, and damned if I didn't come to like it a whole
lot -- the combo of flexible modules and flexible classes flatly <wink>
proved much more satisfying to me as soon as I got the hang of it.

So for several years Guido & I took turns arguing against "deep" lexical
scoping on the Python mailing list, and, later, c.l.py.  But not often!  It
just didn't come up much.

And then "lambda" was introduced.  The combo of lambda and "two-level"
scoping basically sucks, and there's been a scoping roast ever since.  It's
never going to go away by itself, so one of the two has to bend.

I still like Python's scoping rules, and am now so used to default-arg abuse
(to leak names across the scope boundary) that I'm barely conscious of its
strangeness anymore.  It clearly isn't a real problem for Guido either.  But
it was and remains unboundedly surprising to newcomers, and that's
un-Pythonic too.

Guido has since said publically that he may have overreacted to "Pascal
abuse", but that's the closest he's come (that I've heard, anyway) to saying
he now likes traditional lexical scoping.  On the basis of "if the
implementation is easy to explain, it may be a good idea", he tried
implementing it.  As several people have been surprised to discover, lexical
closures are laughably easy to add to Python -- frames are already heap
allocated, and it's mostly just a matter of adding a "static link" for name
resolution.

So that's my guess on Guido's current opinion:  while he still doesn't
*like* closures, he's no longer deeply opposed to adding them; sees that the
current combo of features is harmfully confusing to newcomers; and has to
acknowledge that the simplicity of the required changes argues for an
inherent naturalness.

But, as I said, I have a hard time distinguishing Guido's opinion from mine
<wink>.

What I worry about (unsure about Guido on this one) is that adding closures
will also intensify clamoring for more Lispish features.  While I love them
in their native languages, they grate against Python's approach, and every
such addition to date has (in my arrogant opinion) done more harm to Python
than good.

List comprehensions are an exception:  they'll work out great.  To know the
reason why, just ask a Scheme fanatic why Haskell is a terrible language
<wink>.

it's-not-for-lack-of-lambdas-ly y'rs  - tim





From darrell at dorb.com  Thu Jan  6 11:58:46 2000
From: darrell at dorb.com (Darrell)
Date: Thu, 6 Jan 2000 11:58:46 -0500
Subject: how to include a file -- not import??
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com> <3874F5DA.B73A6039@starmedia.net>
Message-ID: <004801bf5867$4c4a3270$0100a8c0@rochester.rr.com>

From: "Andrew Csillag" <andrew at starmedia.net>
> Darrell wrote:
> > 
> > from filename import *
> 
> Actually, that doesn't do it either if the other module wants to
> reference vars in the importing namespace.
> 
> The way to make it so that the other code can access variables in the
> including files namespace is:
> exec open('filename').read() 
> 

Got me on that one.
It never occured to me to do something like this :-/
I'm thinking this isn't a good thing. 

--Darrell





From michelorengo at netscape.com  Wed Jan 19 15:52:30 2000
From: michelorengo at netscape.com (Michel Orengo)
Date: Wed, 19 Jan 2000 20:52:30 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com> <slrn88c42t.53m.grant@grante.comtrol.com>
Message-ID: <388624F4.8B2EB257@netscape.com>


Grant Edwards wrote:

> OK, I did that with msoutlk85.olb.  I'm assuming that a message
> object is of type _MailItem.  I've got a big list of properties
> and methods that _might_ work, but how do you find out which
> ones are implimented by a particular object?
>
> For example, the messages in my inbox implement properties
> "Subject" and "Size", but not "Body", "To", or "SenderName".
> I guess I could iterate the the properties dictionary and try
> all of them...

Well, here is an extract of the file produced by makepy (under gen_py):
- BTW, I selected the "OLE/Messaging 1.0 Object Library" not sure what is the
msoutlk85.olb.-

class Message(DispatchBaseClass):
 """OLE/Messaging Message"""
 CLSID = pythoncom.MakeIID('{3FA7DEAA-6438-101B-ACC1-00AA00423326}')

 def Delete(self):
  """ """
  return self._ApplyTypes_(0x6a, 1, (12, 0), (), 'Delete', None)

 def Options(self, parentWindow=defaultNamedOptArg):
  """ """
  return self._ApplyTypes_(0x7d, 1, (12, 0), ((12, 17),), 'Options', None,
parentWindow)

 def Send(self, saveCopy=defaultNamedOptArg, showDialog=defaultNamedOptArg,
parentWindow=defaultNamedOptArg):
  """ """
  return self._ApplyTypes_(0x83, 1, (12, 0), ((12, 17), (12, 17), (12, 17)),
'Send', None, saveCopy, showDialog, parentWindow)

 def Update(self, makePermanent=defaultNamedOptArg,
refreshObject=defaultNamedOptArg):
  """ """
  return self._ApplyTypes_(0x84, 1, (12, 0), ((12, 17), (12, 17)), 'Update',
None, makePermanent, refreshObject)

 _prop_map_get_ = {
  "Application": (20, 2, (12, 0), (), "Application", None),
  "Attachments": (5, 2, (12, 0), (), "Attachments", None),
  "Class": (6, 2, (12, 0), (), "Class", None),
  "Conversation": (8, 2, (12, 0), (), "Conversation", None),
  "ConversationIndex": (55, 2, (12, 0), (), "ConversationIndex", None),
  "ConversationTopic": (54, 2, (12, 0), (), "ConversationTopic", None),
  "DeliveryReceipt": (50, 2, (12, 0), (), "DeliveryReceipt", None),
  "Encrypted": (10, 2, (12, 0), (), "Encrypted", None),
  "Fields": (12, 2, (12, 0), (), "Fields", None),
  "ID": (14, 2, (12, 0), (), "ID", None),
  "MAPIOBJECT": (49, 2, (12, 0), (), "MAPIOBJECT", None),
  "Parent": (25, 2, (12, 0), (), "Parent", None),
  "ReadReceipt": (30, 2, (12, 0), (), "ReadReceipt", None),
  "Recipients": (28, 2, (12, 0), (), "Recipients", None),
  "Sender": (32, 2, (12, 0), (), "Sender", None),
  "Sent": (33, 2, (12, 0), (), "Sent", None),
  "Session": (34, 2, (12, 0), (), "Session", None),
  "Signed": (35, 2, (12, 0), (), "Signed", None),
  "Size": (36, 2, (12, 0), (), "Size", None),
  "Submitted": (18, 2, (12, 0), (), "Submitted", None),
  "TimeReceived": (40, 2, (12, 0), (), "TimeReceived", None),
  "TimeSent": (41, 2, (12, 0), (), "TimeSent", None),
  "Unread": (42, 2, (12, 0), (), "Unread", None),
  "folderID": (51, 2, (12, 0), (), "folderID", None),
  "importance": (27, 2, (12, 0), (), "importance", None),
  "storeID": (52, 2, (12, 0), (), "storeID", None),
  "subject": (38, 2, (12, 0), (), "subject", None),
  "text": (39, 2, (12, 0), (), "text", None),
  "type": (17, 2, (12, 0), (), "type", None),
 }
 _prop_map_put_ = {
  "Application" : ((20, LCID, 4, 0),()),
  "Attachments" : ((5, LCID, 4, 0),()),
  "Class" : ((6, LCID, 4, 0),()),
  "Conversation" : ((8, LCID, 4, 0),()),
  "ConversationIndex" : ((55, LCID, 4, 0),()),
  "ConversationTopic" : ((54, LCID, 4, 0),()),
  "DeliveryReceipt" : ((50, LCID, 4, 0),()),
  "Encrypted" : ((10, LCID, 4, 0),()),
  "Fields" : ((12, LCID, 4, 0),()),
  "ID" : ((14, LCID, 4, 0),()),
  "MAPIOBJECT" : ((49, LCID, 4, 0),()),
  "Parent" : ((25, LCID, 4, 0),()),
  "ReadReceipt" : ((30, LCID, 4, 0),()),
  "Recipients" : ((28, LCID, 4, 0),()),
  "Sender" : ((32, LCID, 4, 0),()),
  "Sent" : ((33, LCID, 4, 0),()),
  "Session" : ((34, LCID, 4, 0),()),
  "Signed" : ((35, LCID, 4, 0),()),
  "Size" : ((36, LCID, 4, 0),()),
  "Submitted" : ((18, LCID, 4, 0),()),
  "TimeReceived" : ((40, LCID, 4, 0),()),
  "TimeSent" : ((41, LCID, 4, 0),()),
  "Unread" : ((42, LCID, 4, 0),()),
  "folderID" : ((51, LCID, 4, 0),()),
  "importance" : ((27, LCID, 4, 0),()),
  "storeID" : ((52, LCID, 4, 0),()),
  "subject" : ((38, LCID, 4, 0),()),
  "text" : ((39, LCID, 4, 0),()),
  "type" : ((17, LCID, 4, 0),()),
 }

This give you the methods (Send, Delete, Update...) + attributes (Application,
Attachements,...) [and actually some you can get and/or put]
so you in my previous example I used  "msg.subject", you can also get msg.text
to get the body of the message.
What is less obvious is thing such as Recipients, because Recipients is a
collection of objects (ie Recipient), but you can find the description of the
class Recipient in the same .py generated by makepy.

Or again look at CDO (under "Platform SDK, Database and Messaging Services") in
MSDN Library.




From quinn at amatho.ugcs.caltech.edu  Thu Jan 20 17:01:30 2000
From: quinn at amatho.ugcs.caltech.edu (Quinn Dunkan)
Date: 20 Jan 2000 22:01:30 GMT
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu> <FonA21.B26@world.std.com>
Message-ID: <slrn88f1gg.rq9.quinn@amatho.ugcs.caltech.edu>

On Thu, 20 Jan 2000 17:32:24 GMT, Will Ware <wware at world.std.com> wrote:
>Paul M (paul.magwene at yale.edu) wrote:
>: Dear Pythoneers...
>
>A great evil has been perpetuated in the design of the Python
>language. It grieves me to be the one to bring this to the world's
>attention but it must not continue to go unacknowledged. Persons of
>right indentation have once again been marginalized and excluded
>because _every_editor_ever_written_ indents from the LEFT SIDE OF THE
>PAGE.

In vim:
:set rl

I'm sure any editor which supports hebrew, arabic, and like will have no
problem, provided you've compiled it with support.  I admit, however, this
still doesn't fix python (mis) behaviour.  While one could use your script as
an automatic preprocessor, it would be much more user-friendly to just make
right-to-left the python default.

And it should print Content-type before every script, to make it suitable for
CGI.


From pierre at saiph.com  Tue Jan 25 19:10:23 2000
From: pierre at saiph.com (Pierre Imbaud)
Date: Wed, 26 Jan 2000 01:10:23 +0100
Subject: Need Preprocessor Module!
References: <388E124C.B6B9BF9A@pk.highway.ne.jp>
Message-ID: <388E3B6F.C035D5A@saiph.com>

Akinori Tsuchida wrote:
> 
> I would like to use preprocessor-like functionality in Python
> like we use in C/C++.
> 
> Many corresponding functionalities are available in Python, but
>  I think preprocessor is still useful.
> 
> For example, you can comment multiple lines just by "if 0" and "endif"
> with no need to consider indenting.
> You don't have to take care of not embracing another comment section, I
> mean...
> 
> #if 0
> #if 0
>    not required part.
> #endif
> #endif
> 
> Above is treated as expected but
> 
>   """
>   """
>   not required part.
>   """
>   """
> not the case with above. (unwanted part still remains.)
> 
> Well, Any module available ?
> like,... input strings containing preprocessor lines
> and then output processed strings.
> 
> #ifdef YOU_KNOW
> #include INFO
> Please_Answer(INFO)
> #endif
> 
> Thanks for reading this article.
> 
> Akinori Tsuchida.
Well, m4, available on any unix platform, can do what you ask for, and
more.
But you should think twice: adding a layer always mean adding trouble,
source of bugs, loss of readability...
somehow, think of python as of risc processors: give up some fancy
possibilities,
keep a basic set that lets U do everything, and U will be rewarded for
choosing
simplicity
-- 
Pierre Imbaud <pierre at saiph.com>
12 Rue des Bosquets 91480 Quincy Sous S?nart France
Tel:  01 69 00 94 57 Fax 79 65


From nickb at earth.ox.ac.uk  Fri Jan 28 07:09:55 2000
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Fri, 28 Jan 2000 12:09:55 +0000
Subject: Charts
References: <3890C74D.786411BC@acm.org> <87n1pr6m0i.fsf@iname.com> <389118DB.4B7997AE@wjk.mv.com>
Message-ID: <38918712.F53002D2@earth.ox.ac.uk>

"Wm. King" wrote:

> Other would be to use Tkinter and create a canvas ( have to do math as well)

If you are intending going to the level of Tkinter Canvases then you might want to look at BLT.

This is attached to the PMW module stuff and supports some rather nifty plotting  with

Graphs, Stripcharts, Histograms etc

cheers

Nick

--
Dr Nick Belshaw                                    Dept of Earth Sciences
Tel : 01865 272000                                 Oxford University
Fax : 01865 272072                                 Parks Rd, Oxford, OX2 3PR





From phd at phd.russ.ru  Thu Jan 13 10:17:35 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Thu, 13 Jan 2000 15:17:35 +0000 (GMT)
Subject: Cookies
In-Reply-To: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com>
Message-ID: <Pine.LNX.4.21.0001131516440.18609-100000@fep132.fep.ru>

> import os, Cookie
> 
> kaka = Cookie.Cookie(os.environ["HTTP_COOKIE"])
> try:
> 	nPos = kaka['nPos'].value
> except:
> 	nPos = 0

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From sjohnson at csnet.net  Wed Jan 26 21:06:33 2000
From: sjohnson at csnet.net (Scott Johnson)
Date: Thu, 27 Jan 2000 02:06:33 GMT
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
References: <816010E2456BD111A48700805FBBE2EE01C60408@ex-quebec-u1.baan.com>
Message-ID: <JKNj4.855$yY6.12970@news.goodnet.com>

I'm saying you have a good idea.

----------
Gaetan Corneau <Gaetan_Corneau at baan.com> wrote...

> I'm really tempted to hack this guy's machine. 
> Destroy everything. 
> Or at least bust his mailbox.
> 
> The kind of spammer I hate the most.
> Sorry, I know this is off-topic.
> ______________________________________________________
>    Gaetan Corneau
>    Software Developer 
>    Software Engineering Process Group
>    BaaN  Supply Chain Solutions  
>    http://www.baan.com
>    E-mail: Gaetan_Corneau at baan.com                  
>    Tel: (418) 266-8252         
> ______________________________________________________
> "Profanity is the one language all programmers know best"
> 
> 
> -----BEGIN GEEK CODE BLOCK-----
> Version: 3.1
> GCS/B/MU d- s+:++ a+ C++ 
> UL+ P--- W+ N- K- W++ t-- 
> !5 X- R+ tv-- b++ DI++ 
> G e++ h---- r+++ y++++
> ------END GEEK CODE BLOCK------
> 
> 
> > -----Original Message-----
> > From: 
> > 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > [mailto:24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.
> > pacbell.ne
> > t]
> > Sent: Sunday, January 23, 2000 3:45 PM
> > To: python-list at python.org
> > Subject: DARREN MULLIGAN SEEKING MODELS! 89582
> > 
> > 
> > My name is Darren Mulligan and I am an aspiring photographer 
> > from California.
> > I will get right to the meat and gravy of this post, so as 
> > not to waste your time!
> > 
> > +++++WHAT AM I SEEKING?+++++
> > 
> > I am looking for young models (prefer early teen 14-16 year 
> > old female) for nude and semi-nude photography. This is a 
> > great way for me to update my portfolio, and also a great way 
> > for you to get a free portfolio, too! This could lead to 
> > employment in the Adult XXX entertainment business. What a 
> > great way to make money!
> > 
> > Please respond to one of the following e-mail addresses. I 
> > run my very own mail server on an ADSL connection so you can 
> > be assured to complete discretion and privacy. If one e-mail 
> > address does not work, simply try the next! 
> > 
> > Sorry for the long list! My ADSL connection goes down from 
> > time to time, so I posted all the e-mail addresses I can be 
> > reached at. If one does not work, just move on to the next!
> > 
> > nanae at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > usenet at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > news.admin.net-abuse.email at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > spamlinkpage at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > alt.make.money.fast at nntp.8m.com  nanas at nntp.8m.com   
> >   hostmaster at FREESERVERS.COM
> > 
> > THIS IS NOT A SCAM! IT IS NOT CHILD PORN!!
> > There is nothing wrong or illegal about the artistic 
> > representation of a young body. If you have a problem with 
> > this post, please e-mail me and I will not post to your 
> > newsgroup again!
> > 
> > Sincerely,
> > Darren Mulligan
> > 
> > -- 
> > http://www.python.org/mailman/listinfo/python-list
> > 
> 



From rob at hooft.net  Mon Jan 31 10:52:16 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: Mon, 31 Jan 2000 16:52:16 +0100 (CET)
Subject: IOError: (0,"Error") when closing "os.popen"ed file?
In-Reply-To: <000c01bf6c00$ecced140$3acbd9c2@peridot.optichrome.com>
References: <14485.41767.332703.614501@temoleh.chem.uu.nl>
	<000c01bf6c00$ecced140$3acbd9c2@peridot.optichrome.com>
Message-ID: <14485.44976.913646.228004@temoleh.chem.uu.nl>

>>>>> "AE" == Adrian Eyre <a.eyre at optichrome.com> writes:

 >> But maybe I should repeat the flush() until it succeeds, and only
 >> then close() the file. Since I wrote my first message I have
 >> already found occasions where even the sleep(0.1) doesn't solve
 >> the problem. Bad. I really hope I can properly fix this.

 AE> TBH it sounds more like a bug in one of the following:

 AE> Python, Standard C libraries, OS

 AE> I suspect one of the latter two.

 AE> Are you not able to trace where/why this odd error occurs?

I could try to upgrade this machine (it appears to be a RH6.0 and not
6.1 as I thought earlier). But if that doesn't fix it, how would I go
about tracing the problem? Do I have to edit "printf's" into the
python source? I'm running quite a large package (controlling a data
acquisition robot, 45.1k lines of Python) to produce this problem, and
the occurrence seems to be timing-dependent.  Horrible debugging
scenarios doom in front of me....

Rob

-- 
=====   rob at hooft.net          http://www.xs4all.nl/~hooft/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========



From arcege at shore.net  Tue Jan 18 20:20:27 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 19 Jan 2000 01:20:27 GMT
Subject: simple question about raw_input
References: <1Zyg4.1372$Pb4.11218@newsfeed.slurp.net>
Message-ID: <vj8h4.3862$Ef6.1020706@news.shore.net>

TechSpood <ScrewMe at mypad.com> wrote:
: Ok I am about as new a you can get to programming and I have a question
: which should be very simple.

: I am trying to make a simple program which ask the user for his/her name
: then says
: "hello (name) what would you like to do something"
: after it displays this it is supposed to take the users input and store it
: in a variable... all in 2 lines (which I think is possible)

: I tried this but it doesn't quite work right...can you tell me what I am
: doing wrong?

: name=raw_input ("What is your name?")
: choice1=raw_input "hello, %s, would you like to do something?" % name

The raw_input function needs the parentheses aroudn the rest of the
line:
  choice1= raw_input ("hello, %s, would you like to do something?" % name)

The parentheses are necessary so the language to do things like:
  
  >>> round(10.0 / 4.0) * 5.0
  15.0
  >>> round(10.0 / 4.0 * 5.0)
  13.0
  >>>

Where the parentheses start and end tells what to pass to the round()
function.

  -Arcege



From BgPorter at NOacmSPAM.org  Mon Jan 31 10:54:27 2000
From: BgPorter at NOacmSPAM.org (Brett g Porter)
Date: Mon, 31 Jan 2000 15:54:27 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <874sby5m6l.fsf@iname.com> <20000131153329.B19725@xs4all.nl>
Message-ID: <Teil4.12357$t_.270545@news.rdc1.nj.home.com>

"Thomas Wouters" <thomas at xs4all.net> wrote in message
news:20000131153329.B19725 at xs4all.nl...
> """
> Tabs are 8 characters, and thus indentations are also 8 characters.
> There are heretic movements that try to make indentations 4 (or even 2!)
> characters deep, and that is akin to trying to define the value of PI to
> be 3.
>
> Rationale: The whole idea behind indentation is to clearly define where
> a block of control starts and ends.  Especially when you've been looking
> at your screen for 20 straight hours, you'll find it a lot easier to see
> how the indentation works if you have large indentations.
>
...except that studies of actual human programmers say otherwise. From Steve
McConnell's "Code Complete" (pp 409-410):
"The same study ['Program Indention and Comprehensibility', Miaria, et al,
1983] found...[t]he lowest comprehension scores were achieved on programs
that were not indented at all. The second lowest were achieved on programs
that used six-space indentation. The study concluded that two to four space
indentation was optimal. Interestingly, many subjects in the experiment felt
that the six-space indentation was easier to use than the smaller
indentations, even though their scores were lower."

The original study was printed in the Nov 1983 Communications of the ACM.





From gerrit.holl at pobox.com  Sun Jan 16 13:59:19 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 16 Jan 2000 19:59:19 +0100
Subject: Breaking out of walk()
In-Reply-To: <3881f20c.844046140@news1.on.sympatico.ca>; from rjroy@takingcontrol.com on Sun, Jan 16, 2000 at 04:36:10PM +0000
References: <38811854.F2BC197E@rl.gov> <3881f20c.844046140@news1.on.sympatico.ca>
Message-ID: <20000116195919.A7714@stopcontact.palga.uucp>

Robert Roy wrote on 948036970:
> if 'cvs' in names: names.remove('cvs')
> 
> will skip over directories named 'cvs'

It won't:
>>> l = ['foo', 'cvs', 'bar', 'cvs']
>>> l.remove('cvs')
>>> l
['foo', 'bar', 'cvs']

use:
>>> while 'cvs' in l:
...     l.remove('cvs')
...
>>> l
['foo', 'bar']

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From backov at nospam.csolve.net  Fri Jan  7 13:12:53 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Fri, 07 Jan 2000 18:12:53 GMT
Subject: Getting SWIG to work with C++/Python
References: <9mad4.42353$Ke.27158@news21.bellglobal.com> <yPmd4.37$ed.1161540647@news.euroconnect.net>
Message-ID: <F0qd4.43681$Ke.27628@news21.bellglobal.com>

Brian Dam Pedersen <brian.pedersen at mail.danbbs.dk> wrote in message
news:yPmd4.37$ed.1161540647 at news.euroconnect.net...
> SWIG is very picky with its C++ syntax. For example it chokes on the
> following code from VC++
>
> class NOFFER_API CNoffer {
> public:
>  CNoffer(void);
>  // TODO: add your methods here.
> };
>
> The NOFFER_API part bothers it. You have to make a .i file containing a
> cleaned-up version of your classes or functions. This will also give you
the

Yep, on more careful analysis of the error messages, I noticed that the
lines it were choking on were the function declarations, which often have
stuff like __fastcall, which is just a BCB way to tell the compiler what
calling conventions to use. I guess it's going to be a lot more work than I
thought to use swig. ;>

Cheers,
Jason





From soundhack at hotmail.com  Mon Jan 17 17:52:04 2000
From: soundhack at hotmail.com (Robert)
Date: Mon, 17 Jan 2000 17:52:04 -0500
Subject: PythonWin and Dialogs in DLLs
Message-ID: <38839d87$0$22473@senator-bedfellow.mit.edu>

Hi all,
    After getting bored creating dialogs by making dialog templates
programmatically,
I would instead like to use VC++'s dialog editor to format the dialog.
Looking at the dialog.Dialog class in PythonWin, apparently
it accepts a DLL filename to get resource information.

   My question is this: How can I make this DLL that contains
only the dialog resource? Using the appwizards to make a DLL
project, I either get a big DLL file (which works, but it has DLLMain() and
run time libraries linked in that i dont really need) or an error linking,
saying there is an unresolved external
reference _DllMainCRTStartup. The ideal situation would be just
to save the resource file and have pythonwin read that, but if i
have to use DLLs, I would like to keep unnecessary runtime stuff out of it.

Thanks in advance,
Robert




From wware at world.std.com  Wed Jan 12 22:31:03 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 13 Jan 2000 03:31:03 GMT
Subject: Mem leak fix in uthreads
Message-ID: <Fo98Fx.39x@world.std.com>

Jeff Collins identified and fixed some memory leaks in my
microthreads package. The patched version is available at
ftp://ftp.std.com/pub/wware/uthreads.tgz.

I am hoping before the conference to make some progress on
merging microthreads with Christian Tismer's Stackless Python.
Things are a bit busy but I'll see what I can do.
-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From tim_one at email.msn.com  Mon Jan 24 19:45:45 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jan 2000 19:45:45 -0500
Subject: Multidimensional Dictionaries: Possible?
In-Reply-To: <slrn88ppsg.49t.w6uv@voyager.home.com>
Message-ID: <001101bf66cd$84128bc0$a42d153f@tim>

[Jerry Gardner]
> Are multidimensional dictionaries possible in Python?

Certainly.

> I'm sure they're possible, but I can't figure out the syntax.
>
> Here's how I do it in Perl:
>
> %dict = ();  # create empty dictionary
>
>
> $dict{$key1}{$key2} = $string;       # make an entry
> $newstring = ${dict{$key1}{$key2}};  # access an entry

Python doesn't have autovivification (unless you subclass UserDict to supply
it yourself).  You have to create intermediate data structures explicitly:

dict  = {}
dict[key1] = {}
dict[key1][key2] = string
newstring = dict[key1][key2]

Python rarely assumes you know what you're doing <wink -- but while that
sometimes adds more typing, it often saves your butt>.

Note too that you can use dict[key1, key2, ...] to "fake" a multidimensional
dict.  Sometimes that's more appropriate.  The equivalent in Perl
interpolates the multiple subscripts into a string; Python's dicts are more
general (can be indexed by any immutable object, not just strings), and
Python constructs a *tuple* of keys under the covers.  That's faster and
can't suffer accidental "superkey" collisions.





From tye4 at yahoo.com  Thu Jan 20 15:23:29 2000
From: tye4 at yahoo.com (tye4)
Date: Thu, 20 Jan 2000 12:23:29 -0800
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu>
Message-ID: <867qt1$a49@news.or.intel.com>

I have Python installed on Linux Redhat 6.0. Does anyone know where the
source code for the compiler is located (any .tar file?). I intend to modify
the language's syntax to add
'end' delimiters for the blocks.
And since Python is copyrighted, I don't intend to flood out versions to
anyone else... strictly for personal use.

class TestNew

    def foo(s):
        for i in range(0, 10):
            print i,
        end loop

        if i == 10:
            print 'internal error'
        end if

        while i < 10:
            print i
        end loop
    end def

end class

My end delimiters are inspired by Ada, which in my personal opinion has best
overall syntax.

Thx,

-tye4




From aahz at netcom.com  Fri Jan 21 10:16:36 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 15:16:36 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com>
Message-ID: <869t8k$sb9$1@nntp3.atl.mindspring.net>

In article <388521FB.BCC45087 at callware.com>,
Ivan Van Laningham  <ivanlan at callware.com> wrote:
>
>Years ago, Susan Sontag discovered Science Fiction.  Being a big name
>with a large reputation, she thought it incumbent upon her to call
>attention to the small, overlooked ghetto of SF.  That would have been
>fine, but when she wrote articles about SF, she failed to do even the
>minimal research needed to find out what fans of SF called it.  So she
>made up a name, and referred to it as "Sci-Fi" in her essays.

Like Brett Porter, I thought that Forrey Ackerman laid claim to the
coinage; got a source?
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From vetler at ifi.uio.no  Sun Jan  2 06:53:24 2000
From: vetler at ifi.uio.no (Vetle Roeim)
Date: 02 Jan 2000 12:53:24 +0100
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84mh65$2uv$1@harvee.billerica.ma.us>
Message-ID: <yayvh5cg05n.fsf@ringhorni.ifi.uio.no>

* Eric S. Johansson
> The XML based code suggestion was also fairly interesting.

Indeed. But the first thought that struck me, was that if such a
language was created, it would most certainly be hyped into 
oblivion ;)

XML is a major buzzword, and too much hype is a bad thing.
 


vr


From rgruet at ina.fr  Wed Jan 26 10:19:08 2000
From: rgruet at ina.fr (Richard Gruet)
Date: Wed, 26 Jan 2000 16:19:08 +0100
Subject: Global mutex/lock
References: <388CB723.7EE04D7C@ina.fr> <388ED782.E1B8F00A@inrialpes.fr>
Message-ID: <388F106C.320C4A4C@ina.fr>

<French>
(Vladimir , bonjour puisque tu ?cris de France, et je sais que tu es sur
python-fr)
</French>  ;-)

By global locks I mean (as you guessed) inter-process locking.

On Win32 there are synchronization objects such as mutex that fit perfectly,
but I wanted to avoid requiring the install of the Mark hammond's Win32
extensions. Unfortunately, with the solution relying on file locking via the
msvcrt module ( function: locking), I had to solve the case where the same
thread locks may times without unlocking meanwhile (the msvcrt.locking
function blocks in that case). I used the same approach than in the
threading.RLock class (a pair of variables: lockOwner, lockCount) and it
seemed to work at 1st glance. Unfortunately, more careful tests revealed that
there were some subtle race conditions that lead to interblocking, so I
eventually adopted the Win32 Mutex solution.

On Unix, the file-locking approach based on the fcntl.flock(fd, fcntl.LOCK_EX)
seems to handle correctly the pb of the recursive calls stated above.
I don't know about the existence of Python modules that would allow usage of
more 'ipc' oriented solutions.

Cheers,

Richard

Vladimir Marangozov wrote:

> Richard Gruet wrote:
> >
> > Hi Python gurus, Is there a portable global mutex implementation somewhere
> ?
>
> What do you mean by global?
>
> > AFAIK, the locks in the threading/thread modules work only between threads
> of the same module.
>
> They work for all threads within a given (Python) process, yes.
>
> > OTOH, there are file locks for Windows and Unix (respectively in modules
> msvcrt and fcntl), that can serve as locks.
>
> Yes. Locking a shared resource can be used for inter-process
> synchronization.
>
> > Am I right or am I missing something ?
>
> It depends on what you're looking for, but it sounds like you need
> inter-process synchronization of some sort... Different solutions
> exist, some are "portable", some aren't.
> For more feedback, tell us more about your problem. :-)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000126/347c76c0/attachment.html>

From m.faassen at vet.uu.nl  Sat Jan 15 11:36:16 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jan 2000 16:36:16 GMT
Subject: Any Newbies?
References: <s2Wf4.974$sv3.3886@newsfeed.slurp.net>
Message-ID: <85q7m0$kio$2@newshost.accu.uu.nl>

TechSpood <ScrewMe at mypad.com> wrote:
> Anyone besides me out there almost a complete newbie to all programming and
> is learning python for a first language?

You may want to check out the 'tutor' list, you can find more info here:

http://www.python.org/psa/MailingLists.html#tutor

Also, this index at the Vaults of Parnassus is useful, listing various
tutorials:

http://www.vex.net/parnassus/apyllo.py?i=636055170

Anyway, there are definitely other newbies around, and Python is a good
language to have picked! Good luck! Be sure to keep up asking questions
whenever you are unclear about something; you're bound to receive help.

Regards,

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


From alex at magenta.com  Thu Jan  6 05:20:43 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 6 Jan 2000 11:20:43 +0100
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <005501bf582f$b2860ce0$c02b2bc1@martelli>

Chuck Esterbrook writes:


> Seems like many web host providers, such as MindSpring, offer CGI with
> Perl, but not other languages, notably Python.
> 
> Does anyone know who would host a CGI Python site?

I've noticed the same thing, and posed the same question, a
month ago, or so.

The only answer I got on this group was to join the PSA and
get a Starship account -- I did the former, but I'm waiting
for the latter to happen when the Starship admin gets some
time.

I also sent requests to all providers of free web hosting that
currently house Perl CGI only.  The only one which gave a
(belated) favourable response was www.prohosting.com;
their help-person eventually promised they would install
Python too (saying that they used to have it, but nobody
was using it so they took it off).  That was a week or so ago,
and nothing yet has happened, but, one can hope.

I suggest you do as I did -- write everybody you can think
of... eventually, I think we'll see some action!  Also, joining
the PSA is a good idea even if one has to wait a while for
the Starship account -- it _does_ help support Python:-).


Alex





From jjlucsy at concentric.net  Thu Jan 27 09:04:00 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Thu, 27 Jan 2000 09:04:00 -0500
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk>
Message-ID: <2eYj4.1931$UH3.230869@news4.usenetserver.com>

It's at http://www.concentric.net/~jjlucsy/python15.dll

"Paul Robinson" <paul.robinson at quantisci.co.uk> wrote in message
news:38900E46.909421F8 at quantisci.co.uk...
>
> Joel Lucsy wrote:
> >
> > I have a version of python15.dll that I've compiled using the delayload
> > mechanism. Basically this means that it will only load a dll when it
needs
> > it. Running a very basic test on my system (400Mhz celeron,128Mb,Win2K
RC3)
> > the load time went from over 2 seconds to under half a second. If you
wish I
> > can email it to you, or if someone wants, I'll upload it somewhere.
>
> I'd be very interested in that!
>
> If it's about the size of the standard dll then that's pushing it a bit
> on our mail server, so if you could upload it somewhere that would be
> great.
>
> Paul.
>






From akuchlin at mems-exchange.org  Sat Jan  8 14:44:05 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 08 Jan 2000 14:44:05 -0500
Subject: RegExp, Python and strings
References: <857hhb$3kmi$1@www.univie.ac.at>
Message-ID: <3d7lhkqrga.fsf@amarok.cnri.reston.va.us>

"Matthias Huening" <matthias.huening at univie.ac.at> writes:
> In PERL I can do things like this (in one line):
> $A = "Rossum, Guido van; Harms, Daryl; Python, Franz-Josef";
> $A =~ s/([A-Z])[\w]+(?![ \w\-]+,)([ ;-]|$)/$1.$2/g;
> This RegExp results in: "Rossum, G. van; Harms, D.; Python, F.-J."

re.sub will do it; consult the documentation for the 're' module.

>>> import re
>>> A = "Rossum, Guido van; Harms, Daryl; Python, Franz-Josef";
>>> re.sub(r'([A-Z])[\w]+(?![ \w\-]+,)([ ;-]|$)', r'\1.\2', A)
'Rossum, G. van; Harms, D.; Python, F.-J.'

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
    "Hmph. What the hell would you know? You're a dog."
    "Did I ever say I wasn't?"
    -- Destruction and Barnabas, in SANDMAN #43: "Brief Lives:3"



From tjg at avalongroup.net  Sat Jan 22 21:57:27 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Sat, 22 Jan 2000 18:57:27 -0800
Subject: Python online docs are great!
References: <slrn88kouc.pej.nobody@isis.visi.com> <86dpnm$dtg$1@nntp9.atl.mindspring.net>
Message-ID: <388A6E17.DEBA3B7E@exceptionalminds.com>

> >I just want to compliment GvR on the Python online documentation files.
> >They are amazingly well done. They are without a doubt the best online
> >documentation I've used for anything, anywhere, ever.  Often online
> >documentation has been chopped up into pages that are way too small,
> >and there's no useful index or navigation aids.  The result is you
> >spend all your time thrashing around because the needed information
> >is spread over a half-dozen different pages.  I've found the Python
> >documents very well organized, comprehensive, and easy to use.
> 
> Fred Drake deserves much of the credit here. I can say from personal
> experience that the docs have improved greatly over just the last year.

I started using Python four years ago or so, Bought the Python
Programming book which I have this amazing hate/love relationship with,
but most of what I learned as a newbie was from those excellent on-line
docs, and yes the improvements over the last year have made them even
better.

However, I'm still thankful for David Beazley's new book, it's easier to
read in the bathtub.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<



From m.faassen at vet.uu.nl  Mon Jan  3 12:30:35 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 3 Jan 2000 17:30:35 GMT
Subject: Considering using python for something, and need a bit of info :)
References: <0.3903b60a.25a2245f@aol.com>
Message-ID: <84qmbr$ctg$1@newshost.accu.uu.nl>

PartyG2816 at aol.com wrote:
[snip description of muds]

For a mud done in Python, check out:

http://www.accessoft.com/moop/

Not that I ever used MOOP (I did use its predecessor, POO). I just used
that great index of all things Pythonic, the Vaults of Parnassus:

http://www.vex.net/parnassus

There's also a thread involving someone who wants to make a new
Python interpreter that compiles to MUQ bytecodes; apparently also a
mud system.

There's also PyGCS, a tiny one-room mud written in Python. I've used
this one and it does what it promises to do quite well (which isn't
too much, but it's fun, and it's probably interesting to disect it's
source to learn more about using networking with Python).

http://sourceforge.net/project/?form_grp=378

>     Could the Python interpreter be embedded and used in a way that would 
> allow me to use it like this within my application?

Sure, it can. There's I think one problem remaining, which I'll describe
later on.
 
>  I've downloaded the 
> interpreter and gone through parts of the tutorial... I love it. *chuckles* 

I had the same feeling.

> I'm a C++ programmer at heart,

Came from the same direction. I must admit I don't do too much C++ these
days. :)

> but I can definately see some advantages to 
> using Python for osme things.. I think it would be fairly easy to teach a 
> non-programmer how to use it within the context of my application would be 
> fairly easy with a person dedicated to helping the nephytes and a few docs.

You may end up coding most of your mud system in Python. I warned you. :)
(not that this is a bad idea, it's just that you're probably thinking about
coding it with C++ mostly still -- this could very well change).

>     Another concern of mine is: how secure can it be? IE can i restrict 
> certain aspects of it based upon a condition? Say I have a a normal builder, 
> ie someone who simply makes areas and may perform a few scripts for them, I 
> wouldn't want to let him do certain things... Whereas me as head programmer, 
> I would have access to many more options within my scripts :) Or take a 
> normal "player".  I'm going to be allowing small scripts and macros and such 
> for my players to automate a few 'boring' things, I would want to give them 
> access to a limited set of use, and not let them say re-arrange everything in 
> the room. *chuckles*

There's the rexec and Bastion modules that may help here, but I think you
need to implement most of the security system itself. This can definitely
be done; I imagine MOOP has one, and I'm sure Zope does:

http://www.zope.org

Zope isn't a mud, but it's in Python and does allow people to execute
limited Python expressions in a secure environment. You may want to look
at their security model. Also interesting in this respect is PythonMethods,
a Zope extension that allows people to use Python securely over the
web. I can't explain what Zope does briefly, but be assured people have
looked into this.
 
>     Ah well, maybe you peoples understand what I'm asking and can help, I'd 
> appreciate any response :) If you have questions about something I may need 
> to clarify, then by all means ask and I'll try my hardest. ;)

Okay, the trickiest part which I'm not sure any of the things I listed
solves is the 'infinite loop' problem; what if your ordinary user does:

while 1:
   pass

Does the entire mud hang now? How to manage this?

The classical way an lpmud (for instance) does it is by allocating a limited
number of 'ticks' to each execution thread; if the thread takes too many
ticks then it's automatically aborted. Using ticks also helps in 
simulating multithreading. Default Python however does not support such
ticks, but possibilities you may wish to look at are:

Stackless Python
http://www.pns.cc/stackless/stackless.htm

I'm excited about this development even though it makes my brain explode
so far. Someone more familiar with it could chime in on if this really
would solve the 'infinite loop' problem.

There's also Python Microthreads, which implement an efficient way to
do lots of tiny threads in Python. Again I'm not sure if this will help
for muds, but it's neat anyway. Last I heard was that its functionality
will be folded into Stackless Python. Look here:

http://world.std.com/~wware/uthread.html

The Zope way to avoid this problem is simply to restrict the kind of code
you can write, I think, and to add extra checks to loops. I'm not sure
how far this is along, though.

> p.s. - If it turns out that yall think Python could be suitable for osmething 
> like this,

Python would definitely be great for this! I've been managing an lpmud
for years and I've been vaguely toying with the idea of adding Python
to the driver (and then slowly displace the legacy lpc code :). Not that
I'll ever get time to do this. I've had at least one lpmud guru agree
with me, after he took a look at Python. 

> could you include a few linkst o places that discuss and give some 
> working examples of embedding python in another application and running 
> scripts through it using whatever API is inherent in Python for that use? :)

You may want to check out the Vaults, and there's the Extending and 
Embedding documentation on www.python.org. Also you may want to check out
Swig, at www.swig.org, though that's more to extend Python with
your own C or C++ modules than the other way around.

PS, I love the Vaults. I can sound like Tim Peters without actually
having to know anything. :)
 
Regards,

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


From hz at lexgen.com  Wed Jan  5 12:15:25 2000
From: hz at lexgen.com (Hao Zhou)
Date: Wed, 5 Jan 2000 11:15:25 -0600
Subject: TypeError: slice index must be int ?
Message-ID: <20000105111525.C1172@lexgen.com>

Is it possible to use long int (say 2000000000L) to 
slice a string with gigabytes size?

Thanks.
Hao



From mgushee at havenrock.com  Fri Jan  7 09:58:52 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: Fri, 7 Jan 2000 09:58:52 -0500 (EST)
Subject: XML to dictionary?
In-Reply-To: <20000107092832.A25967@vislab.epa.gov>
References: <8518kq$59n$1@nnrp1.deja.com>
	<m3puvf5p4h.fsf@havenrock.com>
	<20000107092832.A25967@vislab.epa.gov>
Message-ID: <14453.65324.60677.448505@gargle.gargle.HOWL>

Randall Hopper writes:
 > Matt Gushee:

 >  |data = string.split(data)
 >  |
 >  |dict = {}
 >  |for line in data:
 >  |    if line[0] = '(':         # '(' at the beginning of the line in 
 >  |			      # PYX indicates an element
 > 
 > Pyxie is a good tool.  But just to pick nits, using string.split isn't safe
 > with Pyxie output.  :-)
 > 
 > Consider:
 > 
 >     <?xml version = "1.0"?>
 >     <Greeting>
 >     Wishing You A Merry Christmas!
 >     From all of us (here at the White House)
 >     Author: --Unknown
 >     </Greeting>
 > 
 > string.split gives you a new start tag, attribute, and text element you
 > didn't know you had ;-)

Ah. I see what you mean. Well, how about:

data = string.split(data, "\n")

Matt Gushee



From garabik at melkor.dnp.fmph.uniba.sk.spam  Tue Jan 18 04:44:47 2000
From: garabik at melkor.dnp.fmph.uniba.sk.spam (Radovan Garabik)
Date: 18 Jan 2000 09:44:47 GMT
Subject: Tkinter hangs again under Windows
References: <85v092$3n4$1@nnrp1.deja.com>
Message-ID: <948188619.449404@cdwork.cvt.stuba.sk>

jbhr at my-deja.com wrote:
 : By searching the FAQ and Deja News, the last time I saw this
 : mentioned was May 1999.

 : Python 1.5.2, newest Windows Installer with Tcl/Tk 8.0.5, Windows 95b

 : Well, now I have the problem: tk.mainloop() seems to just ignore
 : any event, but only under certain circumstances.

 : My program opens a file and displays some information of it in a
 : Listbox. If the program is started with a filename argument, this
 : file is read, otherwise a filename is prompted via
 : tkFileDialog.askopenfilename(). The info is then shown.
 : Now, if the program is started with a filename, everything (well,
 : almost :-)) is working fine - I can select items in the listbox,
 : use the menu or the button. If the program is started without a
 : filename, the file select box comes up, lets me select a file, and
 : the info is then displayed as well. The difference starts when I try
 : to select something - nothing happens. A Ctrl-C always breaks in
 : Tkinter.py, line 492: mainloop.
 : This problem doesn't exist under Linux. And it's the second Windows PC
 : with the very same problem.

I have the same problem. I have not investigated it further, though.
However, I do not use tkFileDialog.askopenfilename, but
tkMessageBox.showwarning. (and the program stops responding when there is no
gnuplot module, i.e. when it shows tkMessageBox.showwarning)


the relevant parts of the program:

#!/usr/bin/python
from Tkinter import *
import string, pickle
import tkSimpleDialog, tkMessageBox, tkFileDialog

have_gnuplot = 1
try:
    import Gnuplot
except:
    have_gnuplot = 0

...

class App(Frame):
    def __init__(self, master=None):
        self.list=[] 
        self.frame=Frame(master)
        self.frame.pack()

        if not have_gnuplot:
            self.g = Canvas(self.frame, width="4i", height="4i")
            self.g.pack(side=RIGHT)

        scrollbar = Scrollbar(self.frame, orient=VERTICAL)
        self.listbox = Listbox(self.frame, width=35, yscrollcommand=scrollbar.set)
        self.listbox.bind("<Double-Button-1>", self.edit)
        scrollbar.config(command=self.listbox.yview)
        scrollbar.pack(side=RIGHT, fill=Y)
        self.listbox.pack(side=LEFT, fill=BOTH, expand=1)

        self.loadb = Button(self.frame,
                        text="Load",
                        width=7,
                        command=self.load)
        self.loadb.pack(side=TOP)

        ....

        if not have_gnuplot:
            tkMessageBox.showwarning("Gnuplot missing", "Gnuplot.py not found, falling back to tkinter")


if have_gnuplot:
    g = Gnuplot.Gnuplot()
    g.title('title')
    g.xlabel('x')
    g.ylabel('y')


root = Tk()
app = App(root)
root.title("title")
root.mainloop()

 

-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From mstenber at cc.Helsinki.FI  Mon Jan  3 11:54:59 2000
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 03 Jan 2000 18:54:59 +0200
Subject: Considering using python for something, and need a bit of info :)
References: <0.3903b60a.25a2245f@aol.com>
Message-ID: <al8k8lrrt7g.fsf@sirppi.helsinki.fi>

PartyG2816 at aol.com writes:
<snip>

> the main reason for this post: I wanted to know how suitable Python would be 
> to embed into an environment like I mentioned earlier to be used by mainly 
> non-programmers to perform scripting goals.

Not very. If you want security, and crash-proofness (see the other thread,
and recent post by Thomas Wouters. I'm too lazy to dig it up).

<snip>
>     Could the Python interpreter be embedded and used in a way that would 

embedded. yes. 

> allow me to use it like this within my application?  I've downloaded the 

used. yes.

> interpreter and gone through parts of the tutorial... I love it. *chuckles* 
> I'm a C++ programmer at heart, but I can definately see some advantages to 
> using Python for osme things.. I think it would be fairly easy to teach a 
> non-programmer how to use it within the context of my application would be 
> fairly easy with a person dedicated to helping the nephytes and a few docs.

easy. yes. nice - hell, yes.

>     Another concern of mine is: how secure can it be? IE can i restrict 
> certain aspects of it based upon a condition? Say I have a a normal builder, 

not easily. some things cannot be restricted at the moment without
custom extensions (memory use, CPU use).

<snip>

> p.s. - If it turns out that yall think Python could be suitable for
> osmething like this, could you include a few linkst o places that discuss
> and give some working examples of embedding python in another application
> and running scripts through it using whatever API is inherent in Python
> for that use? :)

-Markus

-- 
	We are MicroSoft.  You will be assimilated.  Resistance is futile.


From ysamama at cybercable.fr  Mon Jan 17 15:41:25 2000
From: ysamama at cybercable.fr (Yann SAMAMA)
Date: Mon, 17 Jan 2000 21:41:25 +0100
Subject: WIN32 : how to list the current processes by their executable name?
Message-ID: <85vur9$1prj$1@news5.isdnet.net>

Hi everybody!!

Is there a simple way to implement an os.kill() that works on all WIN32
platforms?
I know about the FAQ article, but I want the kill function to take the name
of the executable file as parameter, not the Process ID.

It should work like this :


def kill(exe_file):
# assumes we're on a Win 32 platform
    import win32api

    mylist = getcurrentprocesses(exe_file)  # that's the key function, it
retrieves a list of corresponding PID

# if exec_file is "myprog.exe", we kill all instances of this program that
are actually running...

    for pid in mylist:
        handle = win32api.OpenProcess(1, 0, pid)   # that's the FAQ
implementation of kill function
        win32api.TerminateProcess(handle, 0)


Any idea about how writing this function "getcurrentprocesses()" ?


I've heard about a POSIX port of "kill" on NT that accepts a file name
with -f option, but I'm afraid I won't be allowed to install it on my NT
platform...

Thanks in advance,

Yann.




From vetler at ifi.uio.no  Wed Jan 19 17:22:43 2000
From: vetler at ifi.uio.no (Vetle Roeim)
Date: 19 Jan 2000 23:22:43 +0100
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu>
Message-ID: <yayd7qxk8gc.fsf@ringhorni.ifi.uio.no>

[warning: the humour impaired should skip this and go back to
discussing what they think is wrong with python.. preferably somewhere
far away from here]

* Paul M.
> Dear Pythoneers,
> 
> Inspired by recent postings to this newsgroup I've decided that
> Python, and in particular the entire Python user base, has been going
> about things all wrong.  I've decided I would like to be able to code
> in python something like this:
> 
> def fun(a,b,c)
> {
>     if a eq b:
>   if b ne c
>                  print "Oooh, random changes are fun"
>                    a = b = c
>                  end
> end
> }
> 
> Important points to note:
> 
> 1) I can use whatever god-awful indentation scheme I want
> 2) Colons after "def" and "if" are purely discretionary
> 3) The compiler can decide what I mean when I use "="
> 4) Readability is greatly improved
> 
> I can't believe that nobody has changed this before - can't you see
> that MILLIONS of users are crying out for these features!

No, no, no! You've got it _all_ wrong.
It should be like this:

PROCEDURE FUN(A,B,C)
BEGIN
  IF A EQ B THEN
  BEGIN
    IF B NE C THEN
    BEGIN
      PRINT "Bah! This is better!"
      A := B := C
    END
  END
END

*My* points are much, much better than your pathetic points:

1) There is NO ambuiguity as to where each block starts.
2) All names are written with big letters, to make the code easier to
   read. (This obliterates your point about readability my method is
   gazillion times better.)
3) THEN is used instead of those idiotic colons.
4) I say so. So there.


*sigh* .. sorry; i just couldn't leave it alone.


<i-wanna-discuss-this-too>-ly y'rs - vr


From rhww at erols.com  Wed Jan 19 22:56:31 2000
From: rhww at erols.com (Robert Wentworth)
Date: Wed, 19 Jan 2000 22:56:31 -0500
Subject: How to copy a binary file?
Message-ID: <3886876F.20FFEA27@erols.com>

It seems like a simple enough task: I want to copy a
set of JPEG image files, unmodified.  My first attempt 
looked like:

def file_copy(source, dest):
    in_file = open(source, 'r')
    out_file = open(dest, 'w')
    temp = in_file.read()
    out_file.write(temp)
    in_file.close()
    out_file.close()

The library documentation says that "read" without an argument
reads "until EOF", though "for some files like tty's it makes
sense to read beyond EOF."  However, the above function fails
because the initial read() reads only a small number of bytes.
If I try to read() again, there are more bytes to be read.

I suppose I could keep calling read() and writing out the
results until read() returned nothing more.  But this seems
clunky and counterintuitive, and without further testing I'm
not sure this wouldn't leave out some character that needs to
appear between read() results.

What is the preferred way of copying a binary file?

Bob Wentworth

[Please reply to rhww at erols.com]


From piet at cs.uu.nl  Sat Jan 29 11:39:47 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 29 Jan 2000 17:39:47 +0100
Subject: Imagemagick
Message-ID: <wzzotoal2k.fsf@sunshine.cs.uu.nl>

Does anybody have a copy of Python/Imagemagick or pymagick or a working link?
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From milek at task.gda.pl  Wed Jan  5 10:41:38 2000
From: milek at task.gda.pl (Robert Milkowski)
Date: 5 Jan 2000 15:41:38 GMT
Subject: NEWS + threads
Message-ID: <0t387365ddi452en79%milek@korweta.task.gda.pl>

HI.
	Does somebody make a small news reader in python?
Or only threads like tin (based on subject and references)?
Or meybe small cook-book how to make threads... step by step...

-- 
							Robert Milkowski
							milek at task.gda.pl



From mlh at vier.idi.ntnu.no  Wed Jan  5 08:03:19 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 05 Jan 2000 14:03:19 +0100
Subject: Simple cartesian product
References: <000801bf571c$0cf448a0$1a2d153f@tim>
Message-ID: <iacembwr7qg.fsf@vier.idi.ntnu.no>

"Tim Peters" <tim_one at email.msn.com> writes:

> Search in DejaNews for p13x25.
> 
> c'mon-folks-new-millennium-new-topics<wink>-ly y'rs  - tim

Now - you're not going to get that "new millennium"-discussion
started, are you? We are still in the same millennium, century, and
decade as last year. The third millennium starts with 2001.01.01.

<wink>

(And - yes - I should have checked deja befor rambling. But then again
- I like to ramble ;)

> 
> 
> 

--

  Magnus
  Lie
  Hetland


From news at dorb.com  Wed Jan 12 02:20:39 2000
From: news at dorb.com (Darrell)
Date: Wed, 12 Jan 00 07:20:39 GMT
Subject: XMLTreeCntrol 0.1
Message-ID: <mt2.0-27553-947661639@news.informatik.uni-bonn.de>

I mentioned my xmlP tree control as a code example and started getting lots
of hits. But it wasn't really in any shape for public consumption. I fixed a
couple of things and packaged all the required files into a zip.
http://www.dorb.com/darrell/xmlP/

Also a screen shot http://www.dorb.com/darrell/xmlP/scrn.jpg the numbers to
the right are the line number.

It will view anything with tags in it.
Things like this aren't handled: <![CDATA[         expr="age > 18" ]]>
That came from the Zope site, so I have some more work todo someday.

If you have wxPython it should work at least as good as it does for me.

== 
--Darrell

<P><A HREF="http://www.dorb.com/darrell/xmlP/">XMLTreeCntrol
0.1</A> - a simple graphical XML tree control for
wxWindows/wxPython.  (08-Jan-2000)

-- 
----------- 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 jody at sccsi.com  Fri Jan 21 00:51:38 2000
From: jody at sccsi.com (Jody Winston)
Date: 20 Jan 2000 23:51:38 -0600
Subject: Talking to embedded python
References: <m3snzsdq7l.fsf@brain.acme.org>
Message-ID: <wkiu0odlat.fsf@sccsi.com>

Joe Cotellese <joec at brain.acme.org> writes:

> My question is how can I communicate with the python interpreter built
> into my application? The idea being that once the program is running I
> can send commands in the form of python scripts to the running
> program.

Look at the medusa code.  It has a method of talking with your program.
Maybe you want to use XML-RPC or COM.

-- 
Jody Winston


From WalterChristler at nhaooyubcg.com  Fri Jan  7 12:03:53 2000
From: WalterChristler at nhaooyubcg.com (WalterChristler at nhaooyubcg.com)
Date: Fri, 07 Jan 2000 17:03:53 GMT
Subject: .,.IF AOL WAS A CAR..,,
Message-ID: <7454F3DDB36D3.kyqmn@ynrndrelsno.com>

1.  The AOL car would have a TOP speed of 40 MPH yet have a 200 MPH Speedometer.  

2.  The AOL car would come equipped with a NEW and fantastic 8-Track tape player.  

3.  The car would often refuse to start and owners would just expect this and try again later.  

4.  The windshield would have an extra dark tint to protect the driver from seeing better cars.  

5.  AOL would sell the same model car year after year and claim it's the NEW model.  

6.  Every now and then the brakes on the AOL car would just "lock-up" for no apparent reason.  

7.  The AOL car would have a very plain body style but would have lots'a of pretty colors and lights. 

8.  The AOL car would have only one door but it would have 5 extra seats for family members.  

9.   AOL car mechanics would have no experience whatsoever in car repair.

10. If an AOL car owner received 3 parking tickets AOL would take the car off of them.  

11. The AOL car would have an AOL Cell phone that can only place calls to other AOL car cell phones.  

12. AOL would pass a new car law forbidding AOL car owners from driving near other car dealerships.   
13. Younger AOL car drivers would be able to make other peoples AOL cars stall just for fun.  

14. It would not be possible to upgrade your AOL car stereo.  

15. AOL cars would be forced to use AOL gas that cost 20% more and gave worse mileage.  

16. Anytime an AOL car owner saw another AOL car owner he would wonder, M/F/age?  

17. It would be common for AOL car owners to divorce just to marry another AOL car owner.  

18. AOL car owners would always claim to be older or younger than they really are.  

19. AOL cars would come with a steering wheel and AOL would claim no other cars have them.  

20. Every time you close the door on the AOL car it would say,"Good-Bye." 

Ewembis qoztp hndp esjlvdz mtsvy?

Wyfuei msfjfni ratihb kkelr lvsd
isxp telu soil clrv lrlp
lre wkebu pxp upl hfkj clck
eely etisfp vlbo co
mob nsij eyj fs lu?

Qlp felvsa dvheiy xsap tplq a gaxt
ejp ofgebr bd keli fbfe!

Vezl yku iogtjom cr
oxmp ekww plcx eec ssp ppbs
uklot frprre ir mum o ehlef yvp
yrrd sul zulmi esj eekt
iiejdh il ddes o xzhm
asqee twdsd knse mrae dpeui rq.

Lpil eln nlp yk.

Afoe upecu sjlf xwx
scmk vpll lxjoz dfs o ew
mllv en ffsnk yi ej
dld ckmsl uri lvasm fmssn hewfi.

Dkf slcm fgya ke eisk?

Teus jyl ndkwpo lus
esi gtt kyd i rlsls
amfp y ohy yele mtpek eikuq
fmay edxc lbftkbd aerp ope rp
npc ls fg kf bs
vfe ry kb vi au
kyel zesoa tkee cmew mku
cue brwal libeso isghe rswum wn
ce kif eke uy
upoe lemed qybw a mej.

Upas elefg stqml kduj if
kkovr iy a fkmshe kbb
rii tfp bif mpf lytec o ddnl
crkp pmeh tiekps eur iqzk
yokp cxmlex hcj cpg?

Wemui spke eoflfl rklid kpp.

Cdkol mssl usffw fdeju
xlj lllcf fiz ycsbo eer
dt fnui iyh jfi
kdloul eps raeiom o rwlexb fbso
clx evfclm fsee pm?

Ata ksfsbeq rfnvrzc xitxlw i dtirt
adl elfete kiudnm eeeyup dkten inkz
si eb skpjt eksgn nyz
llmrk blnp sham csm
yct bgcf ralv a iicwe fk
ekkmkf fekxk qa nodl.

Pldpneo lslise ueh kiad cmeb
oditfv ekvox nzunx jpfxg
fbalipe gwc eermi ps.

Cbfepcl a cdu bit sfreh ttln un
bun rtek ngfmto eifsa
spf nud a iov gfc!

Qlnye ufk mdm rys rldfkm ma!




From eddy at vortigen.demon.co.uk  Mon Jan 17 19:52:45 2000
From: eddy at vortigen.demon.co.uk (Edward Welbourne)
Date: Tue, 18 Jan 2000 00:52:45 +0000 (GMT)
Subject: Suggestions for python 2
In-Reply-To: <200001170221.VAA10441@eric.cnri.reston.va.us> (message from
	Guido van Rossum on Sun, 16 Jan 2000 21:21:35 -0500)
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us>
Message-ID: <m12AMt3-00007rC@vortigen.demon.co.uk>

OK, some examples would be a good idea, and Evan has shown me some ways
of saying tunneling clearer ... Evan has at least one further change he
wants here but I'd rather leave it to him to describe ... if only for
the hope that he'll express why this is a useful idea - albeit he spells
it different - and he expresses it better than I do.
I only know how to design and implement stuff.
Computers understand my better than folk do.

Three new kinds of argument come into being (with my given spelling):
  * safe tunnel: after both * and **
  * keyword-only: after * but not **
  * positional-only: after ** but not *

I shall not discuss the last: they arise only because `the obvious'
structure (Evan may disagree on that) leads to them, but I can't think
what they'd be for.  I'd be delighted to hear if anyone else can ...

First, tunnelling, then some keyword-only stuff.

Safe tunnel:

You want to let your function see some value you can compute, but you
want it to be able to take arbitrarily many positional args; or you need
to be sure your caller can't mess with one of your args.

  def filename(*fragments, **, root=os.getcwd()):
      """Resolves relative paths."""
      # Assume os.path.join joins arbitrarily many fragments.
      return apply(os.path.join, (root,) + fragments)

  def register(name, callback, *, **, book=registry, checker=check_callback):
      grief = checker(name, callback)
      if grief: raise TypeError, 'Unsuitable call-back'
      book[name] = callback

Having someone over-ride the book argument wouldn't be our problem -
it'd mess up their registry, we don't care - so book doesn't need to be
a safe tunnel, an ordinary keyword parameter would do: but we're not so
calm about them messing with the checker we use before storing stuff in
*our* registry.

Keyword-only:

You want to be able to configure some aspect of your function's
behaviour, but you want it to accept arbitrarily many of the arguments
it primarily accepts (and you don't want to have to supply them as a
list, because your users *are* going to forget and/or find it
anti-intuitive).

Have you ever wanted to be able to call string.joinfields without
needing to pack the args up as a list ?

  def join(*fragments, glue=' '): 
      return string.joinfields(fragments, glue)

or wanted to pass popen its command in argv form, and it wouldn't hurt
to be obliged to give mode and bufsize in keyword form, i.e.
  def popen(*argv, mode='r', bufsize=2048): ...
so that some entries in argv can have spaces in them without obliging
the caller to construct a suitably-quoted string,
  popen('sed', '-e', r's:\\:/:', filename)
as opposed to
  os.popen(r'sed -e "s:\\:/:" "%s"' % filename)
and how many folk might have forgotten that filename might be
r'\\samba\shared programs\path builder.com', so missed off the "" round
the %s ?

Or you just want to insist on keyword-form for some of your arguments,
e.g. (future-proofing) because any code which relies on its positional
order would be an albatross round your neck when, at a later date, you
want to add an optional argument before it that really does have a right
to be positional.  It's bad enough knowing you'll always have to support
the keywords that seemed like a good idea in the first version without
having them prevent you using the much better approach that came along
later.  There's loads of places where obliging your callers to use
keyword-arguments would make life much easier for folk reading their
code; and I care deeply about maintainability, which this aids.

Coming up with examples of future-proofing is hard: I'd have to find
Real Code in which I'd needed the functionality, but I could only find
that if I had used it which, of course, I didn't.  It wasn't available.
Especially as most of what I've ever written is 1.3-compatible, and
where I need functionality I don't have I tend to route round the need
for it, so that the resulting code contains no hint of what I (somehow)
avoided needing.  Here's two examples out of a trawl:

class File:
    # ...
    def lock(self, *, limit=None,):
        # ...

some day, we may be adding configuration options to File.lock() (well,
OK, File.__meth__.lock) but all we know as we implement our first cut at
it is that we're going to allow the caller to limit how long it waits
for the file to come free before giving up.  We know we're probably
going to have some fancy options - at least on OSes which don't support
locking functionality of their own - for identifying how we'll do the
locking; but we're only interested, in the first cut, in something
that'll work.  So we insist that the limit argument be given as a
keyword, so that code using the first version can be supportable by
later versions, without messing up the call-interface of File's .lock in
later versions.

def rmdir(name, *, recurse=None): ...

OK, so we're happy about calls to rmdir('~/junk') but who on earth is
going to understand rmdir('~/junk', 1) - what on earth does that 1 mean ?
so we *force* the caller to make itself intelligible by saying
rmdir('~/junk', recurse=1), which at least more of the code's readers will
understand.  [Aid to maintainability.]

There are more examples in python.py, which may be more intelligibly
read (given the high verbosity levels of my comments and docs) as
justthecode.py, linked to from
http://www.chaos.org.uk/~eddy/dev/toy/python2.html and in the same
directory as it.  Sorry, still no py2html prettified forms.

Examples of using generators will have to follow when I have more time.
However: they have to exist because what else is `class', and the
initspace generator illustrated (aka submodule under Novelties) is the
`most obvious' example of a useful generator - you get to assign to and
def whatever names you like within its suite, thereby creating an
immutable sub-domain of whatever context did it, with a minimum of fuss.

	Eddy.
--
Recipient-list truncated to python-list,



From emile at fenx.com  Mon Jan 17 20:39:22 2000
From: emile at fenx.com (Emile van Sebille)
Date: Mon, 17 Jan 2000 17:39:22 -0800
Subject: Sucking e-mail from Exchange Server vis HTTP?
Message-ID: <034a01bf6154$dba16d80$01ffffc0@worldnet.att.net>

If you really want to write a program to pull the mail, then targeting
the web interface probably makes as much sense as anything, but if what
you want is to get your mail, why not use a rule or the out-of-office
thingy to bounce your mail off the 'net to a *real* mailbox?

your-sys-admin's-already-shut-off-out-of-office-anyway-ly y'rs

Emile van Sebille
emile at fenx.com
-------------------



Grant Edwards <grant at nowhere.fenx.com> wrote in message
news:<slrn886kml.284.grant at grante.comtrol.com>...
> I just found the http server page on Exchange server that
> allows access to e-mails via a web browser.  What I'd really
> like is to write a Python program that just sucks the mail out
> of the inbox and delivers it locally.
>
> I've never done any Python code that accesses an http server.
> Are there any applications which would be good examples?
>
> --
> Grant Edwards                   grante             Yow!  Oh, I get
it!! "The
>                                   at               BEACH goes on,"
huh,
>                                visi.com            SONNY??
> --
> http://www.python.org/mailman/listinfo/python-list
>





From claird at starbase.neosoft.com  Sun Jan 23 02:51:37 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Jan 2000 01:51:37 -0600
Subject: Python compiler?
References: <86duuq$8a1@OM9.omantel.net.om>
Message-ID: <5DDFB8C6100AAA81.F1EF5089AEE94B83.043E498410671F19@lp.airnews.net>

In article <86duuq$8a1 at OM9.omantel.net.om>, J Dubal <khimj at gto.net.om> wrote:
>Hello good people!
>
>I am new to python and I understand that although it is a free-to-use
>language, it can be used to build commercial, propritary applications.
>Is this correct?
Absolutely.
>Does it have a compiler for Linux and/or windows that can produce native
>executables?
Yessss--that is, yes, although explaining it
is a bit more involved than just saying, "Here's
the compiler."  Start with <URL:http://
www.deja.com/=dnc/getdoc.xp?AN=570043368>.
>If not, do we have to distribute python application sources?
>Are there modules or libs available to bulid data entry/query applications
>for character terminals?
Kind of.  Not particularly.  <URL:http://
www.deja.com/=dnc/viewthread.xp?AN=558223468>
hints at several approaches to the character-
terminal.  Nothing is wrapped up in a neat
package, though.
			.
			.
			.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From wlfraed at ix.netcom.com  Thu Jan 20 23:28:04 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Thu, 20 Jan 2000 20:28:04 -0800
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu>
Message-ID: <vtyHOHySq7+oAiq8rRQbHOYqDgQA@4ax.com>

On Thu, 20 Jan 2000 11:43:21 -0500, Mark Hathaway
<hathawa2 at marshall.edu> declaimed the following in comp.lang.python:


> 
> Also, to counter the argument of someone who gave a wild example of
> using
> the "end", but not enforcing the indentation rules; enforce the
> indentation
> rules AND use "end". It's even clearer that way.

	But...

	If you are inforcing the indentation rules, then the "end"
becomes superfluous to the language parser.

	Being superfluous to the parser means it is effectively a
comment.

	Therefore, it is no different from the suggested usage of
"#end". Using this structure would mean one could also use "#fi",
"#elihw", "#end if", "#wend", or "#end while" (or any other preferred
block delimiter -- heck, even "#{" and "#}" for C-like start and end
markers).

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From tjg at avalongroup.net  Wed Jan 26 12:05:13 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Wed, 26 Jan 2000 09:05:13 -0800
Subject: strftime and windows
Message-ID: <388F2949.88B3517C@exceptionalminds.com>

Hi all,

I have some really nice code that uses the mxDateTime and strftime to
produce lovely formatted dates. I just discovered that it won't work on
a Win box. Am I missing something, or do I actually have to write a
function to format my dates?

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>Linux...Because rebooting isn't normal<<<<<<<<<<<<



From ivanlan at callware.com  Thu Jan 20 15:49:11 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Thu, 20 Jan 2000 13:49:11 -0700
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com>
Message-ID: <388774C7.1784B662@callware.com>

Hello all--

tye4 wrote:
> 
> I have Python installed on Linux Redhat 6.0. Does anyone know where the
> source code for the compiler is located (any .tar file?). I intend to modify
> the language's syntax to add
> 'end' delimiters for the blocks.
> And since Python is copyrighted, I don't intend to flood out versions to
> anyone else... strictly for personal use.
> 

RTFM:  http://www.python.org

If you had bothered to use Python, you would already know that it is not
a compiler but an interpreter.

If you had bothered to read the copyright notice, you would have learned
something.  It allows you to do any gross disgusting thing you want with
your copy of Python, including modify it and sell it.  I encourage you
to sell the results of your experimentation.  If you ask a high price of
those hundreds or thousands of programmers out there clamoring for it,
you will be able to retire, or at the very least branch off and form
your own newsgroup.

However, I should remind you again.  Shit is not Python.

> class TestNew
> 
>     def foo(s):
>         for i in range(0, 10):
>             print i,
>         end loop
> 
>         if i == 10:
>             print 'internal error'
>         end if
> 
>         while i < 10:
>             print i
>         end loop
>     end def
> 
> end class
> 
> My end delimiters are inspired by Ada, which in my personal opinion has best
> overall syntax.
> 

<so-use-ada>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From juxjux at hotmail.com  Wed Jan 12 08:31:52 2000
From: juxjux at hotmail.com (jux)
Date: Wed, 12 Jan 2000 14:31:52 +0100
Subject: Newbie: Python Accepted?
Message-ID: <85hvor$sno$1@zingo.tninet.se>

Hi.

I've just made my first Python-script and signed up at
free.prohosting.com........ (I can't afford to pay for webspace)
Since they don't accept Python-scripts but only Perl-scripts I wonder if you
can help me.
Is there anyway to convert my script to a Perl-script (So I don't need to
learn Perl),
OR do you know any web-server where I can get some space for free and they
accept Python-scripts?


I hope you can help me! (A hopeful Newbie)

In case you want to look at the script (Maybe you could help me to make it a
Perl-script?) i put it here:



import string
import cgi

form = cgi.FieldStorage()

if form.has_key("email"):email = form[email].value

if form.has_key("firstname"):firstname = form[firstname].value

if form.has_key("lastname"):lastname = form[lastname].value

if form.has_key("text"):text = form['text'].value

filehandle = open ('contact/' + email+ '.txt', 'w')
filehandle.write(firstname)
filehandle.write(lastname)
filehandle.write(text)
filehandle.close()





From NOSPAM_rdrew at cia-group.com  Fri Jan 14 08:04:05 2000
From: NOSPAM_rdrew at cia-group.com (Ray Drew)
Date: Fri, 14 Jan 2000 13:04:05 -0000
Subject: JPython with VisualAge for Java
Message-ID: <85n6tg$aat$1@soap.pipex.net>

Is anyone using JPython with IBM's VisualAge for Java? If so, in what way
are you using it?

Thanks

Ray Drew




From digitome at iol.ie  Mon Jan 17 16:45:11 2000
From: digitome at iol.ie (Sean Mc Grath)
Date: Mon, 17 Jan 2000 21:45:11 GMT
Subject: re-opening stdin in raw (binary) mode?
Message-ID: <388388c9.12001026@news.iol.ie>

The subject says it all. Can some kind soul shed light
on how to re-open sys.stdin in raw mode?

regards,
Sean McGrath




From scarblac-spamtrap at pino.selwerd.nl  Mon Jan 31 12:52:47 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 31 Jan 2000 17:52:47 GMT
Subject: list -> dict help
References: <3895BE15.A7CF569B@bam.com> <slrn89bidu.p04.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <slrn89bipe.p04.scarblac-spamtrap@flits104-37.flits.rug.nl>

Remco Gerlich wrote in comp.lang.python:
> def convert(l):
>     if len(l) == 1:
> 	   return l[0]
> 	else:
> 	   return { l[0] : convert(l[1:]) }
> 	   
> But this copies the list every call, and recursion can use a lot of stack.
> Since it's tail recursion, there must be an iterative version.

Actually, this being Python, it's all done by reference and the lists
aren't copied all of the time. My brain is still thinking in terms of
other languages now and then...

That means the recursive solution is quite good already, I think.
Can't benchmark them at the moment though.


Also, my editor treats .py files and Usenet posts differently space/tab-wise, 
and copying the functions from my testfile mangled the indentation :(.


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
Hi! I'm a .sig virus! Join the fun and copy me into yours!


From tseaver at aftershock.neosoft.com  Fri Jan  7 13:46:43 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 7 Jan 2000 12:46:43 -0600
Subject: lazy (?) evaluation, request for advice
References: <000701bf57e5$9aa20560$b1a0143f@tim> <006001bf5839$ade66180$c02b2bc1@martelli>
Message-ID: <68D3964FA65CA972.FD609140774C2452.C12C753C503ED414@lp.airnews.net>

In article <006001bf5839$ade66180$c02b2bc1 at martelli>,
Alex Martelli <alex at magenta.com> wrote:

<snip timbot's underappreciated "you-want-advice-with-your-irony?">

>Grown-ups, too, appreciate convenience -- which, inter alia,
>means "appropriate level[s] of simplicity/power trade-off[s]".
>
>Rather than offering just one (rich, powerful, complex) way
>to express deal selection, I would prefer to offer users the
>convenience of choosing different points along the curve of
>power vs simplicity according to current needs -- from just
>clicking a few radiobuttons on a GUI, to entering a plain and
>simple selection expression about one hand, all the way to
>subclassing the deal-generator class to tweak and customize

I think maybe you've hit it on the head here -- put the effort into the
expression builder GUI, so that the only time they need to enter the
expression by hand, the additional effort of using object notation may
actually pay off in increased clarity (e.g., 'north.points() > 16 and
south.losers() > 5').  If your GUI generates the expression, and they
can see the result, they get to learn by example anyway.

Sounds like a fun project (I'm back twiddling my Delphi IRC poker client
again this morning),

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From herzog at online.de  Fri Jan  7 05:56:40 2000
From: herzog at online.de (Bernhard Herzog)
Date: 07 Jan 2000 11:56:40 +0100
Subject: Reading tar/gz-files
References: <3874DDAA.E14CC23B@bibsyst.no> <p6qzoujouig.fsf@informatik.hu-berlin.de> <m3u2kqvl82.fsf@greebo.nodomain.de> <3875A663.CB015AFA@bibsyst.no>
Message-ID: <m3puve5evr.fsf@greebo.nodomain.de>

Thomas Weholt <thomas at bibsyst.no> writes:
[...]
> work down to a minimum. Ok, so I`m lazy. The reading tar.h-stuff: eh ...
> I`ll skip that. TarIO implements some of the functionality you say, just
> how much is my question. All I want is to get a list of the files in a
> tar.gz/tgz-archive.

I haven't actually used it, I merely remembered having seen in the PIL
sources. AFAICT, it's used to extract files from a tar file, and one
thing it has to do is search the file in the tar-archive. ISTM that it
would be easy to modify the code to produce a list of filenames.

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


From a.eyre at optichrome.com  Mon Jan 31 10:36:21 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 15:36:21 -0000
Subject: IOError: (0,"Error") when closing "os.popen"ed file?
In-Reply-To: <14485.41767.332703.614501@temoleh.chem.uu.nl>
Message-ID: <000c01bf6c00$ecced140$3acbd9c2@peridot.optichrome.com>

> But maybe I should repeat the flush() until it succeeds, and only
> then close() the file. Since I wrote my first message I have already
> found occasions where even the sleep(0.1) doesn't solve the problem.
> Bad. I really hope I can properly fix this.

TBH it sounds more like a bug in one of the following:

* Python
* Standard C libraries
* OS

I suspect one of the latter two.

Are you not able to trace where/why this odd error occurs?

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From Alex.Martelli at think3.com  Mon Jan 17 12:09:05 2000
From: Alex.Martelli at think3.com (Alex Martelli)
Date: Mon, 17 Jan 2000 18:09:05 +0100
Subject: Global help
Message-ID: <6D8A17398E28D3119F860090274DD7DB4B3DDE@pces.cadlab.it>

Shaun Allen Dishman writes:

> ----- x.py -----
> 
> from y import *
> 
> def x(filename):
> 	file = open(filename, 'w')
> 	y()
> 
> ----------------
> 
> ----- y.py -----
> 
> def y():
> 	# must be the same file object as in x.py
> 	file.write('blah blah blah \n')	
> 
> ----------------
> 
> I have fiddled around with the global settings for the file object but
> cannot seem to get things to work.  All I need is to be able to access the
> file object from one module inside of another one, without passing it as a
> parameter.  Is this even possible?  Thanks in advance.
> 
> 
It's *possible* -- for example, just change the open to

	y.file = open(filename, 'w')

and that, I believe, should "work" -- but I wonder if it's
WISE... you're setting up an extremely strong coupling
between what should be separate modules, by way of
this "shared global variable".  Wouldn't a more explicit
approach be preferable?


Alex




From jjlucsy at concentric.net  Mon Jan 31 08:27:29 2000
From: jjlucsy at concentric.net (Joel Lucsy)
Date: Mon, 31 Jan 2000 08:27:29 -0500
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <949210788.364536@neptune.uniserve.ca>
Message-ID: <J3gl4.8669$iE5.11612400@news4.usenetserver.com>

Try my dll at http://www.concentric.net/~jjlucsy/python15.dll
It uses delay loading to speed startup.

-
Joel Lucsy (jjlucsy at concentric.net)






From grant at nowhere.  Tue Jan 18 17:22:35 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 18 Jan 2000 22:22:35 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com>
Message-ID: <slrn889ptb.3rd.grant@grante.comtrol.com>

In article <slrn889o9j.3oa.grant at grante.comtrol.com>, Grant Edwards wrote:

>------------------------------------------------------------
>import win32com.client
>import win32com.mapi.mapitags
>
>session = win32com.client.Dispatch("MAPI.Session")
>session.Logon("MS Exchange Settings")
>outbox=session.Outbox
>collmsg=outbox.Messages
>msgo=collmsg.Add("subj text","message text",win32com.mapi.mapitags.PR_MESSAGE_CLASS,1)
>msgo.Recipients.Add("grante at visi.com")
>msgo.Sender.Address="grant.edwards at comtrol.com"
>msgo.Send(1,0)
>session.Logoff
>------------------------------------------------------------
>
>It bombs at line msgo.Sender.Address="" line with the following
>error:
>
>com_error: (-2147221233, 'OLE error 0x8004010f', (1271, 'Collaboration Data Objects', ' [Collaboration Data Objects - [MAPI_E_NOT_FOUND(8004010F)]]', None, 0, -2147221233), None)

If I comment out the msgo.Sender.Address line, then I get an
error when I make the msgo.Send(1,0) call:

  File "U:\Grante\python\testSend.py", line 11, in ?
    msgo.Send(1,0)
  File "<COMObject <unknown>>", line 2, in Send
  File "C:\Program Files\Python\win32com\client\dynamic.py", line 207, in _ApplyTypes_
    result = apply(self._oleobj_.InvokeTypes, (dispid, LCID, wFlags, retType, argTypes) + args)
com_error: (-2147467259, 'Unspecified error', (17389, 'Collaboration Data Objects', ' [Collaboration Data Objects - [E_FAIL(80004005)]]', None, 0, -2147467259), None)

???

-- 
Grant Edwards                   grante             Yow!  My Aunt MAUREEN was
                                  at               a military advisor to IKE &
                               visi.com            TINA TURNER!!


From john.mcginnis at usa.net  Mon Jan 17 17:59:48 2000
From: john.mcginnis at usa.net (johnM)
Date: Mon, 17 Jan 2000 14:59:48 -0800
Subject: PYTHONPATH newbie question
Message-ID: <00091c0e.fbb4d1d4@usw-ex0102-012.remarq.com>

I have just recently been twiddling with Python and have to say this is
a great scripting language.

My Question:

How does one make changes to PYTHONPATH permanent? I can import module
sys and call sys.path.append("dir") no problem. I can call sys.path and
see that dir has been added.

But when I exit the shell and reenter the shell the dir entry in
PYTHONPATH has disappeared. I am running Python 1.5.2 on a Windows NT
4.0 SP4 system.

Any hints or am I missing something?

Thanks in advance.

JohnM, the highly impressed Python newbie....



* 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 scarblac-spamtrap at pino.selwerd.nl  Tue Jan 25 04:15:25 2000
From: scarblac-spamtrap at pino.selwerd.nl (Remco Gerlich)
Date: 25 Jan 2000 09:15:25 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be>
Message-ID: <slrn88qq8g.3la.scarblac-spamtrap@flits104-37.flits.rug.nl>

Thomas Hamelryck wrote in comp.lang.python:
> Tim Peters <tim_one at email.msn.com> wrote:
> [snip]
> : like-sex-is-great-but-should-not-be-part-of-life-ly y'rs  - tim
> 
> Good analogy, actually. Sex is great, but should never be forced
> on you.

Noone is forcing Python on you...

> BTW, people here seem to have adopted the idea that if you don't like
> a particular feature of a language, you should abandon it (e.g., "You 
> don't like indentation? Use Perl!"). But many languages have evolved 
> exactly this way: users criticize, complain, praize and whine, thereby
> influencing the shape of the languge in question.

And many of them have become abominations. Python is over a decade old.
Whitespace indentation is an important feature, that makes a lot of sense,
and that makes code much clearer, but is different from other languages.
If you don't like that, there is pindent.py.

That just can't change anymore and won't, so if you really can't live with
it, "use another language" is sane advice, isn't it?

> You can find a very 
> nice description of this process with respect to C++ in "The Evolution 
> of C++ : Language Design in the Marketplace of Ideas", by Jim Waldo 
> (Editor), James Waldo (Editor).

C++ is a good example of an abomination.


-- 
Remco Gerlich,  scarblac at pino.selwerd.nl
 10:12am  up 62 days, 16:17,  6 users,  load average: 2.90, 3.40, 3.62


From jeremy at cnri.reston.va.us  Mon Jan 17 13:54:16 2000
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Mon, 17 Jan 2000 13:54:16 -0500 (EST)
Subject: [PSA MEMBERS] Re: [Types-sig] Suggestions for python 2
In-Reply-To: <200001171837.SAA20866@lsls4p.lsl>
References: <m129zpL-00006JC@vortigen.demon.co.uk>
	<388328F8.85147F85@maxtal.com.au>
	<200001171837.SAA20866@lsls4p.lsl>
Message-ID: <14467.25944.592000.993643@goon.cnri.reston.va.us>

Can we prune the reply-to list?  There's no need for this to be on
psa-members, which is for PSA-related business.  I suggest just the
types-sig, although just python-list would be fine, too.

Jeremy




From tim_one at email.msn.com  Tue Jan 25 03:33:04 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jan 2000 03:33:04 -0500
Subject: In defence of the two-namespace rule
In-Reply-To: <388D5781.79D885F8@pik-potsdam.de>
Message-ID: <000a01bf670e$cce49a00$632d153f@tim>

[Tim]
> Haskell has several shorthand notations for currying of
> various sorts.  For example,
>
>    (- 1)
>
> is the function that subtracts one from its argument, while
>
>    (1 -)
>

[Cezar Ionescu]
> Actually, (-1) is an exception since it refers to the negative
> integer -1 (i.e. 0 - 1)

Ack, yes.  I intended ((-)1) there, as in

? map ((-)1) [1,2,3]
[0, -1, -2]
(41 reductions, 69 cells)
?

but then it wouldn't have looked so obvious -- don't you just hate the way
modern languages treat whitespace <wink>?

> Of course, since the timbot is never wrong, this must have
> been a test!  Do I pass, huh, huh?

Wrong conclusion, alas.  I (Tim) share an email address with the timbot, and
that's how you can tell when I'm posting:  full of errurs an dumb misteakz.
The timbot is temporarily offline to enhance its meanness chip.

and-after-this-public-humiliation-i'm-sorely-tempted-to-
    give-it-a-double-dose<wink>-ly y'rs  - tim





From JamesL at Lugoj.Com  Wed Jan 26 23:39:42 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Wed, 26 Jan 2000 20:39:42 -0800
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <01ea01bf63ec$530bd7e0$f29b12c2@secret.pythonware.com> <86np80$jhp$1@harvee.billerica.ma.us> <86ntlg$2tc$1@news.ycc.yale.edu>
Message-ID: <388FCC0E.C252FF43@Lugoj.Com>

Paul Magwene wrote:
> 
> Eric S. Johansson <esj at harvee.billerica.ma.us> wrote in message news:86np80
> [snip[
> >
> > > hey, who the heck is reading code over the phone
> > > these days?
> >
> > anyone trying to program using speech recognition.  Why would they
> > do this you might ask?  Probably because they were injured from
> > programming using keyboards.
> >
> > yes, it's not strictly reading code over the phone but there
> > are significant similarities...
> >
> 
> Yes, but isn't it possible to set the speech recognition software so when
> you say "indent" it tabs right, and when you say "dedent" it tabs left?

Or for that matter, it maps the phrase "begin block" and "end block" to the
correct key sequences for the target programming language. In "vi" with
auto-indent, this would work fine.


From m.faassen at vet.uu.nl  Mon Jan  3 10:29:34 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 3 Jan 2000 15:29:34 GMT
Subject: Module-repository
References: <386A02A2.5CB32ED9@bibsyst.no> <84ctcp$88k$1@news.glas.net> <386A18E7.99420D1E@bibsyst.no> <386B3871.50EF0915@bibsyst.no>
Message-ID: <84qf8u$e9g$1@newshost.accu.uu.nl>

Thomas Weholt <thomas at bibsyst.no> wrote:
>> > http://www.vex.net/parnassus
[snip]
> Well, after checking the site out it looks great too, but some of the
> links, the ones I tested, seem to be dead or extremely slow.

You can report dead links with the site. The files aren't stored in the
Vaults themselves so sometimes links offsite may 'die'.

So don't blame the wonderful Vaults! :)

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


From tim_one at email.msn.com  Tue Jan 25 20:55:14 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jan 2000 20:55:14 -0500
Subject: Integer literal overflow error with -2147483648 but not with 0x80000000 ?
In-Reply-To: <CC59868A0A76D311B1F50090277B321C1F30A8@VRTMAIL>
Message-ID: <001501bf67a0$63cbeb80$862d153f@tim>

[Mike Fletcher]
> Just wondering if there's a reason for this (doesn't cause a
> problem for me, just seems strange)...
>
> PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)]
> on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
> >>> 0x80000000
> -2147483648
> >>> -2147483648
> Traceback (innermost last):
> OverflowError: integer literal too large
> >>>

Python's grammar is such that negative int literals don't exist:  that's
really the unary minus operator applied to the "one too large" positive int
literal 2147483648.  I once tried twisting Guido's arm into special-casing
the snot out of this snotty case, but he declined the implementation
ugliness.  This is, without doubt, the real reason millions of programmers
refuse to use Python <wink>.  BTW, I believe this is the only exception
other than SyntaxError that can get raised at compile time!

Note that int(-2147483648L) works fine (and for those of you who try that
and get an error, search DejaNews for an explanation of what's wrong with
your obsolete version of gcc <wink>).

every-number-has-its-own-tale-to-tell-ly y'rs  - tim





From moeller at networksplus.net  Sun Jan 23 14:46:49 2000
From: moeller at networksplus.net (Derek Moeller)
Date: Sun, 23 Jan 2000 13:46:49 -0600
Subject: CGI?
In-Reply-To: <OtXZUoZZ$GA.275@net025s>; from vincent_berg@hotmail.com on Sun, Jan 23, 2000 at 01:03:27PM -0000
References: <OtXZUoZZ$GA.275@net025s>
Message-ID: <20000123134649.D27888@networksplus.net>

On Sun, Jan 23, 2000 at 01:03:27PM -0000, Vinnieboy wrote:
> Hi,
> 
> Does anyone know where I can find more about, Python-programming and
> especcially cgi-script!
> 
> Greetz from Vinnieboy

There's a good tutorial at devshed.com.

http://www.devshed.com/Server_Side/Python/CGI/

Although Guido mentions in the comments that the code doesn't look right
when printed, and be careful as the backslashes don't appear to show up in
the code for some reason. Nonetheless it is a good tutorial.

-- 
Derek Moeller
  moeller at networksplus.net
  http://dophnic.yi.org



From ame at swipnet.se  Mon Jan 10 18:27:17 2000
From: ame at swipnet.se (Anders Eriksson)
Date: Mon, 10 Jan 2000 23:27:17 GMT
Subject: starting a cgi with params
References: <iQp6OHyH9j2y64kYNrpGPcKoXKu0@4ax.com> <20000110133338.A15914@quark.emich.edu>
Message-ID: <387a686e.3028554@nntpserver.swip.net>

On Mon, 10 Jan 2000 13:33:38 -0500, Jeff <jam at quark.emich.edu> wrote:

>> How do I send 2 params to the script when I'm using it as an CGI
>> script?
>> 
>I would suggest looking into the 'cgi' module, which comes with the standard
>distribution. there are also a number of good websites that explain how cgi
>works-- the 'protocol' is generic regardless of platform and server
>platform. 
I have looked in the 'manual' and found the cgi, but somehow I don't
understand... I want to be able to call my script from a link, i.e.
http://yourhostname/cgi-bin/cgi.py?xxx.zip&get

The cgi class seems to be form getting info from a form(?)

What am I missing??

>
>by the way, what platform are you using? linux + apache?

To make everything easy. The developing platform is WinNT+Xitami and
the target platform (where the script will run) is Solaris + Apache

// Anders
PS! This is my home account thats why the user changed!



From fredrik at pythonware.com  Sat Jan 22 10:45:06 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 22 Jan 2000 16:45:06 +0100
Subject: Python dumps core with this.. how come?
References: <38891316.393423@news.erols.com> <005301bf64b5$d2f234d0$f29b12c2@secret.pythonware.com> <3889c4b6.242514@news.erols.com>
Message-ID: <004401bf64ef$a8e1e0c0$f29b12c2@secret.pythonware.com>

Roey Katz <katz at glue.umd.edu> wrote:
> >> Also, I have another question (very old, and I can't find this in the
> >> FAQ):  I want to explicitly take a reference to an integer.
> >
> >1. reset your brain ;-)
> >2. variables are named references, not small boxes
> >   that hold data values
> >3. *all* variables in python are references.  there
> >   is no other thing.
> >4. integers cannot be modified in place.
> 
> Ok, OK:  this is what I always get when I ask this question,
> but then why does this happen:
> 
>      >>> a = 3 
>      >>> b = a 
>      >>> b
>      3 
>      >>> a = 5
>      >>> b
>      3

you missed points (1) and (2) in my mail ;-)

to make it clearer, here's one more:

5. simple assignments don't copy any data,
   they rebind the name

(here, a simple assignment is an assignment
of the form "name = expression".  more com-
licated assignments like "name[index] = value"
are syntactic sugar; they map to method calls
that modify the object referred to by "name")

so the second assignment associates the
name "a" with *another* integer value.

it does *not* modify the contents of the
small box that hold the data value...

any name (like "b") that refers to the
old value still refers to the old value.

got it?

> although that I cannot modify numbers, strings
> or tuples in-place seems to me inconsistent

why on earth do you need to change the value
of "1"?

(if that question doesn't make sense, reset your
brain and start over from (1) ;-)

(sorry, no time to elaborate further.  gotta run.
see you all at the conference!)

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! &smiley; -->




From mikael at isy.liu.se  Thu Jan 20 05:15:11 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 20 Jan 2000 11:15:11 +0100 (MET)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <Pine.SOL.3.91.1000119233305.25615A-100000@sunse>
Message-ID: <XFMail.000120111511.mikael@isy.liu.se>

On 19-Jan-00 Evguenii Smogailov wrote:
 >  New operators:
 >  1) 'eq' (like '==');
 >  2) 'ne' (like '!=','<>');
 >  3) 'last' (like 'continue');
 >  4) '..' for generating sequences (like '1..20' or 'a..z'),
 >  like in Perl.

Just my 17 ?re (Swedish equivalent of 2 cents):
I could live with eq and ne, but I really think == and <> are more 
readable. I've gotten used to !=, but I prefer <>. On the other hand, 
if what I am comparing isn't numbers, then != makes more sense than <>.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    11:09:40

This message was sent by XF-Mail.
-----------------------------------------------------------------------



From volumeregeling at hotmail.com  Tue Jan  4 10:45:19 2000
From: volumeregeling at hotmail.com (volumeregeling)
Date: Tue, 04 Jan 2000 15:45:19 GMT
Subject: Quitting my program
Message-ID: <3872148f.18589850@122.0.0.250>

Hi,

I have a problem ,I don?t know how to stop my program from a function.
I?ll show the little piece of code which I have now. (added extra
spaces on the subject fields)

  Command_button.menu.add_command(label='Quit', underline=0, 
				    command=   Command_button.quit   )

I would like to have the following :

  Command_button.menu.add_command(label='Quit', underline=0, 
				    command=    stop_ program  )

def stop_program()
      command_button.quit


I can?t seem to get the command_button in the stop_program function.

I don't think this is a difficult problem, but i don?t think the help
files from python are the best around (sorry creators !!??!?)

thanks,

Volumeregeling



From sposhua at my.pc  Thu Jan 13 12:46:09 2000
From: sposhua at my.pc (Sposhua)
Date: Thu, 13 Jan 2000 17:46:09 +0000
Subject: file width and height
Message-ID: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com>

Is there a module to find gif/jpeg width and height?




From backov at nospam.csolve.net  Mon Jan 10 14:37:42 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Mon, 10 Jan 2000 19:37:42 GMT
Subject: Tricky interpreter initialization question..
Message-ID: <ayqe4.51620$Ke.35799@news21.bellglobal.com>

Ok, in the past I've been asking a few questions about embedding.. After
reading the responses and continuing research, I was somewhat stumped.. But
now I've gotten to the point in my project where getting Python embedded is
pretty much necessary. So, some questions:

I have a running application, with lots of objects that Python needs access
to, so here's what I think I can do (but am not quite sure how to do it):

Start up my app.

Initialize the Python interpreter and then call one or a series of init
procedures that calls Py_InitModule(), without having used "import spam" to
create the module object. Can I get this to work?

Here's my reasoning:

I don't have any code (.o or .lib) to provide to an import function. I don't
see how I could use import as it's supposed to reference code that's already
running, not that the python interpreter runs itself.

I don't want to shove my skeleton into a .dll and thus "extend" Python.
That's just too messy to even contemplate.

I figure it should be possible to initialize these modules at run-time and
thus get around the first two problems. Writing the glue code isn't a
hardship, but again, the interpreter needs to be pointed to already
instantiated interfaces, not stuff it has to instantiate itself.

Any help?

Cheers,
Jason





From tim_one at email.msn.com  Sun Jan  9 01:38:11 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 9 Jan 2000 01:38:11 -0500
Subject: fastest way to merge lists.
In-Reply-To: <roy-D565B0.21255908012000@netnews.nyu.edu>
Message-ID: <000201bf5a6c$1a133f20$f62d153f@tim>

[Roy Smith]
> I've got 5 (unsorted) lists of strings.  Each list has
> O(10^4) strings in it.  Within each list, each string appears
> exactly once.  Also, there is a large overlap between the lists.
>
> I want to form a single list which contains exactly one copy
> of each string from the combined lists.  In unix, this would be:
>
> cat list[12345] | sort | uniq

Thanks for the careful description!  It helps.

> I'm not yet sure if care if the final list is sorted or not.
> For now, I think it doesn't matter.  What's the fastest way
> to do that?
>
> It seems to me:
>
> d = {}
> for s in list1:
>    d[s] = 1
> for s in list2:
>    d[s] = 1
> for s in list3:
>    d[s] = 1
> for s in list4:
>    d[s] = 1
> for s in list5:
>    d[s] = 1
> list = d.keys()
> del d   # if I care about memory
>
> would be pretty good.  Anything better?

Can't beat it, asymptotically speaking -- it's a linear-time algorithm.
Practically speaking, Python's dicts are highly optimized too.  You can save
a little typing via e.g.

for s in list1, list2, list3, list4, list5:
    for x in s:
        d[x] = 1

If you're keen on small speedups, you can set the dict elements on the same
line as the inner "for" (maybe save a percent or two due to avoiding
SET_LINENO opcodes), or play with operator.setitem to obfuscate <0.7 wink>
the inner loop away.  That's all the tricks I can think of; I wouldn't
bother with either.

> Now, if I want it sorted, I could just sort the above list.

Yes.

> I suspect that if I knew ahead of time I wanted the final
> list sorted,  sorting each individual list then doing a linear
> merge might make more sense, if there was little overlap
> between the lists.

It's much more likely to pay off in another language.  There's no free lunch
here:  if you're sorting a total of N distinct elements via comparisons,
it's going to take order N*log2(N) time in all whether you do them all at
once or sort them in pieces and merge at the end.  It's true that

    N*log2(N)   [time to sort the combined list]

is "bigger than"

    5 * (N/5)*log2(N/5) = [time to sort 5 lists 5 times smaller]
    N*log2(N/5)

but the 5-way merge at the end of the latter takes an average of log2(5)
comparisons (don't think about that one too hard <wink>) at each of the N
merge steps, and

    N*log2(N/5) + N*log2(5) = N*(log2(N/5) + log2(5)) =
    N*log2(N)

That is, same thing!  The key point in *Python* is that list.sort() works at
C speed, but the 5-way merge will work at Python speed, so the latter
strategy will almost certainly be significantly slower.

> But, since I know they overlap greatly (i.e. len(list) will
> be approximately the same as len(listN)), I'm guessing doing
> it as described above and sorting the result might just work
> out faster.

Right, and no question about it in this case -- for large enough N, sorting
dominates the time (it's the only worse-than-linear part of the algorithm),
and sorting one list of len N/5 is going to go about about 5 times faster
than sorting 5 of them.

don't-even-need-to-time-it<wink>-ly y'rs  - tim





From culliton at clark.net  Wed Jan 12 11:29:44 2000
From: culliton at clark.net (Tom Culliton)
Date: Wed, 12 Jan 2000 16:29:44 GMT
Subject: Perl speed vs. Python speed
References: <85hjfa$mlp$1@nnrp1.deja.com> <pd0f4.26480$W2.358425@iad-read.news.verio.net> <00f801bf5d0e$ae601140$f29b12c2@secret.pythonware.com>
Message-ID: <YZ1f4.26511$W2.359579@iad-read.news.verio.net>

In article <00f801bf5d0e$ae601140$f29b12c2 at secret.pythonware.com>,
Fredrik Lundh <fredrik at pythonware.com> wrote:
>Tom Culliton" <culliton at clark.net> wrote:
>> The "new" Python re module needs to be optimized and possibly pushed
>> into C code, the older regex module is about twice as fast.
>
>if everything goes according to plan, 're' (and possibly also
>'regex') will be replaced by a new unicode-aware engine in
>1.6.  I've attached some (somewhat outdated) benchmarks.

Cool.  How much rewriting is it going to require in existing scripts?

>note that 'regex' is faster than 're' only if you apply simple
>regular expressions many times.

Which, I think you'll have to admit, is probably the most common usage.

>if you can reorganize the
>code to use a single regular expression on a larger string,
>'re' beats the hell out of 'regex'.  ...and if you make things
>complicated enough, 'regex' stops working... (but as usual,
>some people prefer to get the wrong answer quickly ;-)

Big complicated regular expressions are a recipe for insanity.  Ask
Tim, and he'll recommend using small simple regular expressions 9
times out of 10.  At least thats how he's responded every time I've
asked about problems with some excessively clever regular expression
here.  Heck, I think you've even told me the same thing on at least
one occasion. ;-)



From skip at mojam.com  Thu Jan 27 12:14:59 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 27 Jan 2000 11:14:59 -0600 (CST)
Subject: newbie question: stdin, stdout, stderr?
In-Reply-To: <hE_j4.7222$S64.211505@nnrp1.uunet.ca>
References: <hE_j4.7222$S64.211505@nnrp1.uunet.ca>
Message-ID: <14480.32019.528356.873165@beluga.mojam.com>

    Warren> How do you first USE, and second REDIRECT stdin, stdout, and
    Warren> stderr from within a Python program?

stdin, stdout and stderr are available as file objects in the sys module:

    >>> import sys
    >>> sys.stdin
    <open file '<stdin>', mode 'r' at 8111298>

You can redirect them by assigning an open file object to them:

    >>> sys.stderr = open("/tmp/errors", "a")

then when you're finished, reassign them using the __ versions:

    >>> sys.stderr = sys.__stderr__ 

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098
"Languages that change by catering to the tastes of non-users tend
not to do so well." - Doug Landauer



From bjorn at roguewave.com  Mon Jan 17 13:17:40 2000
From: bjorn at roguewave.com (bjorn)
Date: Mon, 17 Jan 2000 11:17:40 -0700
Subject: Read In Directory Contents
References: <388358F1.F1771D82@netins.net>
Message-ID: <38835CC4.9FEB9496@roguewave.com>

import glob
list = glob.glob("*.txt")

(note that naming a variable list, is generally not a good idea -- it
shadows a builtin)

-- bjorn

Scott Billings wrote:

> For a program I'm working on, it would be very useful if I could use an
> expression such as:
>
> list = os.getcwd() + "*.txt"
>
> to have the names of all text files in the current dir placed into the
> list. However, when I try this method, I get an error about no such file
> existing.
>
> So, how would I go about doing this?
>
> -Scott Billings-
>
> --
> http://www.python.org/mailman/listinfo/python-list




From pinard at iro.umontreal.ca  Sat Jan 15 20:15:08 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 15 Jan 2000 20:15:08 -0500
Subject: International Python
In-Reply-To: Martin von Loewis's message of "15 Jan 2000 21:29:01 +0100"
References: <oqk8ldg2jp.fsf@titan.progiciels-bpi.ca> <p6qwvpbxenm.fsf@informatik.hu-berlin.de>
Message-ID: <oqr9fi4y1v.fsf@titan.progiciels-bpi.ca>

[My notes tell me I replied to Martin only, while my intent was to include
 python-list as well.  Sorry if I'm mistaken and this happens to be a
 duplicate.]

Martin von Loewis <loewis at informatik.hu-berlin.de> ?crit:

> Are there plans to have the interpreter put the translations into the
> __doc__ strings?  If so, how would that work?

None that I know.  So far that I see, the main (other) difficulty would
be to statically attach a textual domain to the doc string.

If you have a complex Python package made up of components, or sub-packages,
each using a different textual domain, (this is where we all go), it may
be hopeless to try dynamically switch the "default" translation domain.
I would have a suggestion, however.  `_' could be bound, near the very
beginning of each module, by something resembling:

        _ = intl.translator(DOMAIN)

with DOMAIN being component-specific string, that is, always the same for
all modules which are part of a sub-package.  This, in turn, would check
if a Translator object for DOMAIN already exists (just looking it up in a
dictionary).  If not, a new such object would be created.  self.domain would
be bound to DOMAIN, and the `translator' function from that object would
be returned by `intl.translator'.  The Translator.translator function would
be nothing else than a call to `dcgettext' with domain set to self.domain.

Then, to print any string from a given MODULE from outside, including its
doc string, would could merely and safely use MODULE._(MODULE.STRING).

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





From morse at harborcom.net  Fri Jan  7 08:03:45 2000
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Fri, 07 Jan 2000 13:03:45 GMT
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <3875e3f1.230595438@news.oh.verio.com>

Chuck Esterbrook <echuck at usit.net> wrote:

>Seems like many web host providers, such as MindSpring, offer CGI with
>Perl, but not other languages, notably Python.
>
>Does anyone know who would host a CGI Python site?

Netpedia offers free hosting for open source projects.
They've told me they support Python CGI, but I've not
yet tried it.


From garry at sage.att.com  Thu Jan 20 09:41:37 2000
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Thu, 20 Jan 2000 14:41:37 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com> <slrn886oku.2ac.grant@grante.comtrol.com>
Message-ID: <38871EA1.8AF33147@sage.att.com>

Grant Edwards wrote:

> The admin shut off both POP and IMAP to try prevent people from
> using mail clients other than Outlook.  He's convinced
> management that people who don't use Outlook were increasing
> his workload.
> 
> IMAP used to be enabled to allow people to check e-mail from
> home using Outlook Express, but once he figured out how to
> enable the web interface, he told people to use that from home
> and then he shut off IMAP (I was using fetchmail and mutt
> at the time).

while i expect you'll learn a lot solving your technical
problem, you should also pay some attention to your political
problem.  an out of control sysadmin doesn't do anyone any good.
you might try and engage your other colleagues, gather some data,
and enlist some management support.  f'rinstance, why is *his*
workload the dominant factor?  have him defend and document what
actual, measured effects there would be to his workload if he
enabled POP and IMAP.  gather info from outlook admins to 
challenge his assumptions.

or just get some big guys and go kick his ass.

-- 
Garry Hodgson			Every night 
garry at sage.att.com		a child is born
Software Innovation Services	is a Holy Night.
AT&T Labs			 - Sophia Lyon Fahs


From gtalvola at NameConnector.com  Fri Jan 28 18:36:36 2000
From: gtalvola at NameConnector.com (gtalvola at NameConnector.com)
Date: Fri, 28 Jan 2000 23:36:36 GMT
Subject: question about COM and ADO
Message-ID: <86t965$vlq$1@nnrp1.deja.com>

I'm using ADO from Python, and I'm having a problem figuring out
how to do the equivalent of the VB code

    Dim cnn as New ADODB.Connection
    cnn.Open "Provider=Microsoft.Jet.OLEDB.4.0;Data Source=c:\db1.mdb"
    dim rs as New ADODB.Recordset
    rs.CursorLocation = constants.adUseClient
    rs.Open "SELECT * FROM Table1", cnn, adOpenStatic,
adLockBatchOptimistic
    set rs.ActiveConnection = Nothing

in Python.  I'm trying to create a disconnected client-side recordset,
which requires that after I open the recordset, I set its
ActiveConnection property to Nothing (at least that's how it's done in
VB).  I can't figure out the equivalent in win32com.  In particular, it
chokes if I try to set it to None:

>>> from win32com.client import *
>>> cnn = Dispatch('ADODB.Connection')
>>> cnn.Open( "Provider=Microsoft.Jet.OLEDB.4.0;Data Source=c:\db1.mdb"
)
>>> rs = Dispatch('ADODB.Recordset')
>>> rs.CursorLocation = constants.adUseClient
>>> rs.Open( "SELECT * FROM Table1", cnn,
constants.adOpenStatic,constants.adLockBatchOptimistic )
>>> rs.ActiveConnection = None
Traceback (innermost last):
  File "<pyshell#8>", line 1, in ?
    rs.ActiveConnection = None
  File "E:\Program
Files\Python\win32com\gen_py\00000201-0000-0010-8000-00AA006D2EA4x0x2x1
.py", line 390, in __setattr__
    apply(self._oleobj_.Invoke, args + (value,) + defArgs)
com_error: (-2146825287, 'OLE error 0x800a0bb9', (0, 'ADODB.Recordset',
'The application is using arguments that are of the wrong type, are out
of acceptable range, or are in conflict with one another.', '', 0,
-2146825287), None)


Anyone have an idea of how to do this?


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


From tseaver at starbase.neosoft.com  Fri Jan 21 01:04:59 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 21 Jan 2000 00:04:59 -0600
Subject: procmail-like spamfilter ?
References: <3884EA56.4A3C29C8@lemburg.com>
Message-ID: <F80D660560697939.8484D455C9EF0522.8557F3B8383C42F6@lp.airnews.net>

In article <3884EA56.4A3C29C8 at lemburg.com>,
M.-A. Lemburg <mal at lemburg.com> wrote:
>Does anyone know of a procmail like spam filter that I could
>use to prevent all these spam mails arriving at my mail box ?
>
>Ideally it should be written in Python, of course :-) Snakes
>eating spam... you know.

It should be called Lumberjack (except Sean McGrath has already
reported a name clash on that one).

Best,

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From thantos at chancel.org  Fri Jan 28 21:16:33 2000
From: thantos at chancel.org (Alexander Williams)
Date: Sat, 29 Jan 2000 02:16:33 GMT
Subject: statements and blocks
References: <86t0ov$4en$1@news.psc.edu>
Message-ID: <slrn894jc1.g0f.thantos@brimstone.mecha>

On 28 Jan 2000 21:13:03 GMT, John Kochmar <kochmar at pitfiend.psc.edu>
wrote:

>	print "a"
>	print "b"
>		print "c"
>		print "d"
>		print "e"
>	print "f"

Maybe this begs the question, but why not:

    print "a"
    print "b"
    
    # Operational block -- Outputs Crap Level Two
    print "c"
    print "d"
    print "e"

    print "f"

...  That is, use vertical space and a verbose comment to denote the
block.

Python drives you to make form follow function.  You're not actually
setting that block off for any functional reason, its connotational,
not denotational; as such, you should use connotational methods to
convey the difference, in essence, comments and vertical whitespace.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From DOUGS at oceanic.com  Mon Jan  3 15:53:38 2000
From: DOUGS at oceanic.com (Doug Stanfield)
Date: Mon, 3 Jan 2000 10:53:38 -1000 
Subject: Python Books & improving PyAwareness - How to ?
Message-ID: <5650A1190E4FD111BC7E0000F8034D26A0F216@huina.oceanic.com>

[Jason Cunliffe said:] 
> Sad to say there is a very low profile of python books there. Only a
couple of
> copies at best in most stores...compared to shelf-fulls of the other P and
J 
> lanaguage books. Even worse [but perhaps significant] is that in most
stores
> the Python books were extremely oddly placed - sometimes a volume or two
in 
> the multimedia section, sometimes a tome in Internet, somtimes a copy wher
you
> would expect it a thte end of the Perl shelf. In no store did I find all
the
> current published Python books together. 
> 
> I noted that ther was no Python books in Linux sections - which I think is
> rather surprisin and dissappointing, especially given the enthusiasm now
> for Linux sections in these stores.

[to which jblaine said:]
> Why would you expect to find Python books in the Linux sections?  It's
> a programming language available for a pile of operating  systems.

Despite what we know, thats the only place I've found them prior to the
appearance
of 'Python Essential Reference' at Border's.  Apparently all the O'Reilly
books
are about Unix. <wink>

[Jason]
> I doubt that anyone here can directly affect much of this - but I still
> think it is worth some consideration:

I tried a little guerrilla marketing, moving the 'Learning Python' book from
the
Linux section into the programming, 'P's (not where Perl or Java are, BTW,
they
get their own shelves loosely grouped under "Internet"), next to 'Python
Essential
Reference'.  It appeared to work since that copy, having been on the shelves
for
about two months, disappeared within two days, presumably having been sold.
Just as traffic increases at Burger King when they build a McDonald's next
door,
possibly someone compared the two and decided to start with 'LP'.

> Q1: How to improve the bookstore/publisher handling of Python book
> placement?

Don't all the big chains have 'subject matter experts' that make the buying
decisions?  The number of Python references recently in the magazines might
trigger
some interest by them.  I'm always seeing interest area broadsheets floating
around
these stores also.  If 'the other P' books could get some mention in these
it may
affect those in the franchise stores that stock the shelves.  One theory
would hold
that someone on this list is no more than two people removed from
influencing the
right person in some way.

My 1<<1

-Doug-



From gmcm at hypernet.com  Wed Jan  5 08:51:37 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 5 Jan 2000 08:51:37 -0500
Subject: Importing a dicrionary from a file
In-Reply-To: <3872C4EC.20ADE046@trojanslair.zzn.com>
Message-ID: <1265043092-3792484@hypernet.com>

Black Mantis wrote:

[BTW, this is not comp.lang.python.announce material]

> I am a newbie, and i am working on a program that uses a really
> big dictionary. i was wondering if i could store the dictionary
> in a seperate file, let's say dict.dat. Now, at the beggining of
> the program i want to import that dictionary and get it ready,
> how would i do this? is there a way to read the whole input file
> into one variable?

Some simpler alternatives:

- save it with
  open('mydict','w').write(repr(dict))
- load it with
  dict = eval( open('mydict','r').read() )

- save it with
  open('mydict.py', 'w').write( "d=%s" % repr(dict) )
- load it with
  from mydict import d




- Gordon



From tim_one at email.msn.com  Mon Jan 24 02:22:57 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 24 Jan 2000 02:22:57 -0500
Subject: Stackless 1.01 Homepage
In-Reply-To: <388BE68E.B26985CA@tismer.com>
Message-ID: <000101bf663b$d68afdc0$022d153f@tim>

[posted & mailed]
> "Why should I care about this, isn't my Python fine"?
>
> If you know what Generators, Coroutines and Continuations
> are about and want to have them at your fingertips,
> this is the real thing for you.
>
> If not, how about speed?
> About 5 percent faster, binary compatible as a drop-in
> replacement for your python15.dll, unlimited recursion,
> why not give it a try? Just click on the installer :-)
>
> Stackless Python 1.01 with Continuations 0.6
> can now be downloaded from a web page:
>
> http://www.tismer.com/research/stackless/
>
> Fact sheet, links to documentation, source and
> binaries can be found there.

Thanks for setting up a homepage, Chris!  This has had more homes than an
impoverished schizophrenic <wink>.

> ...
> SLP has been successfully compiled on Solaris (Thanks to Jeff
> Collins). It will be used on PalmIII, to solve the 4k stack
> problem.

Excellent.  Take a day off after the conference <wink>.





From tim_one at email.msn.com  Thu Jan 20 15:19:19 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 20 Jan 2000 15:19:19 -0500
Subject: no multithreading?
In-Reply-To: <38877222.9F5BBBA9@iqsoft.hu>
Message-ID: <000601bf6383$a234d420$b72d153f@tim>

[Nemeth Miklos]
> This can only be implemented using real threads.

Python threads are real threads.

> If a Python thread is about to execute a long operation
> implemented in C (for example a database call), this
> thread will be blocked untill the called procedure returns.

The thread isn't blocked, it's off running the C code.  *Other* threads will
be blocked from executing Python code for the duration unless the author of
the C code did some macro magic to release Python's global interpreter lock
(which most C extension authors are pretty good about doing).

> What operating system threads are going to execute the
> other Python threads while this proc is running?

This is unanswerable because it assumes a false dichotomy <wink>.  Python's
threads build on whichever flavor of native platform threads it was compiled
to use; it's a config option at the time you build Python, and varies across
platform and (on platforms that support more than one thread package)
personal preference.

> By the way I bought Mark Lutz's Programming Python, and
> I am a bit disappointed that he does not even mention
> multi-threading in the 870-page book.

Threading still has a relatively small audience (e.g., check out the usual
(lack of!) traffic on Python's Thread-SIG).

> Is there another book that explains more advanced
> features of Python?

A list of current Python titles can be found at

    http://www.python.org/psa/bookstore/

Don't know whether any cover thread programming specifically.  threading.py
is the preferred Python thread API, and was modeled on Java's thread
gimmicks, so Java books are also relevant.

I think Python is a good language in which to learn threading, but so far I
don't know of any intro material written to such an audience.  Guido started
to write a thread tutorial a few years ago, but it's a large, intricate
topic and he didn't get much beyond the first "print" stmt <0.9 wink>.

threading-works-best-for-those-who-already-know-it-ly
    y'rs  - tim





From shouldbe at message.com  Fri Jan 21 08:23:19 2000
From: shouldbe at message.com (Arinte)
Date: Fri, 21 Jan 2000 13:23:19 GMT
Subject: A modest Error (I mean Proposal)
References: <MODh4.1808$D4.62243@newsread1.prod.itd.earthlink.net> <000901bf6405$f7692360$f29b12c2@secret.pythonware.com>
Message-ID: <b5Zh4.3741$L4.141488@newsread2.prod.itd.earthlink.net>

You'll have to excuse me because, I am a newbie to python and I don't do a
lot of doc reading.

This is a embedded python project, lots of calls back and forth from c++ to
python.  How do you know when to call PyErr_Clear?  Do I call it when I do a
rets = PyLong_AsLong(comm);
when comm is suppose to be a string?  I do this, because I don't know how to
type check the PyObjects in C/c++.  I know it is in there somewhere, but I
was trying to be quick and it is like which ever one isn't null will be the
one I use.

"Fredrik Lundh" <fredrik at pythonware.com> wrote in message
news:000901bf6405$f7692360$f29b12c2 at secret.pythonware.com...
[snip]

is this the entire program, or are any user-written
C extensions involved in this?

a great way to get utterly weird errors is to have
a C extension function which ignores an internal
exception, but returns to Python without clearing
the error state (PyErr_Clear).

if you cannot fix the C code, you can clear the
error state from Python, using something like:

    mymodule.badfunction()
    hasattr(None, "none")

just after the C call.

</F>







From tim_one at email.msn.com  Sun Jan 23 16:24:44 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 23 Jan 2000 16:24:44 -0500
Subject: Python Rocks! - get rid of colons
In-Reply-To: <86f1o4$am8$1@thoth.cts.com>
Message-ID: <000901bf65e8$44a75be0$022d153f@tim>

[Tim]
>  ... the connection between colons and blocks was adopted
> from Python's closest predecessor, ABC.  Formal usability
> studies [blah blah blah]

[Will Rose]
> Cool.  But if Python is designed for newbies,

ABC was designed for newbies.  Python != ABC.  Although his work on the ABC
implementation team clearly had strong influence on Guido's philosophy, the
languages are very different.

> does that mean we all have to give up using it after a month
> (or three months, in my case) when we actually have some
> experience with it?
>
> There's certainly a niche for learning languages: but is that
> really where Python is headed?  From the libraries and tools
> available, I hadn't thought so.

Me neither, although I reject the assumption that a language good for
learning is necessarily bad for "real life"; to the contrary, I think Python
already stands as a strong counterexample (although ABC does not).  Guido's
CP4E proposal

    http://www.python.org/doc/essays/cp4e.html

is definitely aimed at newbies, but is much more concerned with the
development environment than with the language per se.  There are several
reasons to suspect that a few aspects of Python today are "plain wrong" for
newbies (case sensitivity and "lossy" integer division get mentioned most
often) -- but many experienced programmers gripe about the same things.

Unfortunately, "experienced programmers" are all over the map, usually
griping about whatever doesn't match the last language they used <0.4 wink>.

e.g.-you're-probably-irked-about-indentation<2.0-wink>-ly
    y'rs  - tim





From hartmut at oberon.noris.de  Sat Jan 15 10:40:35 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Sat, 15 Jan 2000 16:40:35 +0100
Subject: Parser Generator for Python
References: <387E8B81.23C91488@attglobal.net>
Message-ID: <16600443@oberon.noris.de>

Wolfgang Roesner (wolftex at attglobal.net) schrieb:

 >I haven't found a good reference so far for a parser generator (a la
 >ANTLR)

If you are going to build a compiler, I recommend
http://www.csc.uvic.ca/~aycock

 >Wolfgang Roesner


+++hartmut


From just at letterror.com  Tue Jan 18 21:09:06 2000
From: just at letterror.com (Just van Rossum)
Date: Wed, 19 Jan 2000 03:09:06 +0100
Subject: Python Rocks!
In-Reply-To: <87ya9muahl.fsf@fun77.kivikko.hoas.net>
References: <860ftk$bas@news.or.intel.com>
 <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk>
Message-ID: <l03102800b4aac96334f7@[193.78.237.138]>

At 3:16 AM +0200 19-01-2000, Erno Kuusela wrote:
>>>>>> "Nick" == Nick Belshaw <nickb at earth.ox.ac.uk> writes:
>
>    Nick> ! - whats wrong with tabs?
>
>Nothing at all, since no sane editor would ever use a tab
>width other than 8, an no sane person would ever want anything
>else than 8-character indents.

Unix myth. Ancient historic nonsense. A religion nevertheless ;-)

Insane is:

     define indentation as a N spaces and then
     use a tabsetting of M spaces where N <> M

Tabs are meant to be flexible (hey, you *can* change the tab setting in
most sane editors, can't you? Should ring a bell.), so therefore mixing
tabs and spaces is asking for trouble.

To *many* people 8 spaces indentation is just too much. Look at the Python
library: 4 spaces all over! (But had they used tabs anyone would've been
able to decide for themselves...)

Sorry, nothing else to flame about right now.

Just





From Harald.Terkelsen at adm.hioslo.no  Thu Jan 27 11:22:29 2000
From: Harald.Terkelsen at adm.hioslo.no (Harald Terkelsen)
Date: 27 Jan 2000 17:22:29 +0100
Subject: imaplib and fetch results
Message-ID: <m3wvovscvu.fsf@kiwi.P52.hioslo.no>

I've been playing with the imaplib.py module and found some difference
in the returned data of the fetch function depending on the
message_part argument.

The command:

typ, data = M.FETCH('13:17', '(FLAGS envelope)')

returns a list where data[0] is the flags and envelope data from
message 13, data[1] for message 14 and so on. This is what i would
expect.

But the result of the following command is somewhat different:

typ, data = M.FETCH('13:17', '(FLAGS body.peek[header.fields (From Subject)])')

Now data[0] contains the information from message 13 as in the previous
fetch command. I would then assume data[1] to contain the requested
information for message 14, but it doesn't. Instead data[1] contains
the character ')'. data[2] contains the data for message 14 and
data[3] gives me again an ')' and so on.

Does anybody know why it is like this, or is it a bug?

Harald






From michelorengo at netscape.com  Thu Jan 13 11:05:48 2000
From: michelorengo at netscape.com (Michel Orengo)
Date: Thu, 13 Jan 2000 16:05:48 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in>
Message-ID: <387DF8C4.4B3214CE@netscape.com>

I've just started to play around with MAPI. I'm far from being a expert but here
is a sample of what you can do to send a message:

import win32com.client
session = win32com.client.Dispatch("MAPI.Session")
session.Logon("Michel Orengo")
outbox=session.Outbox
collmsg=outbox.Messages
msgo=collmsg.Add("my subject","my
text",win32com.mapi.mapitags.PR_MESSAGE_CLASS,1)
msgo.Recipients.Add("Michel","morengo at morganlabs.com")
msgo.Sender.Address="morganea at bellatlantic.net"
msgo.Send(1,0)

If I've time, I'll write a more useful example. For now, I'm still playing...
Hope this will help

Michel

"George Thom at s" wrote:

> Hi everyone,
> I'd like to know if anyone has worked with MAPI using Python with the aid of
> either Mark Hammond's win32 extensions or Sam Rushing's calldll/dynwin
> package. Some sample code would also be useful, since there is none in the
> distribution.
> Regards,
> George
> --------------------------------------------------------------------------
> George Thomas
> Resident,Planet Earth,Third Rock from the Sun
> georgethomas at pspl.co.in
>
> Flat No. FI-5, Fifth Floor,
> Himagirinath Co-op. Hsg. Society,
> 486A Elphinstone Road,
> Kirkee, Pune - 411003,
> Maharashtra State,
> India.
> --------------------------------------------------------------------------



From JamesL at Lugoj.Com  Thu Jan 20 23:22:37 2000
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 20 Jan 2000 20:22:37 -0800
Subject: Using tabs effectively. Was: Why not use tabs?
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <20000118.19182100@robert.polytopic.com> <38854A89.B3FBFA12@Lugoj.Com> <38858B76.AC22E62B@bogusaddress.com>
Message-ID: <3887DF0D.B102B357@Lugoj.Com>

Joe Smith wrote:
> 
> If one is using all tabs, then putting a comment at the top of the file
> indicating the tab setting seems like a waste.  If one is mixing tabs and
> spaces, then it is needed.  Maybe the comment should be one of the following:
> using all tabs, using all spaces or mixing with a tab setting of X.

Unfortunately, this wont work in general. You may be forgetting that tabs
may also be embedded. For example, consider a file that contains the
following text lines, where ^I indicates a tab:

A^I^Ix
AA^I^Ix
AAA^I^Ix
AAAA^Ix
AAAAA^Ix
AAAAAA^Ix
AAAAAAA^Ix

The intent is to have that look like:

A       x
AA      x
AAA     x
AAAA    x
AAAAA   x
AAAAAA  x
AAAAAAA x

There is only ONE tab setting that will produce that result. You would have
to be either very clever, plow through the possibilities, or be told (via a
comment) for you to discover the original tab setting. (The discovery of the
tab setting used in the example is left as an exercise for the reader.)

Hopefully that explains the reason for my desire to see a comment at the top
of all source files indicating the tab setting, even when spaces aren't
being used.


From hellan at acm.org  Mon Jan 10 08:12:03 2000
From: hellan at acm.org (Jon K Hellan)
Date: 10 Jan 2000 14:12:03 +0100
Subject: dlclose problem
Message-ID: <8766x25avw.fsf@parus.parus.no>

I am confused!

I am helping out with gnumeric, the Gnome spreadsheet. One feature of
this is "plugins". There are plugins for extension languages - Python
and Guile so far, and plugins for various file formats. Plugins are
dynamically linked with dlopen. The problem is that if initializing
the plugin fails, we close it with dlclose, but dlclose gives a
segmentation violation.

When plugins are loaded, the code looks roughly like this:

    handle = dlopen(filename, 0);
    if (!handle) goto error;
    discover init_plugin symbol
    if (!found) goto error;
    res = init_plugin();
    if (res != OK) goto error;
    return;

error:
    dlclose(handle);

init_plugin calls Py_Initialize, registers functions, etc. It
basically works.

But if something goes wrong in init_plugin, we close the plugin with
dlclose. My problem is that this causes a segmentation violation. When
run from a debugger, we get "Cannot access memory at address
0x406f2e34". The address is apparently where the python library was
linked in. I get this even if I immediately return from init_plugin,
without making python calls at all!

To isolate the problem, I made a test program with no application
code. All it does is dlopen the plugin, call init_plugin and
close. However, this program hangs in Py_Initialize, apparently
waiting for a lock. The backtrace is:
(gdb) bt
#0  0x4005d67a in sigsuspend () from /lib/libc.so.6
#1  0x40195703 in __pthread_lock () from /lib/libpthread.so.0
#2  0x4019306a in pthread_mutex_lock () from /lib/libpthread.so.0
#3  0x40195041 in flockfile () from /lib/libpthread.so.0
#4  0x400886bb in getc () from /lib/libc.so.6
#5  0x4015a4f6 in r_long ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#6  0x4015ad1c in PyMarshal_ReadLongFromFile ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#7  0x40157219 in check_compiled_module ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#8  0x4015753d in load_source_module ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#9  0x40157cfe in load_module ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#10 0x401588d5 in import_submodule ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#11 0x401584d3 in load_next ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#12 0x40158162 in import_module_ex ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#13 0x401582a3 in PyImport_ImportModuleEx ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#14 0x40158100 in PyImport_ImportModule ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#15 0x4015c403 in initsite ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#16 0x4015c094 in Py_Initialize ()
   from /usr/lib/gnumeric/plugins/0.47/libgnum_python.so
#17 0x401391e0 in init_plugin (context=0x0, pd=0x804ae50) at python.c:309
#18 0x8048ce1 in main (argc=1, argv=0xbffff994) at plugin.c:49

Is this somehow related to the original problem?

When I linked the plugin code into the test program and just called
init_plugin, everything worked as expected.

To summarise, what we are doing is to dynamically link in code which
embeds python in gnumeric. Are there any suggestions as to what is
going wrong?

Jon


From loewis at informatik.hu-berlin.de  Wed Jan 12 15:00:04 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 12 Jan 2000 21:00:04 +0100
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim>
 <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov>
 <20000111114923.B42270@vislab.epa.gov> <87zouclb8m.fsf@baal.winnegan.de>
Message-ID: <p6qaembhxh7.fsf@informatik.hu-berlin.de>

Siggy Brentrup <bsb at winnegan.de> writes:

> Can you point us to a formal definition? Seems like you initially got into
> trouble explaining it by example :)

I can try to produce one right here; please be aware that languages
may use variations of that.

Let A, B, C, T1 .. Tn and Q1 .. Qn denote types. Let A => B denote
specialization, in the sense "A is more special than B" (i.e. A
inherits from B). For purpose of this definition, let A => A (i.e. a
type is a specialization of itself).

Notice that '=>' defines a half-ordering on types: If A=>B and B=>C,
than A=>C; however, there may be A, B so that neither A=>B, nor B=>A.

Now consider a set of signatures, s1 .. sn. All have the same function
name (say, f), but different argument types. We can now extend '=>'
onto signatures as well: s1(T1 .. Tn) => s2(Q1 .. Qn) iff Ti => Qi for
all t from 1 to n.

Given an operation invocation f(a1 ... an), multi-dispatch selects a
signature from the set of all signatures of f, and invokes the
corresponding operation definition.

To do so, the candidates are selected first from the set of
signatures. A signature si(T1..Tn) is candidate iff for each i,
type(ai) => Ti.

If there is candidate, the invocation is in error (no matching function)

If there are candidates, select the maximal specific candidates from
the set S of all candidates. A candidate si is maximal specific, if
there is no other element sk of S so that sk => si.

If there is more than one maximal specific candidate, the invocation
is in error (ambiguous invocation).

If there is exactly one maximal specific candidate, this candidate is
invoked.

In languages with static typing, it is possible to statically verify
that there will be always exactly one most specific candidate. A
sufficient condition is that the set of signatures with the same name
forms a lattice: If there are two signatures sa and sb so that neither
sa=>sb nor sb=>sc, there must be a signature sc so that sc=>sa and
sc=>sb.

Is that what you wanted to hear? :-)

Regards,
Martin

P.S. This is exactly the procedure that is also used for overload
resolution in Java, only that this is all done statically in Java, and
not via dynamic dispatch.


From aahz at netcom.com  Sun Jan 30 13:27:41 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 30 Jan 2000 18:27:41 GMT
Subject: nesting class definitions?
References: <20000127155551.A2069@stopcontact.palga.uucp> <20000128182031.C16093@xs4all.nl>
Message-ID: <871vqt$a8d$1@nntp8.atl.mindspring.net>

In article <20000128182031.C16093 at xs4all.nl>,
Thomas Wouters  <thomas at xs4all.net> wrote:
>
>Nesting doesn't really *do* anything in python, except bind the class to a
>different namespace. And there is no difference between classes, modules,
>functions, lists, dictionaries or any of the other objects. For instance:
>
>class Viking:
>	class Spam:
>		pass
>
>is effectively equivalent to:
>
>class private_Spam:
>	pass
>
>class Viking:
>	Spam = private_Spam

Yup.  But try this one on for size:

def SpamGen ( foo ):
	class Spam:
		bar = foo

class Viking:
	def __init__(self, foo):
		self.spamClass = SpamGen(foo)

You can now have multiple instances of the Spam *class*, each with
different values for the class-level variables.  I think this is what
people are often trying for when they nest things in Python; you just
need an additional level of nesting in order to generate the multiple
namespaces.

(Yeah, this is essentially meta-class stuff, which I know you know
about, but I think presenting it slightly differently like this will
make sense to more people.  It certainly makes more sense to me --
metaclasses usually make my head hurt.  ;-)
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From news at dorb.com  Tue Jan 11 20:54:08 2000
From: news at dorb.com (Darrell)
Date: Tue, 11 Jan 2000 20:54:08 -0500
Subject: Please, someone tell me what's going wrong here. ;>
References: <GzLe4.54805$Ke.42098@news21.bellglobal.com> <udQe4.3161$Ef6.805744@news.shore.net> <ylQe4.55014$Ke.43355@news21.bellglobal.com>
Message-ID: <t6Re4.2051$mj7.43546@typhoon.nyroc.rr.com>

Pasted this much into the initModuleName function of an existing extension
module and it worked just fine. Also removed everything before it which
caused it to print the path over and over.

     int a;
     Py_SetProgramName("xyz");
     a=PyRun_SimpleString("import sys"); /* import into the namespace */
     a=PyRun_SimpleString("print sys.path");


--
--Darrell
"Jason Maskell" <backov at nospam.csolve.net> wrote in message
news:ylQe4.55014> > void TForm1::Button1Click(TObject *Sender)
> > {
> >     int a;
> >     Py_SetProgramName(Application->ExeName.c_str());
> >     Py_Initialize();
> >     a=PyRun_SimpleString("import xyzzy"); /* import into the namespace
*/
> >     a=PyRun_SimpleString("xyzzy.foo()");
> >     Py_Finalize();
> > }
> >
> > The module never got imported into the __main__ module, only
> > initialized.
>





From dheise at debitel.net  Sun Jan 16 23:20:33 2000
From: dheise at debitel.net (Dirk Heise)
Date: 16 Jan 2000 23:20:33 MEST
Subject: Compiling Python under Borland C++ Builder 4
References: <9nKf4.59481$Ke.56920@news21.bellglobal.com>
Message-ID: <01bf606c$0950fbe0$LocalHost@dreadzone>

I patched and compiled Python under
BCB3, but not into a LIB or DLL, i made
it part of my app. (If you want the 
patch, mail me.)

I don't have BCB4 by now, but i know they
claim to have some new features to import
Microsoft projects. I would be very interested
to hear whether you get that to work...

Dirk

Jason Maskell <backov at nospam.csolve.net> schrieb im Beitrag <9nKf4.59481$Ke.56920 at news21.bellglobal.com>...
[...]
> So, I'm just curious: Are there any C++ Builder types here who have compiled
> it under BCB4? I imported the VC++ project and compiled it successfully, but
> I get a 10k DLL, which I assume is probably incorrect. ;>
> 
> Cheers,
> Jason



From MSteed at altiris.com  Tue Jan 25 18:36:03 2000
From: MSteed at altiris.com (Mike Steed)
Date: Tue, 25 Jan 2000 16:36:03 -0700
Subject: vi and python?
Message-ID: <65118AEEFF5AD3118E8300508B124877073D3C@webmail.altiris.com>

> From: Ben Beuchler [mailto:insyte**NOSPAM**@emt-p.org]
> Sent: Tuesday, January 25, 2000 3:22 PM
> To: python-list at python.org
> Subject: vi and python?
> 
> 
> There seems to be a general preference for emacs in the 
> python community.  I am personally much more comfortable 
> working with vim.  Does anyone have any particular tips for 
> python development within vim? Particular shortcuts to test 
> code, autoindentation settings, etc...  I would particularly 
> like to be able to use some of the idle features with, 
> perhaps, gvim...

Nothing too fancy, but here are the settings I have accumulated for *.py
files.  The ^H and ^M are actually literal control characters.

" indentation
set shiftwidth=4 softtabstop=4 tabstop=8 noexpandtabs
set nocindent smartindent
set cinwords=if,else,elif,for,while,def,class,try,except,finally
set formatoptions=croq2

" disable dedent when '#' is hit
inoremap # X^H#

" python versions of [[ and ]]
map ]] /^[^ \t#]\+.*:^M
map [[ ?^[^ \t#]\+.*:^M

--
M.



From markwatson1 at my-deja.com  Sun Jan 30 13:34:35 2000
From: markwatson1 at my-deja.com (markwatson1 at my-deja.com)
Date: Sun, 30 Jan 2000 18:34:35 GMT
Subject: Q: JPython: AttributeError: java package 'org.python.modules' has no attribute __path__
Message-ID: <87207s$2sd$1@nnrp1.deja.com>

Sorry in advance if this is a FAQ (I looked, but could not
find anything on the JPython bug list).

I am trying to use the jpythonc compiler with the -c or -a
options and I always get this error:

 AttributeError: java package  'org.python.modules' has no attribute
"__path__"

I get this same error under Linux and under NT.

Any ideas?

Thanks!,
Mark Watson    www.markwatson.com


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


From robin at jessikat.demon.co.uk  Mon Jan 10 04:27:57 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 10 Jan 2000 09:27:57 +0000
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu>
 <slrn87fol9.15t.kc5tja@garnet.armored.net>
 <WWIo4DAGXDe4Ew7B@jessikat.demon.co.uk>
 <slrn87imad.thc.wtanksle@hawking.armored.net>
 <slrn87j11s.vc0.kc5tja@garnet.armored.net>
Message-ID: <Y$u7SGAdYae4EwJv@jessikat.demon.co.uk>

In article <slrn87j11s.vc0.kc5tja at garnet.armored.net>, Samuel A. Falvo
II <kc5tja at garnet.armored.net> writes
...

Don't get me wrong, but to attack me for saying I treat COM as too
object oriented and then attack me for being an OO neophyte seems a bit
silly. I'm from the fortran era, I treat COM objects as exactly that;
blobs with interfaces, that doesn't mean I don't appreciate nice OO
languages like Python. The reason I like Python is that it allows me to
"just do stuff".

As for openness I think having seen a lot of the process of DDE OLE OLE2
COM DCOM COM+ from the outside makes me more than a little doubtful
about commitment to open standards. I have to do other work, not become
an expert in all areas of CS. You have a special interest and expertise;
please don't criticise me for not having the same. I won't slam you for
not understanding linear algebra if you complain it's complicated. The
fact that a document describing the 'latest' version of the API is
available doesn't change the complexity or otherwise of the API. M$
describes the Win32 stdcall interface in about one page to someone who
understands a bit about assembly programming. The equivalent for COM is
how much?

I don't disagree with your CORBA assertion. CORBA being less used and
possibly more academic is even more complicated.

I don't pretend to fully understand the various aggregation mechanisms
in COM/DCOM even though the underlying technology is fairly simple. COM
interaction with threading models in win32 is indubitably complex as are
events, but that's probably inevitable given the various threading
models. 

As for verbiage I guess that's a matter of style. Modern practice calls
these things interfaces etc I come from an earlier era where callables
were functions/procedures etc.
-- 
Robin Becker


From Gareth.McCaughan at pobox.com  Mon Jan 10 18:14:05 2000
From: Gareth.McCaughan at pobox.com (Gareth McCaughan)
Date: 10 Jan 2000 23:14:05 +0000
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
	<14452.45436.641890.197509@beluga.mojam.com>
	<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
	<14452.60226.853060.116123@beluga.mojam.com>
	<38774FBC.EB8281CF@maxtal.com.au>
	<slrn87k49l.pup.grant@grante.comtrol.com>
Message-ID: <86so055xky.fsf@g.local>

Grant Edwards wrote:

> At least with the compilers I've used, that depends on the case
> values.  If the switch case values are sparse, the code
> generated is the same as it would be for If/elif.
> 
> For example:
..
> May generate table-lookup code that is O(1), but
> 
>   switch (c)
>     {
>     case      0: [...] break;
>     case   3450: [...] break;
>     case -49393: [...] break;
>     case 233450: [...] break;
>     case -14000: [...] break;
>     case  -9834: [...] break;
>     }
>     
> Would hopefully generate O(n) linear-search code.

It ought to generate code that does something more
like a binary chop. Looking at the output of gcc -O2
on a big random switch, that's certainly what gcc
does.

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


From pbible at home.com  Sat Jan 15 12:15:56 2000
From: pbible at home.com (Paul E. Bible)
Date: Sat, 15 Jan 2000 17:15:56 GMT
Subject: CGI, Apache, and mxODBC
References: <6p1g4.35$iL4.1041@news.rdc1.tx.home.com>
Message-ID: <gX1g4.48$iL4.1177@news.rdc1.tx.home.com>

Just some more information that I found relating to my problem.  I thought
it might in the diagnosis...
These are the error message lines from Apache:

File "c:/apache/cgi-bin/sql.py", line 13, in Main
db = Odbc.Windows.Connect('Allegro','helistest','oilwell')
mxODBC.OperationalError: ('IM002', 0, '[Microsoft][ODBC Driver Manager] Data
source name not found and no default driver specified', 4265)

Thanks again,
Paul


"Paul E. Bible" <pbible at home.com> wrote in message
news:6p1g4.35$iL4.1041 at news.rdc1.tx.home.com...
> Hi all,
>
> First of all, sorry for the length of this message...
>
> Okay, I've been successful in getting Python CGI scripts to run under
Apache
> 1.3.9 under Windows/NT.  Now it's time to add database functionality using
> mxODBC to the script, but I'm encountering some problems.  I have provided
> my script and the Apache environment debug listing below.
>
> The script, as written, gives me a Internal Server Error.  If I comment
out
> the lines dealing with mxODBC (i.e., db = Odbc.Win..., etc.), the script
> works fine; just doesn't do much.
>
> Thanks in advance your help,
> Paul
>
> --------------------------------------------------------------------------
--
> ---
> #!c:\progra~1\python\python.exe
>
> import os, string, sys, cgi, re, Odbc
>
> def Main():
>      sql = 'Select * from cashacct'
>
>      db = Odbc.Windows.Connect('TestDB','testid','testpw')
>
>      # Add the columns
>      cols = ()
>      c = db.cursor()
>      c.execute(sql)
>      cols = c.description
>
>      # Print the HTML header info
>      print "Content-type: text/html"
>      print
>      print "<HTML>"
>      print "<HEAD>"
>      print "SQL Statement Test"
>      print "</HEAD>"
>      print "<BODY>"
>
>      # Add the columns to the Access table
>      print '<table border="0">'
>      print "<tr>"
>      for idx in range(len(cols)):
>           print "<td>" + cols[idx][0] + "</td>"
>      print "</tr>"
>      print "</table>"
>
>      print "</BODY>"
>      print "</HTML>"
>      sys.exit(0)
>
> Main()
>
> --------------------------------------------------------------------------
--
> ---
>
> Shell Environment:
> COMSPEC
> D:\WINNT\system32\cmd.exe
> DOCUMENT_ROOT
> c:/apache/htdocs
> GATEWAY_INTERFACE
> CGI/1.1
> HTTP_ACCEPT
> image/gif, image/x-xbitmap, image/jpeg, image/pjpeg,
> application/vnd.ms-powerpoint, application/vnd.ms-excel,
application/msword,
> application/x-comet, */*
> HTTP_ACCEPT_ENCODING
> gzip, deflate
> HTTP_ACCEPT_LANGUAGE
> en-us
> HTTP_CONNECTION
> Keep-Alive
> HTTP_HOST
> XX.X.XX.XXX
> HTTP_REFERER
> http://XX.X.XX.XXX/blank.htm
> HTTP_USER_AGENT
> Mozilla/4.0 (compatible; MSIE 5.01; Windows NT)
> PATH
>
C:\orant\bin;D:\WINNT\system32;D:\WINNT;C:\RECYCLER\S-1-5-21-862944694-15761
>
02806-697575874-500\DC0;C:\RECYCLER\S-1-5-21-862944694-1576102806-697575874-
> 500\DC0\Perl;c:\Program Files\Symantec\pcAnywhere;c:\Program Files\python
> PYTHONPATH
> c:\Program Files\python;c:\Program Files\Lib;c:\Program
> Files\python\lib\tkinter;c:\Program Files\python\win32com;c:\Program
> Files\python\win32
> PYTHONUNBUFFERED
> 1
> QUERY_STRING
> sql=select+*+from+cashacct&submit-button=Submit
> REMOTE_ADDR
> XX.X.XX.XXX
> REMOTE_PORT
> 1052
> REQUEST_METHOD
> GET
> REQUEST_URI
> /cgi-bin/sql.py?sql=select+*+from+cashacct&submit-button=Submit
> SCRIPT_FILENAME
> c:/apache/cgi-bin/sql.py
> SCRIPT_NAME
> /cgi-bin/sql.py
> SERVER_ADDR
> XX.X.XX.XXX
> SERVER_ADMIN
> webmaster at home.com
> SERVER_NAME
> XX.X.XX.XXX
> SERVER_PORT
> 80
> SERVER_PROTOCOL
> HTTP/1.1
> SERVER_SIGNATURE
> <ADDRESS>Apache/1.3.9 Server at 24.6.21.112 Port 80</ADDRESS>
> SERVER_SOFTWARE
> Apache/1.3.9 (Win32)
> SystemRoot
> D:\WINNT
> WINDIR
> D:\WINNT
> None
> These environment variables could have been set:
> AUTH_TYPE
> CONTENT_LENGTH
> CONTENT_TYPE
> DATE_GMT
> DATE_LOCAL
> DOCUMENT_NAME
> DOCUMENT_ROOT
> DOCUMENT_URI
> GATEWAY_INTERFACE
> LAST_MODIFIED
> PATH
> PATH_INFO
> PATH_TRANSLATED
> QUERY_STRING
> REMOTE_ADDR
> REMOTE_HOST
> REMOTE_IDENT
> REMOTE_USER
> REQUEST_METHOD
> SCRIPT_NAME
> SERVER_NAME
> SERVER_PORT
> SERVER_PROTOCOL
> SERVER_ROOT
> SERVER_SOFTWARE
> In addition, HTTP headers sent by the server may be passed in the
> environment as well. Here are some common variable names:
> HTTP_ACCEPT
> HTTP_CONNECTION
> HTTP_HOST
> HTTP_PRAGMA
> HTTP_REFERER
> HTTP_USER_AGENT
> None
>
>
>




From piet at cs.uu.nl  Sat Jan  8 15:52:00 2000
From: piet at cs.uu.nl (Piet van Oostrum)
Date: 08 Jan 2000 21:52:00 +0100
Subject: bsddb item size?
References: <mrE03AA5nOd4Ew2R@jessikat.demon.co.uk> <wzd7rd4kcd.fsf@sunshine.cs.uu.nl> <1y9ghGAe1nd4Ewk9@jessikat.demon.co.uk> <wzpuvcn3pt.fsf@sunshine.cs.uu.nl> <HnKQ+CAqXzd4EwiV@jessikat.demon.co.uk>
Message-ID: <wzn1qgmglr.fsf@sunshine.cs.uu.nl>

>>>>> Robin Becker <robin at jessikat.demon.co.uk> (RB) writes:

RB> In article <wzpuvcn3pt.fsf at sunshine.cs.uu.nl>, Piet van Oostrum
RB> <piet at cs.uu.nl> writes
>>>>>>> Robin Becker <robin at jessikat.demon.co.uk> (RB) writes:
>> 
RB> thanks that's a great help. How about collisions are they possible?
>> 
>> yes, you can choose to allow duplicate keys.
RB> thanks very much extremely helpful.

Please note that these answers apply to BSDDB version 2, not (or only
partially) to version 1.85.
-- 
Piet van Oostrum <piet at cs.uu.nl>
URL: http://www.cs.uu.nl/~piet [PGP]
Private email: Piet.van.Oostrum at gironet.nl


From anders.eriksson at morateknikutveckling.se  Thu Jan 13 09:56:25 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Thu, 13 Jan 2000 15:56:25 +0100
Subject: Cookies
Message-ID: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com>

Hello!

I have downloaded the Cookie.py module by Mr Timothy O'Malley, but I
can't figure out how to use it:-(

if I have this script 

# ---------------------------------------
import Cookie

kaka = Cookie.Cookie()
try:
	nPos = kaka['nPos'].value
except:
	nPos = 0

for i in range(20)
	npos = npos + 1
print "Content-type: text/html\n"
print `nPos`

kaka['nPos'] = nPos

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

Why doesn't this script increase the value of nPos with 20 each time I
run it? (well I know that it's because the cookie isn't saved, but
why)

// Anders

---
Enlish isn't my first or second language, 
so anything that you find offending is not there on 
purpose it's just due to the translation ;-)


From matthias.huening at univie.ac.at  Thu Jan 13 07:37:43 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Thu, 13 Jan 2000 13:37:43 +0100
Subject: Z39.50
Message-ID: <85kgkj$4l0u$1@www.univie.ac.at>

Hi,

Im am looking for a possibility to access a Z39.50 server through CGI or
ASP. Would that be possible with Python? Is there a Z39.50-module or some
other Python interface to the Z39.50 standard?

Thanks, Matthias

--
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/dt/





From wc at zyan.com  Mon Jan  3 11:34:10 2000
From: wc at zyan.com (John B. Williston)
Date: Mon, 03 Jan 2000 16:34:10 GMT
Subject: Python on NT systems (help a Unix guy, please)
References: <19991218113757.A637@Ridcully.home>
Message-ID: <3871cf49.3713896@news.dslspeed.com>

On Sat, 18 Dec 1999 11:37:57 +1100, "Malcolm Tredinnick"
<malcolmt at smart.net.au> wrote:

When I still ran NT as my main OS, my Python installation was
consuming about 24 MB of my drive. Bear in mind, however, that I had
several libraries and lots of other third-party code lying around in
addition to the original installation binaries.

John


From rvprasad at ksu.edu  Fri Jan 28 01:04:34 2000
From: rvprasad at ksu.edu (Venkatesh Prasad Ranganath)
Date: Fri, 28 Jan 2000 00:04:34 -0600
Subject: how do i get user input?
References: <389117E3.748B44ED@widomaker.com>
Message-ID: <38913172.44397142@ksu.edu>

Hi,

Use raw_input()
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rvprasad.vcf
Type: text/x-vcard
Size: 405 bytes
Desc: Card for Venkatesh Prasad Ranganath
URL: <http://mail.python.org/pipermail/python-list/attachments/20000128/42230962/attachment.vcf>

From super_banane at yahoo.com  Tue Jan 25 11:17:20 2000
From: super_banane at yahoo.com (super_banane at yahoo.com)
Date: Tue, 25 Jan 2000 16:17:20 GMT
Subject: ConfigParser and .ini files
References: <86a1s2$7ha$1@nnrp1.deja.com> <14472.39533.223900.313932@weyr.cnri.reston.va.us> <388C1FEB.D4CDAD6E@murl.com>
Message-ID: <86kia9$i01$1@nnrp1.deja.com>

In article <388C1FEB.D4CDAD6E at murl.com>,
jim kraai <jkraai at murl.com> wrote:
[...]
> >   Does anyone know of a *real* specification for the .ini format?
> > I didn't know this was allowed.
>
> Reference to a *real* spec and why specs are invalid can be found
> in the first paragraph of the "Introduction" at:
> http://www.cloanto.com/specs/ini.html

So... any chance to see the ConfigParser class updated
to support this good-looking spec? Anyone willing to contribute
something nice and neat? I guess it is mostly a matter of
finding a regexp guru...

--
Nicolas


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


From m.faassen at vet.uu.nl  Wed Jan  5 05:57:17 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 5 Jan 2000 10:57:17 GMT
Subject: CPAN for Python?
References: <38720F70.A5DC6F33@ttm.com> <84t52c$1gg$1@nntp9.atl.mindspring.net> <C28D1A44F2165E48.B5C4E34C4CD957A5.44F995461BA2F6ED@lp.airnews.net> <slrn874buo.2do.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com> <A99BAC4F8CFFB5C8.FB3BECE999060AC1.EAF3C286CA4FB724@lp.airnews.net>
Message-ID: <84v82d$m4u$1@newshost.accu.uu.nl>

Tres Seaver <tseaver at aftershock.neosoft.com> wrote:
> In article <slrn874buo.2do.coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com>,
> Stephen Coursen <coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com> wrote:

[a module to access the Vaults]
>>That would be a useful module to have, but for it to work, there would have
>>to be a standard installation mechanism.  Perl uses:
>>     perl Makefile.pl
>>     make
>>     make install
>>     
>>If python had a standard way of installing modules, writing a parnassus.py
>>script would be easy.  Unfortunately, not every module installs the same way
>>(although there seems to be only a few different methodologies).

> The distutils SIG will be reporting at IPC8/DevDay: here's hoping they are
> thinking along theses lines.

I was at least thinking about these lines. Tim (of the Vaults) and the
distutils people have been talking, so I'm hoping something will come of
this too.

I think there's a good chance we'll see something like this (Vaults +
distutils) within this year. That'd be Way Cool. :)

Regards,

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


From wware at world.std.com  Thu Jan 13 15:18:58 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 13 Jan 2000 20:18:58 GMT
Subject: ANN - PyRecode
References: <20000113144712.A666@starship.python.net>
Message-ID: <FoAJ3M.FDA@world.std.com>

Andreas Jung (ajung at server.python.net) wrote:
: PyRecode is a wrapper around the functionality of
: Recode...
: ... very useful when you are working with texts
: in different formats, languages and character sets.

Funny to see this, immediately after I got an email from a
Portugese student. I needed to use Babelfish to read his email
and respond. Would it make sense to include some kind of
Babelfish wrapper in PyRecode? I ran into a little trouble
with different character sets; Portugese has a rich set of
accent marks. Alas, there is no English<->Dutch translation
in Babelfish.
-- 
   People say that everyone has a few skeletons in their closet. Not me.
   Well, not yet anyway. I mean, the bodies are still decomposing.
Will Ware		email: wware[at]world[dot]std[dot]com


From lvirden at cas.org  Mon Jan 24 15:22:54 2000
From: lvirden at cas.org (lvirden at cas.org)
Date: 24 Jan 2000 20:22:54 GMT
Subject: Looking for Solaris 2.6/Python 1.5.2 make test fix
References: <0001221518.AA16327@cas.org> <m3k8l1voac.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <86icau$6ah$1@srv38.cas.org>

According to Michael Hudson  <mwh21 at cam.ac.uk>:
:When you cd into Lib/test and run ../../python regrtest.py, do you get
:output along the lines of
:
:test_fcntl

:? If so, what is it saying before it blows up?

Yes, I do get that kind of output.  I get quite a ways thru
the list it appears.  Here's where things give up:
test_zlib
Bus Error(coredump)


:Hmm. Which compiler? I seem to recall tales of regexp's failing due to
:buggy optimizations in some compilers. You might want to try turning
:off optimizations or using something like gcc-2.95.

gcc 2.8.1 

I will have to see if my friendly neighborhood gcc maintainer has nearby
plans to upgrade to 2.95.  Thanks for all your kind help.
-- 
<URL: mailto:lvirden at cas.org> <URL: http://www.usenix.org/events/tcl2k/>
<*> O- <URL: http://www.purl.org/NET/lvirden/>  Tcl2K - Austin, Texas, US
Unless explicitly stated to the contrary, nothing in this posting
should be construed as representing my employer's opinions.


From grndayfan6 at aol.com  Sun Jan  9 10:23:14 2000
From: grndayfan6 at aol.com (GrnDayFan6)
Date: 09 Jan 2000 15:23:14 GMT
Subject: Newbie here, a few questions
References: <m3ln5zux6o.fsf@havenrock.com>
Message-ID: <20000109102314.23968.00000360@ng-fg1.aol.com>

Thanks. As soon as I'm able to get 1.5.2, I'll do so. Thanks again


From aa8vb at yahoo.com  Mon Jan 10 07:20:16 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Mon, 10 Jan 2000 07:20:16 -0500
Subject: First try at Tkinter stuff
In-Reply-To: <38762A22.96C79903@exceptionalminds.com>
References: <3875819D.41ECB8BA@exceptionalminds.com> <m37lhm3uey.fsf@havenrock.com> <38762A22.96C79903@exceptionalminds.com>
Message-ID: <20000110072016.A30427@vislab.epa.gov>

Timothy Grant:
 |I did some more reading last night (Tkinter Life Preserver) and one of
 |the things Matt says is that containers shrink to the size of their
 |contents. Is that always the case, or is there a way around it?

You can always turn off geometry negotiation (see attached).
Also, consider using the "place" manager instead of pack or grid.

Though, I'd really recommend taking a harder look at the pack manager and
using the fill, expand, pad, padx, pady, and anchor options.  For example,
resize the window created by the attached code and notice how badly it
behaves.

Next time you're in the bookstore, browse sections 2.1.1-2.1.7 of
"Effective Tcl/Tk Programming" (Harrison & McLennan).  I find this a useful
Tk reference.

-- 
Randall Hopper
aa8vb at yahoo.com
-------------- next part --------------
#!/usr/bin/env python
#
#  no_resize.py - Show how to disable geometry negoation for a Frame widget.
#

from Tkinter import *

root = Tk()

frame = Frame( root, width=200, height=200, background="blue" )
frame.pack_propagate( 0 )
frame.pack()
            
w = Label( frame, text="Hello, world!", background="yellow" )
w.pack( expand=Y, fill=NONE )

root.mainloop()

From chrism at digicool.com  Sun Jan 30 23:41:33 2000
From: chrism at digicool.com (Chris McDonough)
Date: Mon, 31 Jan 2000 04:41:33 GMT
Subject: Iterative argument conversion in C extensions
Message-ID: <389510DC.6C67B965@digicool.com>

Hi,

I'm working on a C extension that must accept arbitrary positional
arguments of Python types string, int, and float and convert those to
their C equivalents (int, char, float).

I'm sort of stuck, however, on creating a C routine which can accept and
deconstruct into C equivalents the elements of the passed-in Python
tuple that may contain arbitrary Python object types of int, float, and
string arbitrarily positioned within the tuple.  It doesn't help that I
don't know much C, either :-)  

To explain a little better, I'd like to be able to call a function
within a C extension with arbitrary arguments like so:

mymodule.function(1,"string",2.0)

or maybe

mymodule.function("string", "another string", 2.0, 1, "wow")

or maybe even

mymodule.function("lonely string")

I imagine the best thing to do here is to construct a wrapper for the
function that determines the number of passed in arguments and their
types.  The wrapper would then shove what it knows about the arguments
into a dictionary of some kind which then would be shoved into the tuple
that is passed to the C extension module.  But deconstructing that
Python dictionary and retrieving and marshalling its contents within C
is sort of beyond me.

Does anyone have an example of doing such a thing?

Many thanks in advance!

-- 
Chris McDonough
Digital Creations, Inc.
Zope - http://www.zope.org


From darrell at dorb.com  Sat Jan  1 17:01:47 2000
From: darrell at dorb.com (Darrell)
Date: Sat, 1 Jan 2000 17:01:47 -0500
Subject: pickle problem:cant reload object saved with binary option
References: <386e1de1.998928692@news-server> <YHpb4.914$Sr1.104202@typhoon.nyroc.rr.com> <386e31c2.1004019140@news-server> <386E4B0E.C9DD0BF@appliedbiometrics.com> <003001bf5492$ada0faf0$0100a8c0@rochester.rr.com> <87aemp4jzk.fsf@baal.winnegan.de>
Message-ID: <003801bf54a3$ccd09f50$0100a8c0@rochester.rr.com>

Real men go straight to the source :)

--Darrell
----- Original Message -----
From: "Siggy Brentrup" <bsb at winnegan.de>
To: <python-list at python.org>
Sent: Saturday, January 01, 2000 3:25 PM
Subject: Re: pickle problem:cant reload object saved with binary option


> "Darrell" <darrell at dorb.com> writes:
>
> > Out of curiosity I looked into this file mode thing. Sure beats cleaning
out
> > the garage !
> >
> > Looks like the '+' allows read/write.
> > >>> open('t','w').write('now is the time')
> > >>> open('t','r+').write('xx')
> > >>> print open('t','r').read()
> > xxw is the time
> > >>>
> > For append you have to use 'a'
> >
> > /////////////////////////////////
> > gcc\libio\fileops.c function _IO_file_fopen
>
> [...]
>
> No need for UTSL, it's available in the python docs:
>
> http://www.python.org/doc/current/lib/built-in-funcs.html
>
> look for "open"
>
> Thanks
>   Siggy
>
> --
> Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
> ****** ceterum censeo javascriptum esse restrictam *******
>
> --
> http://www.python.org/mailman/listinfo/python-list




From drj at mail.bulkley.net  Thu Jan 27 10:47:46 2000
From: drj at mail.bulkley.net (Peter Schopfer (aka Dr. J. - magician))
Date: Thu, 27 Jan 2000 07:47:46 -0800
Subject: compiling python programs
Message-ID: <389068A2.CD0DFCE0@mail.bulkley.net>

Hello

i read somewere that i could "freez" my python program into a c
executable. how can i do this? could someone show me the process for
doing this?
Thanks



From jbauer at rubic.com  Thu Jan  6 15:34:32 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 06 Jan 2000 14:34:32 -0600
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu>
Message-ID: <3874FC58.69F1D59D@rubic.com>

I have written a Python plugin that works with
Netscape Navigator on Win32.  It used to work with
Microsoft IE, but I think Microsoft has dropped
plugin support for their recent browsers.  Can 
anyone confirm?

We used it for a minimal task -- to print forms on
a laser printer from the browser.

I would really like to see Python running inside
(or along side) Mozilla, but I get the heebie-jeebies
every time I try to do a Mozilla build. <no wink>
As Mozilla becomes more stable, I think it will
be the ideal browser to use for Python client-side
scripting.

As an improvement over Java Applets, you might
consider Castanet, a commercial product from 
Marimba, http://www.marimba.com

Jeff Bauer
Rubicon Research

> Is there a (webbrowser) plugin for executing Python scripts?
> I'm having a lot of problem with Java, and I am seeking 
> alternatives.



From jcollins at pacificnet.net  Thu Jan 27 14:32:17 2000
From: jcollins at pacificnet.net (Jeffery D. Collins)
Date: Thu, 27 Jan 2000 11:32:17 -0800
Subject: Python script monitoring another process
References: <389089E8.32BB7710@cs.brandeis.edu>
Message-ID: <38909D41.2027BA8C@pacificnet.net>

Magnus Bjornsson wrote:

> Hi all,
>
> I'm new to Python so this is probably a typical newbie question...
> Anyway, I'm trying to build a Python script which does the following:
>  1. starts a process (that's easy)
>  2. checks the output from the process (from its stdout)
>  3. when a certain token is found in the process' output the script
>     sends a new token to the process' stdin
>
> So, how can I redirect the process' stdout so I can manipulate it in the
> Python script?  And how can I redirect my scripts stdout to the process'
> stdin?
> Using sys.stdin doesn't seem to do the trick since it doesn't affect the
> environment for the running process...
>
> Any help would be appreciated,
> - magnus e.

Here's a little code snippet...
- Jeff



# module proc.py

import sys
def looper():
    fh = sys.stdout
    while 1:
        fh.write("something\n")
        fh.write("<get input>\n")
        fh.flush()
        x = raw_input()
        fh.write("you entered: " + x + '\n')

if __name__ == '__main__':
    looper()

"""
>From the interpreter, run something like:

jdc:/jcollins% python
iPython 1.5.2+ (#9, Nov 27 1999, 12:41:03)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> import popen2
>>> o,i = popen2.popen2("python proc.py")
>>> o.readline()
'something\012'
>>> o.readline()
'<get input>\012'
>>> i.write('jeff was here\n')
>>> i.flush()
>>> o.readline()
'you entered: jeff was here\012'
>>> o.readline()
'something\012'
>>> o.readline()
'<get input>\012'
>>>
"""




From bob at horvath.com  Tue Jan 18 23:12:57 2000
From: bob at horvath.com (Bob Horvath)
Date: Wed, 19 Jan 2000 04:12:57 +0000
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com>
Message-ID: <388539C9.4812249B@horvath.com>

bjorn wrote:

> It might be easier to use the exchange server as either an IMAP or POP server
> instead (both should be on by default).

Or use fetchmail (at least the configuration utility is written in Python).

>
>
> -- bjorn
>
> Grant Edwards wrote:
>
> > I just found the http server page on Exchange server that
> > allows access to e-mails via a web browser.  What I'd really
> > like is to write a Python program that just sucks the mail out
> > of the inbox and delivers it locally.
> >
> > I've never done any Python code that accesses an http server.
> > Are there any applications which would be good examples?
> >
> > --
> > Grant Edwards                   grante             Yow!  Oh, I get it!! "The
> >                                   at               BEACH goes on," huh,
> >                                visi.com            SONNY??
> > --
> > http://www.python.org/mailman/listinfo/python-list



From wclodius at lanl.gov  Fri Jan 21 17:06:07 2000
From: wclodius at lanl.gov (William B. Clodius)
Date: Fri, 21 Jan 2000 15:06:07 -0700
Subject: In defence of the two-namespace rule
References: <000601bf6452$c123ada0$caa0143f@tim>
Message-ID: <3888D84A.DA230AA0@lanl.gov>


Tim Peters wrote:
> <snip>
> Historical note:  Curry had nothing to do with computers.  He worked in the
> foundations of mathematical logic, specifically the branch called
> "combinatory logic".  This branch showed that the concept of (roughly
> speaking) "variable names" isn't necessary, and, indeed, given all the
> gotchas caused by bound and free vrbl names in formal logic, many complex
> proofs become a heck of a lot more transparent without them.  An offshoot
> showed that this is directly applicable to computational theory as well, and
> leads to an amazingly easy and clean (tho not particularly efficient in its
> naive form) implementation technique for functional languages, one that does
> lazy evaluation and many forms of common subexpression elimination "for
> free, by magic".
> <snip>

Minor historical note: Curry was not the discoverer of currying. It was
first discovered by Frege and was ten independently discovered, and more
widely published by Schoenfinkel. Curry "popularized" the concept of
crrying, but, as I understand it, never used the term himself, generally
citing the previous work of Schoenfinkel.

Somehow schoenfinkeling and freging never caught on as terms, at least
partly because the existing meaning of currying (removing the tangles
from an animal's hair with a comb) is slightly analogous to the removal
of tangles of logic introduced by the concept of currying.


From tim_one at email.msn.com  Fri Jan 14 23:30:43 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 14 Jan 2000 23:30:43 -0500
Subject: Nested while-s
In-Reply-To: <8EBBEE15Ejoshtinamecom@209.208.0.7>
Message-ID: <000301bf5f11$48fa28e0$b92d153f@tim>

[Josh Tompkins]
> Is it not possible to nest while's?

Sure!

> Consider this code:
>
> def listOfRadius(radius, xpos, ypos):
> 	list = []
> 	currentX = (xpos - radius)
> 	currentY = (ypos - radius)
> 	while currentX <= (xpos + radius):
> 	    while currentY <= (ypos + radius):
> 		    str1 = `currentX` + " , " + `currentY`
> 		    list.append(str1)
> 		    currentY = currentY + 1
> 	    currentX = currentX + 1
> 	return list

Note that you never reset currentY to ypos-radius, so the inner loop won't
be entered on any but the first iteration of the outer loop.  You probably
want to change

    currentY = (ypos - radius)
    while currentX <= (xpos + radius):

to

    while currentX <= (xpos + radius):
        currentY = (ypos - radius)

(that is, just swap those two lines).

feel-free-to-kick-yourself-now<wink>-ly y'rs  - tim





From phd at phd.russ.ru  Fri Jan 14 03:56:30 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 14 Jan 2000 08:56:30 +0000 (GMT)
Subject: best Template module
In-Reply-To: <85ljq1$mms$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.21.0001140856120.27804-100000@fep132.fep.ru>

On Thu, 13 Jan 2000, Otis Gospodnetic wrote:
> Could anyone recommend a good 'Template module'?

   ZTemplates. www.zope.org.

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From S.I.Reynolds at cs.bham.ac.uk  Wed Jan 26 13:39:51 2000
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Wed, 26 Jan 2000 18:39:51 +0000
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com>
Message-ID: <388F3F77.3E44@cs.bham.ac.uk>

Its a nice article. Does anyone know where I can learn about frameworks?

I recently started using metaclasses (I think) in Python:

  #I have some existing classes (there may be very many)
  classes = [SomeAgentClass1, SomeAgentClass2, SomeAgentClass3]
  AgentClass = superclasses[1]

  # Dynamically add some extra methods and
  # attributes to the class
  AgentSubClass = makeNewSubClass(AgentClass)

  # Instantiate
  agent = AgentSubClass(x,y,z)

In effect, the magical makeNewSubClass function represents a virtual
class insofar as it describes how to form a whole bunch (class?) of
subclasses without writing them all out.

Frameworks sound like they provide a way of thinking about designs which
have this kind of dynamicism.


Stuart


Edward C. Jones wrote:
> 
> I saw the following on slashdot:
> 
> GameSpy posted a six-page essay of Tim Sweeney's entitled "A
> Critical Look at Programming Languages." Tim Sweeney is, of
> course, the lead coder for the Unreal game engine (one of the
> most licensed 3D game engines for the PC.)
> 
> The article is at http://www.gamespy.com/articles/devweek_b.shtm


From ngps at madcap.dyn.ml.org  Sun Jan  9 09:33:55 2000
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 9 Jan 2000 14:33:55 GMT
Subject: pipe problem on FreeBSD....
References: <200001082039.PAA12903@acer.spvi.com>
Message-ID: <38789c53.0@news.cyberway.com.sg>

According to Steve Spicklemire  <steve at spvi.com>:
> tryit.py:
> [snipped]
> fails on FreeBSD when invoked thusly:
> 
> pluto.spvi.com> python tryit.py | head -3
> (83d0"@eKBdP5!!!!-m4F"e&*2!GP8I3!!!!!!!!!!!GD,D#@R`!!!!!!!!!!!!!
> !!!!!!!!!!!!+Mj!!!!!!!2rrrrp"8&"-8(PdB5%!Y(r$aE4r`mF!#5KQ!"-ic`!
> #f3d!"lAp(RrP2!!!!!!!!,iF"X"9[CqGcE0IZ3fdcPNc0pZq!aYXRD4GeXSrbYF
> Traceback (innermost last):
>   File "tryit.py", line 12, in ?
>     sys.stdout.write(s)
> IOError: [Errno 32] Broken pipe

Works for me on FreeBSD 2.2.5.

/tmp:$ cat 1.py
#!/usr/bin/env python

import sys
f=open('/etc/sendmail.cf')
while 1:
    s=f.read(1024)
    if not s:
        break
    sys.stdout.write(s)
f.close
/tmp:$ ./1.py | head -3
#
# Copyright (c) 1983, 1995 Eric P. Allman
# Copyright (c) 1988, 1993


Is there anything special with your file 'blah'?



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



From jblazi at netsurf.de  Sat Jan 29 04:14:10 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Sat, 29 Jan 2000 10:14:10 +0100
Subject: Cutting off trailing zeroes?
Message-ID: <3892ad81$1_2@goliath.newsfeeds.com>

Id like to format a string like this:
  s = "(%f)" % x

Now if x=1.5 then s="1.50000" or something like this. Is there a tirivial
way of cutting of the zeroes, i.e. getting s="1.5" instead of s="1.50000"?

Of course it can be programmed very simply. My question is: can I do it with
specifiers somehow?

Janos Blazi




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From sholden at bellatlantic.net  Mon Jan 10 12:02:47 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Mon, 10 Jan 2000 17:02:47 GMT
Subject: Zope and NT databases
References: <387956BD.CFDF17DD@bby.com.au> <3879CE9B.BE77787C@bam.com> <3879E4E4.B4BD96FD@ipro.de>
Message-ID: <387A10CE.E8CD408F@bellatlantic.net>

Correct me if I'm wrong, but doesn't Zope get around these
connectivity problems by defining an inherently multi-tier
architecture?  Being new to both Zope and Python I may have
misunderstood, but I'm sure there's documentation on this
[... starts Zope ...] [... waits three minutes for under-
powered server to stop swapping ...] aah, yes.  Visit your
Zope server (assuming you have left the default web intact),
click ont he "Get Started" link, and navigate the tree to
"Database Integration | Multi-Tiered".

It appears you could use a Zope instance on the Windoze machine 
to publish the Access database contents to the UN*X box.

regards
 Steve

Jochen Knuth wrote:
> 
> Hi,
> 
> Bill Scherer wrote:
> >
> > You want ZODBCDA:
> >
> > http://www.zope.org/Products/DA/ZODBCDA
> >
> > Sarah Burke wrote:
> >
> > > Hello...
> > > I want to connnect up to a NT Access 97 Database through Zope running on
> > > Unix. Is the mxODBC module in an External Method the way to go?
> > > Thanks in advance
> > >
> 
> if i remember correctly ZODBCDA only works on Win32 platform.
> 
> also Access 97 is not a client/server database, so you cant connect to
> it from unix directly to it.
> 
> perhaps the ODBC Socket Server from http://odbc.linuxbox.com/index.html
> will work (haven't tried it).
> 
> You have to write a python client for the ODBC Socket Server (if noone
> else has written it, don't know).
> 
> Ciao,
> Jochen
> --
> --------------------------------------------------
> Jochen Knuth          WebMaster http://www.ipro.de
> IPRO GmbH             Phone ++49-7152-93330
> Steinbeisstr. 6       Fax ++49-7152-933340
> 71229 Leonberg        EMail: J.Knuth at ipro.de


From rickhigh at aol.com  Tue Jan  4 05:01:49 2000
From: rickhigh at aol.com (RICKHIGH)
Date: 04 Jan 2000 10:01:49 GMT
Subject: JPython newbie, CLASSPATH question
References: <83q2c5$4ac$1@news.adamastor.ac.za>
Message-ID: <20000104050149.13218.00000287@ng-fk1.aol.com>

No. But you can use the URLClassLoader to load the class. URLClassLoader is in
java.net. You can add a directory as a file URL.

Here is an excerpt that uses URLClassLoader:
...

from java.lang import ClassLoader, Exception
...

class AppletBrowser(JFrame, AppletContext, AppletStub):
...
...

        def createClassLoader(self):                
        
                try:
                                # Get a list of URLS for each jar in the
archive tag.
                                # Get the codebase URL, and it to the list of
URLS.
                                # Create a URL jarray to pass to the
URLClassLoader constructor.
                                # Get the system class loader to pass to the
URLClassLoader constructor.
                                # Create an instance of URLClassLoader passing
the urls and
                                # the system class loader.
                        urls = self.getJars()
                        codebase=self.getCodeBase()
                        urls.append(codebase)
                        urls = array(urls, URL)
                        loader = ClassLoader.getSystemClassLoader()
                        url_loader = URLClassLoader(urls, loader)
                        if debug: print "Created class loader with " + `urls`
                        return url_loader
                        
                except Exception, exception:
                        global error
                        error = exception
                        print "Unable to create class loader"

        def loadApplet(self):

                        #Create the class loader and get the class name.       
        
                class_loader=self.createClassLoader()
                class_name = self.props.getProperty("code")
                
                        # Load the class, create an instance of it, and add it
with addApplet.
                if class_name and class_loader:
                        try:
                                if debug: print "Loading the class: " +
class_name
                                clazz=class_loader.loadClass(class_name)
                                if debug: print clazz
                                applet = clazz.newInstance()
                                self.addApplet(applet)
                        except Exception, exception:
                                global error
                                error = exception
                                print "Unable to load the class " + class_name

...
...

The above is taken from an Applet container that allows an Applet to run as an 
application. (Even on a different box.)
It uses URLClassLoader.

If you want to see more, i.e., if you want me to send the rest-- let me know.

There is a book coming on in March on JPython from publisher Addison Wesley.
You should check it out.

--Rick Hightower


From tiddlerdeja at my-deja.com  Mon Jan 31 15:25:47 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Mon, 31 Jan 2000 20:25:47 GMT
Subject: GetObject instead of CreateObject
References: <000401bf6bd5$2a25a720$3acbd9c2@peridot.optichrome.com>
Message-ID: <874r42$3ei$1@nnrp1.deja.com>

We use win32com.client.Dispatch in python to create COM objects. In VB
you can use CreateObject.

In VB, if you want to use an existing instance of a COM object rather
than creating a new one you call GetObject (I think).

I'd like to know what function we call in Python to mirror this?

Any help greatly appreciated.

Kieran

In article <000401bf6bd5$2a25a720$3acbd9c2 at peridot.optichrome.com>,
  "Adrian Eyre" <a.eyre at optichrome.com> wrote:
> > Can someone tell me what function I should call in Python to
perform a
> > similar task to VB's GetObject?
>
> Maybe it would help (not being a VB programmer) to explain what VB's
> GetObject actually does before looking for an analogue in Python.
>
> --------------------------------------------
> Adrian Eyre <a.eyre at optichrome.com>
> http://www.optichrome.com
>
>


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


From mgm at unpkhswm04.bscc.bls.com  Thu Jan 20 08:46:09 2000
From: mgm at unpkhswm04.bscc.bls.com (Mitchell Morris)
Date: 20 Jan 2000 13:46:09 GMT
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000120001532.29235B-100000@sunse>
Message-ID: <slrn88e4cs.ie6.mgm@unpkhswm04.bscc.bls.com>

In article <Pine.SOL.3.91.1000120001532.29235B-100000 at sunse>,
Evguenii Smogailov wrote:
[snip]
>Yes, i'm saying that 'eq', 'ne' are more readable then '==' '!=', '<>',
>'next' shorter then 'continue' and '..' operator can be a good thing to 
>consider.
>
>evguenii

Except that Perl considers both '==' and 'eq' to be operators, and they
do different things.

% perl
print "cool\n" if 'A' eq 'B';
print "bzzzt\n" if 'A' == 'B';
__END__
bzzzt
% python
if "A" == "B": print "not on your life"
^D
% 

I guess we should be glad you didn't learn lisp first, because then you'd
be asking for 'eql', 'equal', and 'equals' as well as 'eq'.

+Mitchell

P.S. I'm not not not asking for references to your usability studies to
back up the claim that 'eq' is more readable (whatever that means) than
'=='. But I might, if pressed.


-- 
Mitchell Morris

Why is the symbol for anarchy always written the same way?


From tiddlerdeja at my-deja.com  Fri Jan 21 05:02:01 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Fri, 21 Jan 2000 10:02:01 GMT
Subject: Python -> EXE?
References: <2B1262E83448D211AE4B00A0C9D61B03BA706D@MSGEURO1>
Message-ID: <869aqk$mlt$1@nnrp1.deja.com>

Unfortunatly it doesn't. I did try that, then it failed because it
couldn't find pythoncom.dll. I have pythoncom15.dll and I then renamed
that too. I then got my EXE but when I ran it it did nothing. No
errors, no nothing. It was of some size (I can't remember now).

Any further help greatly appreciated.

In article <2B1262E83448D211AE4B00A0C9D61B03BA706D at MSGEURO1>,
  Pieter Claerhout <PClaerhout at CREO.BE> wrote:
> Make a copy of the pywintypes15.dll file, and rename it to
> pywintypes.dll and then it should work just fine.
>
> Pieter
>
> -----Original Message-----
> From: tiddlerdeja at my-deja.com [mailto:tiddlerdeja at my-deja.com]
> Sent: Wednesday, January 19, 2000 8:14 PM
> To: python-list at python.org
> Subject: Re: Python -> EXE?
>
> I tried to use the Standalone.py from this distribution, but I got an
> error.
>
>  found OLEAUT32.dll
>  found python15.dll
>  added filter <TypeFilter {'d': 1}>
> Creating createFruitCategories.exe
>   support\Run.exe -> createFruitCategories.exe
>   C:\TEMP\~535-1 -> createFruitCategories.exe
> Gathering components of dist_createFruitCategories
>  added filter <TypeFilter {'d': 1}>
> Creating dist_createFruitCategories
> Traceback (innermost last):
>   File "Builder.py", line 502, in ?
>     main(opts, args)
>   File "Builder.py", line 484, in main
>     target.build()
>   File "Builder.py", line 87, in build
>     self.assemble()
>   File "Builder.py", line 234, in assemble
>     shutil.copy2(path, self.name)
>   File "d:\Program Files\Python\Lib\shutil.py", line 63, in copy2
>     copyfile(src, dst)
>   File "d:\Program Files\Python\Lib\shutil.py", line 17, in copyfile
>     fsrc = open(src, 'rb')
> IOError: [Errno 2] No such file or directory: 'c:\\winnt\\system32
> \\pywintypes.dll'
>
> I have pywintypes15.dll. Could someone tell me what I'm doing wrong?
>
> Thanks.
>
> In article <002801bf60fa$80b4b440$f29b12c2 at secret.pythonware.com>,
>   "Fredrik Lundh" <fredrik at pythonware.com> wrote:
> > Christian Schaller <Christian.Schaller at mchp.siemens.de> wrote:
> > >   Is there a tool similar to Perl2Exe that transforms a Python
> program
> > > to a MS .EXE programc that can be executed without installing
> Python =
> > on
> > > a machine?
> >
> >     http://www.python.org/doc/FAQ.html#4.28
> >     "How can I create a stand-alone binary from a Python script?"
> >
> > </F>
> >
> > <!-- (the eff-bot guide to) the standard python library:
> > http://www.pythonware.com/people/fredrik/librarybook.htm
> > &mdash; see it live on the python conference! -->
> >
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
> --
> http://www.python.org/mailman/listinfo/python-list
>
>


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


From fredrik at pythonware.com  Tue Jan 11 10:51:57 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 11 Jan 2000 16:51:57 +0100
Subject: Newbie simple question - can't find answer
References: <85fhha$6js$1@nnrp1.deja.com>
Message-ID: <00dd01bf5c4b$ccbaccc0$f29b12c2@secret.pythonware.com>

eleisonme at juno.com wrote:
> I do not seem to be able to use any of the math or cmath functions
> either in a script or on the python command line.
> 
> For example if I type:
> >>>import math
> >>>exp (4)
> 
> OR
> 
> >>>exp(4)
> 
> I get a traceback error:
> NameError: exp (or any other math function that I try to use).

>>> import math
>>> math.exp(4)
54.5981500331

>>> from math import *
>>> exp(4)
54.5981500331

additional info:
http://www.python.org/doc/current/tut/node8.html
http://www.pythonware.com/people/fredrik/fyi/fyi06.htm

</F>




From chrisw at nipltd.com  Thu Jan 27 05:06:02 2000
From: chrisw at nipltd.com (Chris Withers)
Date: Thu, 27 Jan 2000 10:06:02 +0000
Subject: Multidimensional Dictionaries: Thanks!
References: <slrn88ppsg.49t.w6uv@voyager.home.com> <slrn88ue3f.ejp.w6uv@voyager.home.com>
Message-ID: <3890188A.CCC7EC86@nipltd.com>

I'll second that, since I had the same problem! :-)

Chris

Jerry Gardner wrote:
> 
> On 25 Jan 2000 00:00:16 GMT, Jerry Gardner wrote:
> >
> >Are multidimensional dictionaries possible in Python? I'm sure they're
> >possible, but I can't figure out the syntax.
> 
> Thanks to all of you who took the time to answer my question,
> especially Richard Brodie, who's answer was exactly what I was looking
> for.
> 
> --
> Jerry Gardner     | "Bill Clinton has all the steely resolve
> w6uv at hotmail.com  | of a kamikaze pilot on his 37th mission."


From parkw at better.net  Fri Jan 14 22:44:00 2000
From: parkw at better.net (William Park)
Date: Fri, 14 Jan 2000 22:44:00 -0500
Subject: Nested while-s
In-Reply-To: <8EBBEE15Ejoshtinamecom@209.208.0.7>
References: <8EBBEE15Ejoshtinamecom@209.208.0.7>
Message-ID: <20000114224400.A1223@better.net>

On Sat, Jan 15, 2000 at 03:33:07AM +0000, Josh Tompkins wrote:
> Is it not possible to nest while's?
> 
> Consider this code:
> 
> def listOfRadius(radius, xpos, ypos):
> 	list = []
> 	currentX = (xpos - radius)
> 	currentY = (ypos - radius)
> 	while currentX <= (xpos + radius):

	    currentY = (ypos - radius)

> 	    while currentY <= (ypos + radius):
> 		    str1 = `currentX` + " , " + `currentY`
> 		    list.append(str1)
> 		    currentY = currentY + 1
> 	    currentX = currentX + 1
> 	return list
> 
> It will list only the values from the nested while.  It was supposed to 
> increment the inner while until currentY <= (ypos + radius), and then 
> increment the outer while (currentX), and then repeat the inner while's 
> code, etc., etc., etc.  I'm not very good at explaining these things, so I 
> hope you can see what I'm trying to do from the code.
> 
> Any help you can give me would be greatly appreciated.
> 
> TIA,
> 
> Josh "Josue" Tompkins
> ________________________________________________________________
> 
> "Destined For Great Things -- but pacing myself."
> - From a t-shirt.
> 
> E-Mail:  josht at iname.com 
> ICQ:  21219667 
> Web:  http://www.crosswinds.net/~josht
> _________________________________________________________________
> -- 
> http://www.python.org/mailman/listinfo/python-list



From philh at vision25.demon.co.uk  Wed Jan 12 21:47:03 2000
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 13 Jan 00 02:47:03 GMT
Subject: Now What? :o)
References: <n2bd4.64$%Ph.18874529@news.randori.com> <85iglp$qiv$1@news1.xs4all.nl>
Message-ID: <947731623snz@vision25.demon.co.uk>

In article <85iglp$qiv$1 at news1.xs4all.nl>
           boud at rempt.xs4all.nl "Boudewijn Rempt" writes:
> There is the rudiments of a drag-and-drop gui-designer, vipyl, but I
> fear the author has lost heart and quit the project. That might well be
> because coding a Python gui can be really easy, and because drawing a
> gui doesn't make much sense with layout managers anyway.

Python also has the rudiments of a text-based GUI designer (Parrot
at http://www.comuno.com/linux/parrot/intro.html). I haven't done
much work on it recently, as I've been busy with other things; 
probably I will do more work on it this month.

-- 
Phil Hunt - - - phil at comuno.com

 See Comuno's Linux coverage at <http://www.comuno.com/linux/> for:
* Links to useful sites * Linux-related news * Open source software *



From News at Titanic.co.uk  Sun Jan  2 05:15:16 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Sun, 02 Jan 2000 10:15:16 GMT
Subject: DB API
Message-ID: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com>

Hi All,

Where can I find the instructions for the DB API?

I have managed to figure out the .connect() and .fetchone() methods
but am flagging with the rest.

The Python Database API Specification 2.0 document does not realy show
me the methods or examples that I need to get started and the MySQLdb
documentation is deliberatly vague to discourage use of the _mysql.
methods (so you use the proper API?)

Apart from that I have managed more in Python this holiday than I have
in Perl in the last 3 months!

Keep up the good work...


TIA

Simon
Simon Faulkner


From g_will at cyberus.ca  Fri Jan 21 13:09:35 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: Fri, 21 Jan 2000 18:09:35 GMT
Subject: comp.lang.python.announce working??
Message-ID: <01bf643a$5039a8a0$3e41c3d1@Cyberus.cyberus.ca>

Hi All,

I havent had any news feed from comp.lang.python.announce since the end of
December.  I contacted my ISP and the news came back from the group that
supplies the news feed that there havent been any new postings to this
group since this time.  I find that strange.

Is anyone else able to access this news group?

thanks,

Gordon Williams


From bsb at winnegan.de  Tue Jan 11 13:22:17 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 11 Jan 2000 19:22:17 +0100
Subject: [ANN] Multimethod.py -- multimethods for Python
In-Reply-To: Randall Hopper's message of "Tue, 11 Jan 2000 11:49:24 -0500"
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov> <20000111114923.B42270@vislab.epa.gov>
Message-ID: <87zouclb8m.fsf@baal.winnegan.de>

Randall Hopper <aa8vb at yahoo.com> writes:

> Randall Hopper:
>  |I had the same question yesterday and did a little surfing.  Basically,
>  |instead of switching off just the type of 'self' to determine which method
>  |to call, you switch off the type of other arguments as well.
>  |
>  |If you think best in implementation as I do:
>  |
>  |   def f( a,b ):
>  |     if   type(a) == Type1 and type(b) == Type1: f_t1_t1(a,b)
>  |     elif type(a) == Type1 and type(b) == Type2: f_t1_t2(a,b)
>  |     elif type(a) == Type1 and type(b) == Type3: f_t1_t3(a,b)
> 
> To be closer to the truth, I really should have said:
> 
>        if   issubclass( a, Type1 ) and issubclass( b, Type1 ): f_t1_t1(a,b)
>        elif issubclass( a, Type1 ) and issubclass( b, Type2 ): f_t1_t2(a,b)
>        elif issubclass( a, Type1 ) and issubclass( b, Type3 ): f_t1_t3(a,b)
> 
> This alludes to one way multimethods can get into method resolution
> troubles.  If none of the multimethods' parameter signatures match the
> argument types exactly, but multiple multimethods accepting compatible base
> class objects as parameters exist, ...

Can you point us to a formal definition? Seems like you initially got into
trouble explaining it by example :)

>From what I see above it might as well be some dynamic variant of
C++'s template functions.

Thanks
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
                  bsb at north.de - http://www.north.de/~bsb/
****** ceterum censeo javascriptum esse restrictam *******



From wlfraed at ix.netcom.com  Tue Jan 25 12:27:51 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Tue, 25 Jan 2000 09:27:51 -0800
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <948807929.472741@news.adtranzsig.de> <Thjj4.6690$S64.195053@nnrp1.uunet.ca>
Message-ID: <uduNOEQ5UiHWtRwJuOY1WhUWSYbc@4ax.com>

On Tue, 25 Jan 2000 10:23:32 -0500, "Warren Postma"
<embed at geocities.com> declaimed the following in comp.lang.python:

> 
> Rather bloody minded of me to point this out, but hey "to each his own"...
> 
> #define BEGIN {
> #define END     }
> 
> Can I do the same somehow in Python?  ;-)
> 
	As has been pointed out once or twice...

	if x:
	#begin
		y()
	#end

	The difference being that you still need to indent between
them..

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From pbible at home.com  Wed Jan  5 23:46:37 2000
From: pbible at home.com (Paul E. Bible)
Date: Thu, 06 Jan 2000 04:46:37 GMT
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <N6Vc4.610$hJ1.7143@news.rdc1.tx.home.com>

Interland.com supports Python on a Unix server (but not on Windows/NT).

Paul

"Chuck Esterbrook" <echuck at usit.net> wrote in message
news:3873EE03.1CE5F8CC at usit.net...
> Seems like many web host providers, such as MindSpring, offer CGI with
> Perl, but not other languages, notably Python.
>
> Does anyone know who would host a CGI Python site?
>
> -Chuck
>
>




From gerrit.holl at pobox.com  Thu Jan 20 15:47:46 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 20 Jan 2000 21:47:46 +0100
Subject: stupid question !
In-Reply-To: <867kic$kb1$1@eng-ser1.erg.cuhk.edu.hk>; from s986001@mailserv.cuhk.edu.hk on Fri, Jan 21, 2000 at 02:41:28AM +0800
References: <867kic$kb1$1@eng-ser1.erg.cuhk.edu.hk>
Message-ID: <20000120214746.A6313@stopcontact.palga.uucp>

Kenny wrote on 948418888:
> Dear all,
>      I am now trying to use python for CGI programming!
> But I don't know how to wake up the python programming by using HTML!
> 
>         <form method=post action=" ....path to the python source code....">
>       It seems to me that the program is not be activated!

Are you sure the path is an url, and not a local path on the machine?
What does the webserver return to the browser? What does the access
and the error logs say about it? Does your program print
"Content-Type: text/html\n" above the HTML output?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From helge at mdlink.de  Thu Jan  6 17:20:30 2000
From: helge at mdlink.de (Helge Hess)
Date: Thu, 06 Jan 2000 23:20:30 +0100
Subject: no return values for __init__ ??
References: <1264926603-6720782@hypernet.com>
Message-ID: <3875152E.98598E4E@mdlink.de>

Gordon McMillan wrote:
> >   def __init__(self...
> >      s = super.__init__(self...
> 
> OK, but it introduces a "gotcha". As in, what happened to the
> stuff I set up before calling super.__init__, (which is not only
> valid, but often useful).

I don't know what you mean with that. It is deallocated if another value
is returned and available to super.__init__ as usual. super.__init__ can
base it's decision about the value to return on state already set up in
self as well.
I cannot see a problem here.

> > BTW: I also always wondered why Python has None as the default
> > return value. 'self' seems better to me as one could make
> > compound method calls without additional work while None has no
> > benefit I can see (maybe better RC behaviour ?).
> 
> It's consistent with function behavior (which has no self to
> return), and after all, a method (when viewed on the class
> object) is a function.

Yes, that's seems understandable to me. Maybe it would have been cool if
functions would have been methods on their namespace/module :-)

> With yet more "gotchas".  The (one and only) base class
> constructor that could change "self" would have to be the first
> one executed.

No, why so ? Do you care about partial state in __del__ ?? The one who
uses such a setup is also responsible for resolving such a setup (and
this is easily possible).
Again that is similiar with any other method and MI.

> > There is no conceptual difference between __init__ returning a
> > value and any other method returning a value.
> 
> Not if you don't accept the axiom that __init__ is "just another
> method". In fact, as an initializer, __init__ is a lot less special
> that constructors are in most languages.

Well, in Objective-C and Smalltalk there isn't a thing like a
'constructor'. In Python and Java 'constructor' IMHO only means 'method
with special syntax for invocation' since such methods are not
participating in allocation.

Greetings
  Helge



From aerogems at netins.net  Mon Jan 17 13:01:22 2000
From: aerogems at netins.net (Scott Billings)
Date: Mon, 17 Jan 2000 12:01:22 -0600
Subject: Read In Directory Contents
Message-ID: <388358F1.F1771D82@netins.net>

For a program I'm working on, it would be very useful if I could use an
expression such as:

list = os.getcwd() + "*.txt"

to have the names of all text files in the current dir placed into the
list. However, when I try this method, I get an error about no such file
existing.

So, how would I go about doing this?

-Scott Billings-



From tratt at dcs.kcl.ac.uk  Wed Jan 26 09:10:39 2000
From: tratt at dcs.kcl.ac.uk (Laurence Tratt)
Date: Wed, 26 Jan 2000 14:10:39 +0000
Subject: Wanted: A python icon for X11
References: <pcosnzl4bpf.fsf@math.ntnu.no>
Message-ID: <388F005F.D5A0BE94@dcs.kcl.ac.uk>

Harald Hanche-Olsen wrote:

> Would anybody perhaps have some pixmap lying about that might be (or
> become) a suitable icon for IDLE windows under X11?  A picture of a
> python (the snake) perhaps, or of a can of spam, or something like
> that?  Right now, I only have the generic X icon, and that is of
> course intolerable.

When I ported Python to RISC OS, I got Mark Hammonds permission and pinched
the bitmap from the Windows port and did a bit of editing to get it to suit
the necessary dimensional requirements. It's really rather nice, and at some
point I did convert it to a gif so I could use it with WindowMaker under
X... I can't remember what I did with it, but it does work and fit's in OK.
It would be good to see a cross platform Python icon and I reckon this is as
good a choice as any :)

> a-portrait-of-John-Cleese-might-also-do-nicely-ly y'rs

in-or-out-of-the-psychiatrists-couch?-ly y'rs - Laurie
-- 
http://eh.org/~laurie/comp/acorn/ropython/ for RISC OS Python


From invalid.address at do.not.email  Tue Jan  4 12:37:50 2000
From: invalid.address at do.not.email (guppy)
Date: Tue, 04 Jan 2000 17:37:50 GMT
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com>
Message-ID: <386fa6b1.58568155@news.telus.net>

Have you looked at the wxPython interface to wxWindows?  It's
cross-platform compatible.  It'd be interesting to hear if it's quick
enough for you.  Certainly at the rate its development is proceeding, it's
set to become the new GUI standard for Python (thank gods!) (or, rather,
thank Robin!)

On Sat, 01 Jan 2000 21:40:47 GMT, randy_shaffer at my-deja.com wrote:
>I have experimented with tkinter and the PMW library.  Things seem to
>operate a bit slow.  The other options/considerations I?ve pondered are:
>
>Move the gui to the Win platform ? I?m not very proficient in MFC-land
>and I?m worried.  Is there Python portability to Windows
>problems/issues as well?
>
>Can Viper help out here?  I get the feeling that it is not quite ready
>for prime time.  Am I wrong?
>
>What about JPython and the fabulous Java Swing library?  Am I asking
>for trouble here?



From wweston at worldnet.att.net  Thu Jan 13 15:06:09 2000
From: wweston at worldnet.att.net (wes weston)
Date: Thu, 13 Jan 2000 20:06:09 +0000
Subject: can't find tk.tcl
References: <387BB7D8.F1888924@worldnet.att.net> <387D4966.7BC156E6@charm.net>
Message-ID: <387E3030.F96BBE2D@worldnet.att.net>

John,,
    I had the same problems installing python you did. The way I got to a fix was by
doing a find of *.dll on my tcl root - yielding a list of about six files. More than
one of these was also in c:\winnt\system32. The one causing the problem was tk80.dll.
I deleted it from sys and bingo; IDLE and Tkinter work.
        wes

John Zouck wrote:

> I had a problem that sounds like this several weeks ago, and finally found that
> there were some tcl/tk .dll files in /windows/system that , when deleted, made
> things work right. Before I found this, I tried every other kind of clean-up and
> reinstall of python to no avail. I had installed tcl/tk months (or years) before,
> which probably caused these files to be there. I don't know why it worked, but it
> did, and this is just a suggestion.
>
> wes weston wrote:
>
> > Guys,
> >         At the command line prompt: "from Tkinter import *" then "w = Tk()" and
> > a "usable" tk.tcl can't be found in the several places looked at. But,
> > it's in one of them: C:\Progra~1\Tcl\lib\tk8.0.  The weirdest thing is
> > I've been running programs that use Tkinter recently on this
> > machine(WinNT Ws). It (Tkinter programs) works at work and on my
> > portable.
> >
> >         I tried a fresh install in several different places. Help!
> >
> >                 wes
>
> --
> ===================
> John H. Zouck
> jzouck at charm.net
> ===================



From dworkin at ccs.neu.edu  Sun Jan 23 22:05:54 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 23 Jan 2000 22:05:54 -0500
Subject: SUGGEST: os.getuser()
References: <20000116174004.A5995@stopcontact.palga.uucp> <000b01bf6049$36532140$f29b12c2@secret.pythonware.com> <20000116195710.A7700@stopcontact.palga.uucp> <Jylh4.3912$Ef6.1036319@news.shore.net> <20000121170432.B843@stopcontact.palga.uucp>
Message-ID: <vndr9f8w4ml.fsf@betelgeuse.ccs.neu.edu>

Gerrit Holl <gerrit.holl at pobox.com> writes:

> a-sysop-letting-multiple-identical-uids-be-in-/etc/passwd-
>     -is-very-paranoid-anyway-ly yours,

On many systems, especially ones with a BSD heritage, this sort of
things is quite common:

root:*:0:0:Charlie &:/root:/bin/sh
toor:*:0:0:Bourne-again Superuser:/root:

If you are careful with how you use and allocate them, multiple
accounts with the same UID can make sense and work just fine.

a-sysadmin-that-isn't-very-paranoid-probably-isn't-a-very-good-one-anyway'ly,

-Justin

 




From tjg at avalongroup.net  Mon Jan 17 17:10:22 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Mon, 17 Jan 2000 14:10:22 -0800
Subject: Tkinter Tab orders and keybindings
References: <3883775C.9369120@exceptionalminds.com> <m3puv0todr.fsf@havenrock.com>
Message-ID: <3883934E.65D15CF1@exceptionalminds.com>

Matt Gushee wrote:

> If that doesn't work for you, you might have to write your own 'next'
> and 'prev' methods, which would use your own list of widgets. If you
> don't need the Tab key for anything else, you could use the following
> bindings in your toplevel window.
> 
> self.bind_all('<Tab>', self.next)
> self.bind_all('<ISO-Left-Tab>', self.prev)

Huge help, thanks Matt, much appreciated.
 

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From loewis at informatik.hu-berlin.de  Wed Jan 12 15:28:09 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 12 Jan 2000 21:28:09 +0100
Subject: Q: Network access & Python
References: <387C68BF.EA30D859@mchp.siemens.de>
Message-ID: <p6q4scjhw6e.fsf@informatik.hu-berlin.de>

Christian Schaller <Christian.Schaller at mchp.siemens.de> writes:

>   Does anyone know if there's a module in Python for getting network
> stats?  I want to write a tool for displaying all used ports and
> services running on them and also if there's a connection on a port,
> from where (ip:port) it originates, but found no function/module
> containing functions for listing/analyzing ports.

How to obtain such information highly depends on the operating system
you are using; I don't know any module covering this for a variety of
operating systems.

For example, if you are using Linux, you can easily obtain this
information from the /proc filesystem; reading /proc/net/tcp will give
you all open sockets.

Regards,
Martin


From smigowsky at dspace.de  Tue Jan 18 03:54:54 2000
From: smigowsky at dspace.de (Stefan Migowsky)
Date: Tue, 18 Jan 2000 09:54:54 +0100
Subject: PythonWin and Dialogs in DLLs
Message-ID: <326536345498D311B3BC00105A39802A07457F@newsintern.dspace.de>

Hi,

Ken Seehof has programmed a Python Resource Wizard

http://starship.python.net/crew/seehof/PyResWizard.html

With this wizard you could design your dialog in VC++ compile
it to a resource DLL and have all the necessary python 
files generated.

Stefan

>-----Original Message-----
>From: Robert [mailto:soundhack at hotmail.com]
>Sent: Monday, January 17, 2000 10:56 PM
>To: python-list at python.org
>Subject: PythonWin and Dialogs in DLLs
>
>
>Hi all,
>    After getting bored creating dialogs by making dialog templates
>programmatically,
>I would instead like to use VC++'s dialog editor to format the dialog.
>Looking at the dialog.Dialog class in PythonWin, apparently
>it accepts a DLL filename to get resource information.
>
>   My question is this: How can I make this DLL that contains
>only the dialog resource? Using the appwizards to make a DLL
>project, I either get a big DLL file (which works, but it has 
>DLLMain() and
>run time libraries linked in that i dont really need) or an 
>error linking,
>saying there is an unresolved external
>reference _DllMainCRTStartup. The ideal situation would be just
>to save the resource file and have pythonwin read that, but if i
>have to use DLLs, I would like to keep unnecessary runtime 
>stuff out of it.
>
>Thanks in advance,
>Robert
>
>
>-- 
>http://www.python.org/mailman/listinfo/python-list
>



From grant at nowhere.  Wed Jan 19 17:30:37 2000
From: grant at nowhere. (Grant Edwards)
Date: Wed, 19 Jan 2000 22:30:37 GMT
Subject: olemsg32 module used in win32com exchange server example?
References: <slrn8897gs.3cm.grant@grante.comtrol.com> <864ult$fot$1@nnrp1.deja.com>
Message-ID: <slrn88ceoc.5ga.grant@grante.comtrol.com>

>>  2) Does the win32com stuff talk directly to the Exchange
>>     server, or does it go through a local copy of Outlook?
>
>It goes through the MAPI DLLs - the same DLLs that the local copy of
>Outlook itself uses.

Cool.  Does that mean that I don't even need to have Outlook
installed in order for a Python/COM/MAPI program to work?

-- 
Grant Edwards                   grante             Yow!  Yow! I forgot my
                                  at               PAIL!!
                               visi.com            


From tim_one at email.msn.com  Sun Jan 30 18:25:41 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 30 Jan 2000 18:25:41 -0500
Subject: [Newbie Problems] AttributeError from CGI script?
In-Reply-To: <20000130.14431716@mis.configured.host>
Message-ID: <000501bf6b79$52a62be0$482d153f@tim>

[David Fisher]
> ...
> Along with most/all of the other functions in string.py, it is
> replaced at the end of the module by another function named replace
> from the strop module, which is written in C.  Now why this is done
> is a complete mystery to me.

Hysterical raisins:  originally, string.py was "it", and there was no strop
module.  What's there today was a way to replace the important string
functions w/ much faster C versions, without breaking any old code.  The
string.py functions also serve as a kind of reference implementation for the
curious, or for an initial port to a weird machine.

those-who-don't-remember-the-past-are-doomed-to-
    be-confused-by-its-ripples-ly y'rs  - tim





From gte838h at prism.gatech.edu  Mon Jan 17 11:18:01 2000
From: gte838h at prism.gatech.edu (Shaun Allen Dishman)
Date: 17 Jan 2000 16:18:01 GMT
Subject: Global help
Message-ID: <85vfbp$f3t@catapult.gatech.edu>

Hello,

I have something like the following situation:

----- x.py -----

from y import *

def x(filename):
	file = open(filename, 'w')
	y()

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

----- y.py -----

def y():
	# must be the same file object as in x.py
	file.write('blah blah blah \n')	

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

I have fiddled around with the global settings for the file object but
cannot seem to get things to work.  All I need is to be able to access the
file object from one module inside of another one, without passing it as a
parameter.  Is this even possible?  Thanks in advance.

Shaun


From tanming_fong at phoenix.com  Tue Jan  4 11:54:16 2000
From: tanming_fong at phoenix.com (TM)
Date: 4 Jan 2000 10:54:16 -0600
Subject: Error in inserting blob into ORACLE using ODBC.Windows
References: <386c0577.107657473@news.phoenix.com> <386CD167.8525C821@roguewave.com> <84ijb0$u46$1@nntp8.atl.mindspring.net>
Message-ID: <38722065.507767100@news.phoenix.com>

Thanx for the leads. I still haven't got a clue of what went wrong, but
here are findings:

Oracle docs says,
"ORA-01461: can bind a LONG value only for insert into a LONG column" - 
Cause: An attempt was made to insert a value from a LONG datatype into
another datatype. This is not allowed. 
Action: Do not try to insert LONG datatypes into other types of columns

The mxODBC test.py run into this problem. i went panic yesterday because of
a looming deadline and threw in some VB(guilty as charged) +ASP ADO code,
the Oracle ODBC driver gave me the same ORA-01461 complaint, and the
MS ODBC driver just got all confused. 

Oracle docs also says,
"ORA-01465: invalid hex numbe" - 
Cause: In an UPDATE statement following a SELECT FOR UPDATE, part of the
ROWID contains invalid characters. ROWID must be expressed in the proper
and expected format for ROWID and within quotes. 
Action: Enter the ROWID just as it was returned in the SELECT FOR UPDATE. 

This is the error I get with the python code. Later on I did try to open
the file in 'rb' mode as bjorn suggested, but still the same thing. BTW,
"print conn.bindmethod" returns a 1, so i must be using BIND_USING_SQLTYPE,
which is what i want, i hope.

Since i got stuck with mxODBC+python  then ADO and my  deadline is two
weeks away, i will surrender and and redirect that part of the web page to
some pieces of Java JDBC code which one my colleagues claimed he saw
working with Oracle BLOB. 

thanks for all the help,

Tanming Fong





From meetpankaj at MailAndNews.com  Mon Jan  3 23:47:40 2000
From: meetpankaj at MailAndNews.com (Pankaj Tolani)
Date: Mon, 3 Jan 2000 23:47:40 -0500
Subject: DB API
Message-ID: <3873F847@MailAndNews.com>

Hi All,

I am new to Python Programming and to the newsgroup as well. I know there is 
a 
DB API in python. But I want to know a few more details 
Is the API implemented using ODBC ? If yes, is there some API implemented 
with 
direct connecttion to the DB ?

Thanks in advance, Pankaj

------------------------------------------------------------
URL : http://pankaj.india.webjump.com
email id : meetpankaj at mailandnews.com
------------------------------------------------------------



From thomas at bibsyst.no  Tue Jan 25 08:05:57 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Tue, 25 Jan 2000 14:05:57 +0100
Subject: Embedding Python in C-code??
Message-ID: <388D9FB5.7389D8E1@bibsyst.no>

Hi,

Just wondered, and it`s probably stupid, if it`s possible to use python
code/objects/modules from a C++/C-program. I mean, Glade and alot of
other GUI-builders are aimed at C/C++ development. If we could somehow
use Python as foundation and just access that from a gui built using
C/C++ it would be cool.

Thomas


From s.schwarzer at ndh.net  Wed Jan 19 20:34:27 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 20 Jan 2000 02:34:27 +0100
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com>
Message-ID: <38866622.95353679@ndh.net>

Hello tye4

(Do you have a real name or did I miss it? If yes, sorry.)

tye4 schrieb:
> Actually, I use something like:
> 
> if x == 10:
>     print 'x is 10'
> #end if
> print 'Hello world'
> 
> I'm more concerned with how many maintenance headaches will result if I were
> to accidently add or delete a space on a line.

Like others I think your problem is not the language but your editing
tool(s). For the code I write alone, I always use spaces (including
such that are inserted by the tab key). One time I had problems when
I edited code from someone else and ran into those troubles you see.
However, after searching the cause for the error for a while, I
selected  :set list  in Vim and saw clearly where tab characters were
and changed them to the equivalent number spaces (for this code).
>From then, I never had such problems, - because I am aware of them
now!

> Also, I would to be very
> careful and choosy about what editor to use.. Some editors expand tabs while
> others don't.. What indentation you see in one editor is sometimes different
> to what is seen in another...

In any good editor (TM) ;-) you can choose whether to insert tab
characters or not. And as said above, it is a much lesser problem _if
you know about its possibility_.

> How hard can it be to add 'end' indicators to the language.. If these end
> indicators are added, the language would still be backward compatible. I
> can't believe nobody has fixed this for so long.

I don't think that this should be "fixed". I've become used to Python
rather quick and I like it very much. Concerning the design of the
language I think there are _many_ things in Python that make
"maintenance headaches" _go away_ (among them the use of indentation
for blocks).

I don't say that Python is completely free of design errors but I found
that Python is so well-designed that I write code much faster, more
elegant (IMHO) and with less errors than with any other languages
that I used before (e. g. Pascal, C++ and Perl).

alt.religion.python-ly y'rs ;-)
 Stefan


From huaiyu_zhu at moon.localdomain  Thu Jan 13 18:52:52 2000
From: huaiyu_zhu at moon.localdomain (Huaiyu Zhu)
Date: 13 Jan 2000 17:52:52 -0600
Subject: A Python bug? (core dump)
Message-ID: <slrn87spbq.16i.huaiyu_zhu@localhost.localdomain>

The following script dumps core when the "print self" line is uncommented,
because print calls __str__ which calls print which calls ...

##################################
class Item:
    def __init__(self, item):
        self._ = item
	    
    def __str__(self):
#       print self
        return `self._`

print Item ('aaa')
##################################

For some other recursive calls I tried that do not involve print and
__str__, the outcomes are sane:

RuntimeError: Maximum recursion depth exceeded

I've read somewhere that "Python never dumps core".  So this must be a bug.
The versions are:

Python 1.5.2 (#1, Sep 17 1999, 20:15:36)  [GCC egcs-2.91.66 19990314/Linux
(egcs- on linux-i386

-- 
Huaiyu Zhu		huaiyu_zhu at yahoo.com


From grant at nowhere.  Tue Jan  4 14:44:32 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 04 Jan 2000 19:44:32 GMT
Subject: Grail
References: <387237F3.C35B7FA1@callware.com> <14450.17021.111722.710187@weyr.cnri.reston.va.us> <3872441A.7E86CCED@callware.com>
Message-ID: <slrn874jd0.l0l.grant@grante.comtrol.com>

In article <3872441A.7E86CCED at callware.com>, Ivan Van Laningham wrote:

>Um, yeah, I was hoping to get a Tk HTML widget, since the entire rest of
>the application is already written in Tk.  An HTML widget would let me
>use the existing HTML file for help without having to fire up netscape
>or IE. ...

STk has a module/widget/thingy that uses a Tk canvas widget
to display HTML.  Don't know how much of a pain it would be to
translate it into Python.

-- 
Grant Edwards                   grante             Yow!  Alright,
                                  at               you!! Imitate a WOUNDED
                               visi.com            SEAL pleading for a PARKING
                                                   SPACE!!


From aa8vb at yahoo.com  Tue Jan 18 12:31:43 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 18 Jan 2000 12:31:43 -0500
Subject: re-opening stdin in raw (binary) mode?
In-Reply-To: <874scbuzw2.fsf@fun77.kivikko.hoas.net>
References: <388388c9.12001026@news.iol.ie> <slrn8876la.bos.kc5tja@garnet.armored.net> <874scbuzw2.fsf@fun77.kivikko.hoas.net>
Message-ID: <20000118123143.A216176@vislab.epa.gov>

Erno Kuusela:
 |>>>>> "Samuel" == Samuel A Falvo <kc5tja at garnet.armored.net> writes:
 |
 |    Samuel> under Linux, you can perform an ioctl() on file descriptor
 |    Samuel> 1 (or is it 0?) to set a binary-mode flag.
 |
 |There is no binary mode in unix/linux, it's a windoze (and mac?) thing.

Sure there is.  It just happens to be a no-op on UNIX. :-)

Yeah, for MSWin/DOS in C, you need something like:

    _setmode(fileno(stdin),O_BINARY);

if you want to keep the current stdin stream (if you want to reopen stdin
as binary to some other file/device, just freopen it).

A quick search of the Python newsgroup archive for _setmode and a grep of
the Python tree turns up the likely answer:

------------------------------------------------------------------------------
1)  Invoke python with the -u option to kick in a BINARY _setmode under the
    hood for both stdin and stdout.
------------------------------------------------------------------------------

Now, having it be unbuffered too may kill your performance (...maybe
there's a Pythonic way to setvbuf an open stream in Python code).

Also, there's also a grep hit in the msvcrt MSWin module.  Check it out.
In particular:

------------------------------------------------------------------------------
2)  http://www.python.org/doc/current/lib/msvcrt-files.html

    setmode (fd, flags) - Set the line-end translation mode for the file
    descriptor fd. To set it to text mode, flags should be os.O_TEXT; for
    binary, it should be os.O_BINARY.
------------------------------------------------------------------------------

Randall



From skip at mojam.com  Tue Jan 11 18:21:15 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 11 Jan 2000 17:21:15 -0600 (CST)
Subject: strange CGI problem
In-Reply-To: <85ga10$p8u$1@nnrp1.deja.com>
References: <85ga10$p8u$1@nnrp1.deja.com>
Message-ID: <14459.47851.670815.455938@beluga.mojam.com>

    >> I'm creating a little web-site; for some reason when I view the page
    >> in Internet Explorer, it's fine, but Netscape shows me the raw html
    >> as if it's a text file.  What would cause this?

Sounds like IE is looking at the contents and saying, "hmmm... looks like
HTML", and displaying it.  Make sure your CGI script is spitting out an
appropriate Content-type header.




From mgushee at havenrock.com  Wed Jan  5 12:31:43 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 05 Jan 2000 12:31:43 -0500
Subject: Seeking feedback on parsing method.
Message-ID: <m3puvgpgqo.fsf@havenrock.com>

Greetings, everybody--

I've written a class for handling CSS stylesheets. The whole module is
included at the end of this message, and you're welcome to use it,
hack it, comment on it, whatever. It has a couple of features that I
think are kinda cool (yes, they're documented). But I'd particularly
like to know what you all think of my parsing method.

I wrote the parser from scratch rather than using an external parsing
library because I wanted something that could be easily included in a
package for public distribution and wouldn't require the user to
download extra stuff. However, I don't have a whole lot of experience
with parsers, so there's probably a lot of room for improvement. I'm
also not positive I've accounted for all features of CSS
syntax. Anyway, here's the parsing code:

    ## These regexps are defined in the class namespace

    ## capture a specification group -- e.g., a group of style
    ## properties pertaining to a particular context
    specgroup = re.compile(r"([^{}]+){([^{}]+)}")
    ## weed out supposed property definitions that are empty
    ## or missing the required colon.
    bogus = re.compile(r"^[^:]*$")

    def _parse(self, cssdata):
        propdict = {}
        while cssdata:
            spec = self.specgroup.match(cssdata)
            if spec is None:
                break
            selectors, stylegroup = spec.groups()
            ## Parse the styles first. That way, if there are
            ## multiple identifiers, we don't have to redo
            ## the styles for each one.
            for style in string.split(stylegroup, ';'):
                if self.bogus.match(style):
                    break
                style = string.split(style, ':')
                prop = string.strip(style[0])
                val = string.split(style[1], ',')
                if len(val) < 2:
                    propdict[prop] = string.strip(val[0])
                else:
                    propdict[prop] = []
                    for item in val:
                        propdict[prop].append(string.strip(item))
            for sel in string.split(selectors, ','):
                sel = string.strip(sel)
                ## Hierarchical selector, e.g. .Body P
                hiersel = string.split(sel)
                if len(hiersel) > 1:
                    sel = tuple(hiersel)
                if not self.has_key(sel):
                    self[sel] = {}
                for prop, val in propdict.items():
                    if not self[sel].has_key(prop):
                        self[sel][prop] = val
            cssdata = cssdata[spec.end():]


And the module:

------------- cut below this line -----------------------------------
#! /usr/bin/env  python

"""

css.py
-----------

Provides a CSS class, which reads in a css specification in
file or string form and wraps it in a dictionary-like object.
See the class documentation for more info.

Copyright 2000 by Matthew C. Gushee

The author hereby grants permission to use, redistribute, or
modify this program for any purpose, on condition that the
above copyright notice and this statement be retained without
alteration. This program carries no warranty of any kind.

Please send comments and questions to <mgushee at havenrock.com>.

"""

import string, re, copy, os
from UserDict import UserDict

class CSS(UserDict):
    """

    A dictionary-like wrapper for CSS style specifications.

    
    Usage:

        mycss = CSS('./mystyle.css')
        mycss = CSS(css_data)

    Instances may be created based on CSS files, or on data
    in CSS or Python dictionary form. See __init__ method
    documentation for details.


    Style data is stored as a dictionary in the form:

        {<selector>: {<prop-name>: [<prop-value>, ...], ...}, ...} 
    
    Each key of the dictionary is a single CSS selector, which may
    be either an identifier in string form (i.e., an element type
    such as 'h1,' 'table,' etc., a class identifier such as '.Menu',
    or a unique ID such as '#my-dog-photo') or a tuple of such
    identifiers. The value of each item is also a dictionary, its
    keys being things like 'font-family', 'margin-l', and so on;
    and its values being either strings or lists of strings.

    Note that if a stylesheet contains a group of properties with
    multiple selectors, the CSS instance will have one key for each
    of these selectors. E.g.,

    .MainMenu, .WhatsNew, H5 { spam: eggs }

    ... becomes

    {'.MainMenu': {'spam': 'eggs'}, '.WhatsNew', {'spam': 'eggs'},
    H5 {'spam': 'eggs'}}

    Hierarchical selectors, on the other hand, are parsed into tuples.
    For example:

    .BodyText A { spam: eggs }

    ... becomes

    {('.BodyText', 'A'): {'spam': 'eggs'}}

    In this documentation, I use the term 'context' to refer to a
    single or hierarchical selector *in the form* used by the CSS
    class methods -- a string like 'BODY', or a sequence like
    ('TABLE', 'A'), or ['.Menu', 'UL']. Although the methods are
    currently written to accept context arguments in list form, I
    may at some point decide to restrict them to tuples.


    Public methods:

    getspec    Return style data for a particular context.

    setspec    Define style data for a particular context.

    merge      Assimilate data from another CSS instance.

    remove     Remove specified data from the current instance.


    Additionally, the CSS class overloads the + and - operators to
    provide a convenient means of adding and subtracting data, e.g.:

        this_style = this_style + that_style

    (that_style may be a CSS class instance or a dictionary in the
    same form as CSS.data) If the two objects have any keys in common,
    the values in the right-hand object take precedence. Thus, the
    above operation may overwrite data in this_style. If you want to
    preserve all existing data while adding new data, simply reverse
    the order:
    
        this_style = that_style + this_style

    The following removes all data contained in that_style from
    this_style:

        this_style = this_style - that_style

    You can even empty all the data by doing this:
    
        this_style = this_style - this_style

    
    The current version of this class will probably fail to parse
    most stylesheets containing syntax errors. It makes no attempt
    to validate the stylesheet, however, so any data following CSS
    syntax will work.

    """

    ## capture a specification group -- e.g., a group of style
    ## properties pertaining to a particular context
    specgroup = re.compile(r"([^{}]+){([^{}]+)}")
    ## weed out supposed property definitions that are empty
    ## or missing the required colon.
    bogus = re.compile(r"^[^:]*$")

    def __init__(self, datasrc, defaultcontext=None):
        """
        Arguments:
        
        datasrc -- The data source for this instance. May be given in
        any of three forms: (1) the name of a CSS file, (2) a string
        containing style data in CSS syntax, or (3) a dictionary in
        the same form as self.data

        defaultcontext -- (OPTIONAL) The default context from which
        others inherit properties. If your stylesheet is for an
        ordinary web page, for example, you might use "BODY" as a
        default context. Must be present in the data source.

        """
        UserDict.__init__(self)
        if type(datasrc) is type({}):
            self.data = datasrc
        else:
            if os.path.isfile(datasrc):
                cssdata = self._readin(datasrc)
            elif type(datasrc) is type(''):
                cssdata = datasrc
            else:
                raise RuntimeError, 'Invalid data type: %s' % type(datasrc)
            self._parse(cssdata)
        self.defaultcontext = defaultcontext

    def __add__(self, other):
        if ((type(other) is type(self) and
            other.__class__ is self.__class__) or
            type(other) is type({})):
            return CSS(self.merge(other))
        elif os.path.isfile(other) or type(other) is type({}):
            temp = CSS(other)
            newcss = CSS(self.merge(temp))
            del(temp)
            return newcss
        else:
            raise RuntimeError, 'Invalid data type: %s' % type(other)

    def __radd__(self, other):
        if os.path.isfile(other) or type(other) is type({}):
            temp = CSS(other)
            newcss = CSS(self.merge(temp, 1))
            del(temp)
            return newcss
        else:
            raise RuntimeError, 'Invalid data type: %s' % type(other)

    def __sub__(self, other):
        if ((type(other) is type(self) and
            other.__class__ is self.__class__) or
            type(other) is type({})):
            return CSS(self.remove(other))
        elif os.path.isfile(other) or type(other) is type({}):
            temp = CSS(other)
            newcss = CSS(self.remove(temp))
            del(temp)
            return newcss
        else:
            raise RuntimeError, 'Invalid data type: %s' % type(other)

    def __rsub__(self, other):
        if os.path.isfile(other) or type(other) is type({}):
            temp = CSS(other)
            newcss = CSS(self.remove(temp, 1))
            del(temp)
            return newcss
        else:
            raise RuntimeError, 'Invalid data type: %s' % type(other)

    def _error(self, strict, errmsg=''):
        if strict:
            raise RuntimeError, errmsg
        else:
            return 0

    def _readin(self, file):
        try:
            f = open(file, 'r')
        except IOError:
            print 'Unable to read file %s' % file
        css = f.read()
        f.close()
        return css

    def _parse(self, cssdata):
        propdict = {}
        while cssdata:
            spec = self.specgroup.match(cssdata)
            if spec is None:
                break
            selectors, stylegroup = spec.groups()
            ## Parse the styles first. That way, if there are
            ## multiple identifiers, we don't have to redo
            ## the styles for each one.
            for style in string.split(stylegroup, ';'):
                if self.bogus.match(style):
                    break
                style = string.split(style, ':')
                prop = string.strip(style[0])
                val = string.split(style[1], ',')
                if len(val) < 2:
                    propdict[prop] = string.strip(val[0])
                else:
                    propdict[prop] = []
                    for item in val:
                        propdict[prop].append(string.strip(item))
            for sel in string.split(selectors, ','):
                sel = string.strip(sel)
                ## Hierarchical selector, e.g. .Body P
                hiersel = string.split(sel)
                if len(hiersel) > 1:
                    sel = tuple(hiersel)
                if not self.has_key(sel):
                    self[sel] = {}
                for prop, val in propdict.items():
                    if not self[sel].has_key(prop):
                        self[sel][prop] = val
            cssdata = cssdata[spec.end():]

    def getspec(self, context=None, strict=0, inherit=1,
                usecurrent=[], set={}):
        """
        Return dictionary of style properties for a given context.

        Arguments (all optional):

        context -- (A dictionary key representing the context where
        this style spec will apply). If omitted, self.defaultcontext
        will be used.

        strict -- (boolean) If true, any properties having values of
        None, and any context names that do not exist in self.data,
        will cause errors. If false, values of None will be returned,
        and non-existent context names will be silently ignored.

        inherit -- (boolean) If true, the returned data will include
        inherited properties. If false, only properties explicitly
        defined for this context will be returned. If true and
        'usecurrent' is empty , all properties applying to this
        context, whether explicitly defined or inherited, will be
        returned.

        usecurrent -- (list of property names) If any names are
        listed, only the listed properties will be returned, using
        inherited values for those not explicitly defined in this
        context.

        set -- (dictionary of property names and values) Sets
        properties to the specified values. This argument overrides
        existing values. If a 'set' argument is supplied, 'inherit'
        is false, and 'usecurrent' is omitted, the method will simply
        return the value of 'set'.

        """
        spec = {}
        base = self.defaultcontext or ''
        if type(context) is type(''):
            context = [context]
        elif type(context) is type(()):
            if self.data.has_key(context):
                context = list(context) + [context]
            else:
                context = list(context)
        if inherit:
            if base and context[0] != base:
                context.insert(0, base)
        else:
            context = context[-1]        
        if usecurrent or not inherit:
            inheritall = 0
        else:
            inheritall = 1
            context = context or base
        if usecurrent:
            ## Work down from current context to default
            context.reverse()
            found = []
            for prop in usecurrent:
                for layer in context:
                    if self.data.has_key(layer):
                        propdict = self.data[layer]
                        if propdict.has_key(prop):
                            spec[prop] = propdict[prop]
                            found.append(prop)
                            break
                    elif strict:
                        raise RuntimeError, "Invalid selector: '%s'." % layer
            for prop in usecurrent:
                if prop not in found:
                    spec[prop] = None
        if inheritall:
            ## Work up from default context
            for layer in context:
                if self.data.has_key(layer):
                    for prop in self.data[layer].keys():
                        spec[prop] = self.data[layer][prop]
                elif strict:
                    raise RuntimeError, "Invalid selector: '%s'." % layer
        if set:
            for prop in set.keys():
                spec[prop] = set[prop]
        if strict:
            for prop in spec.keys():
                if spec[prop] is None:
                    raise RuntimeError, "Property not found: '%s'." % prop
        return spec            
        
    def setspec(self, context=None, strict=0, inherit=1,
                keeponly=[], set={}):
        """
        Define style properties for a particular context.

        [ DON'T USE THIS YET! Something is screwy in the way this
        method calls self.getspec(). To be fixed soon. ]

        Arguments (all optional):

        context -- (A dictionary key representing the context where
        this style spec will apply). The key may, but need not, be
        present in the instance data. If omitted, self.defaultcontext
        will be used.

        strict -- (boolean) See getspec documentation.

        inherit -- (boolean) If true, any existing properties applying
        to the given context, but not specified in 'set', will be
        retained, including inherited properties. If false, the new
        settings will include only the properties explicitly defined
        for the context.

        keeponly -- (list of property names) If empty, all existing
        properties will be retained. If any property names are listed,
        only the properties listed here or in 'set' will be kept; all
        others will be removed.

        set -- (dictionary of property names and values) Sets
        properties to the specified values. This argument overrides
        inherited values.
        
        """
        if context is None:
            try:
                ## MakeError is undefined, so raises exception
                context = self.defaultcontext | MakeError
            except:
                raise RuntimeError, 'No context for style spec.'
        if not self.data.has_key(context):
            self.data[context] = {}
        if keeponly:
            newspec = self.getspec(context, strict,
                                   usecurrent=keeponly, set=set)
        elif inherit:
            newspec = self.getspec(context, strict, set=set)
        elif set:
            newspec = set
        else:
            newspec = self.getspec(context)
        for prop in set.keys():
            self.data[context] = newspec


    def merge(self, cssobj, selfish=0):
        """
        Assimilate data from another CSS instance.

        Arguments:

        cssobj -- a CSS class instance

        selfish -- (optional, boolean) In case of conflicts, this
        argument specifies whether new or existing data takes
        precedence. If false, new data (specified in 'cssobj')
        will take precedence; if true, all existing data will
        be preserved.

        """
        if selfish:
            ## Dunno why I was getting errors with this ...
            ## result = copy.deepcopy(cssobj)
            result = {}
            for k in cssobj.keys():
                result[k] = copy.copy(cssobj[k])
            newdata = self.data
        else:
            ## result = copy.deepcopy(self.data)
            result = {}
            for k in self.data.keys():
                result[k] = copy.copy(self.data[k])            
            newdata = cssobj
        for k in newdata.keys():
            if result.has_key(k):
                for l in newdata[k].keys():
                    result[k][l] = newdata[k][l]
            else:
                result[k] = newdata[k]
        return result

    def remove(self, cssobj, selfish=0):
        """
        Remove specified data from the current instance.

        cssobj -- a CSS class instance

        selfish -- (optional, boolean) If true, the contents
        of self.data will be removed from cssobj.data. If false,
        the contents of cssobj.data will be removed from
        self.data.
        
        """
        if selfish:
            result = copy.deepcopy(cssobj)
            rmdata = self.data
        else:
            result = copy.deepcopy(self.data)
            rmdata = cssobj        
        for k in rmdata.keys():
            if result.has_key(k):
                for m in rmdata[k].keys():
                    if result[k].has_key(m):
                        del result[k][m]
            if not result[k]:
                del result[k]
        return result

            
if __name__ == '__main__':
    import sys, os
    cssfile = raw_input('What file would you like to parse?\n> ')
    cssfile = string.strip(cssfile)
    if string.find(cssfile, '~/') == 0:
        try:
            cssfile = os.path.join(os.environ['HOME'], cssfile[2:])
        except:
            print """This system appears not to support filenames beginning
with '~'. Please try again using the full path.
"""
            sys.exit()
    try:
        mycss = CSS(cssfile)
        print """'%s' successfully parsed. Data is as follows:

%s
        """ % (cssfile, mycss)
    except:
        print """Failed to parse '%s.' Please check for CSS syntax
errors. If your CSS file is correct, please send a
detailed bug report to <mgushee at havenrock.com>, including
a copy of '%s'.
        """

------------- cut above this line -----------------------------------


-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From bciceron at rsn.hp.com  Fri Jan 21 19:20:45 2000
From: bciceron at rsn.hp.com (Benoit Ciceron)
Date: Fri, 21 Jan 2000 18:20:45 -0600 (CST)
Subject: REBOL job (!)
In-Reply-To: <slrn88fmtd.5fj.wtanksle@hawking.armored.net>
Message-ID: <Pine.HPX.4.20.0001211753470.2828-100000@ztt.rsn.hp.com>


http://www.dice.com/DandL/c/cooksys.JM090.html







From mal at lemburg.com  Tue Jan 18 06:59:55 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 18 Jan 2000 12:59:55 +0100
Subject: DBI spec and parameter passing
References: <3883A0CA.AB0D4854@exceptionalminds.com> <14467.42391.905771.673058@beluga.mojam.com> <1263967432-11066499@hypernet.com> <20000118015809.F21652@xs4all.nl>
Message-ID: <388455BB.154F36EC@lemburg.com>

Thomas Wouters wrote:
> 
> On Mon, Jan 17, 2000 at 07:24:21PM -0500, Gordon McMillan wrote:
> > Skip wrote:
> > >               """ country = %s""", (city, state, country))
> 
> > but Skip _meant_:
> 
> > c.execute(""" select id from city where city = ? and state = ?
> > and country = ?""", (city, state, country))
> 
> Doesn't that depend on the DB-API module in question ? As I recall, it's
> module-dependant what kind of arguments it accepts, and you can query an
> attribute to find out the specifics of that module.

You probably mean the parameter marker: yes that is module
dependant. The '?' marker is used by all ODBC drivers and
managers and thus also by mxODBC :-)

Note that one should always use binding parameters if
possible: the driver or interface usually knows better how
to encode the data according to the DBs specs and it's
also much easier to write and more portable. Especially
column types such as date, time and BLOBs are represented
differently among databases.

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




From nemeth at iqsoft.hu  Thu Jan  6 14:29:03 2000
From: nemeth at iqsoft.hu (Nemeth Miklos)
Date: Thu, 06 Jan 2000 20:29:03 +0100
Subject: Python plugin
Message-ID: <3874ECFF.64095C00@iqsoft.hu>

Is there a (webbrowser) plugin for executing Python scripts?
I'm having a lot of problem with Java, and I am seeking alternatives.

Miklos Nemeth



From mgushee at havenrock.com  Sat Jan  8 12:18:45 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 08 Jan 2000 12:18:45 -0500
Subject: Looking for threading tutorial/HOWTO
References: <m3zougva3q.fsf@havenrock.com> <857ph5$jsh$1@nntp3.atl.mindspring.net>
Message-ID: <m3ogawv5vu.fsf@havenrock.com>

aahz at netcom.com (Aahz Maruch) writes:

> This following code example is not well-documented, but should be pretty
> easy to follow; it's designed to test a web server with multiple
> clients:

Thanks! Looks like a good start. If anybody has additional info, I'm
still watching.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From larsga at garshol.priv.no  Fri Jan 14 10:09:43 2000
From: larsga at garshol.priv.no (Lars Marius Garshol)
Date: 14 Jan 2000 16:09:43 +0100
Subject: Perl speed vs. Python speed
References: <85hjfa$mlp$1@nnrp1.deja.com> <pd0f4.26480$W2.358425@iad-read.news.verio.net> <00f801bf5d0e$ae601140$f29b12c2@secret.pythonware.com>
Message-ID: <m366wwr8p4.fsf@lambda.garshol.priv.no>

* Fredrik Lundh
| 
| (searching for the given patterns in strings padded with blanks on
| both sides.  sre8 is the new python 1.6 engine on 8-bit strings,
| sre16 is the same engine using unicode)

Which Unicode encoding will be used in the 16bit version? UCS-2 or
UTF-16?

--Lars M.


From fredp at mygale.org.nospam  Mon Jan 17 03:33:00 2000
From: fredp at mygale.org.nospam (Fred Pacquier)
Date: 17 Jan 2000 08:33:00 GMT
Subject: Python Pre-processor (like php/asp)
References: <6D8A17398E28D3119F860090274DD7DB4B3DD7@pces.cadlab.it> <8EBBD05FBjoshtinamecom@209.208.0.7>
Message-ID: <8EBE6E5BCPaCmAnRDLM@194.2.0.33>

josht at iname.com (Josh Tompkins) said :

>Ummm... I'm not sure how useful this is, but there was an article in a
>recent Dr. Dobbs Journal (January, I think) about something called
>Python Server Pages (PSP).  It is supposed to be mostly compatable with
>ASP and is written in 100% python (for now, at least).

It's JPython, actually. So server-side you need a servlet engine, a JVM, 
and JPython.

-- 
YAFAP : http://www.multimania.com/fredp/


From paul at digicool.com  Sun Jan  9 19:13:03 2000
From: paul at digicool.com (Paul Everitt)
Date: Sun, 09 Jan 2000 19:13:03 -0500
Subject: Text indexing and searching (was "Re: Python stuff")
References: <3875813C.A2F29ECB@trojanslair.zzn.com> <14453.60883.281202.910741@beluga.mojam.com>
Message-ID: <3879240F.4FBCA46@digicool.com>

Skip Montanaro wrote:
> be quite surprised if someone on the list didn't have a better idea about
> text searching or even pointers to modules that solve the problem in a much
> better way than I've outlined.  Accordingly, I've cc'd
> python-list at python.org to shed some more light on the subject.  (The tutor
> list (python-tutor at python.org) is also an excellent source of help.)

Good reply, Skip!  Keagan, if you're interested, Zope has a pretty
comprehensive indexing capability:

  http://www.zope.org/Documentation/How-To/ZCatalogTutorial

(Zope is a modular application server written in Python).

--Paul



From bwarsaw at cnri.reston.va.us  Fri Jan 14 17:14:07 2000
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 14 Jan 2000 17:14:07 -0500 (EST)
Subject: Emacs question
References: <85kr9b$2kh$1@nnrp1.deja.com>
	<002b01bf5de4$2110f7b0$f29b12c2@secret.pythonware.com>
	<jtu2kh52y6.fsf@wazor.biostat.wisc.edu>
	<009501bf5e69$f908e530$f29b12c2@secret.pythonware.com>
	<200001141348.HAA09968@wazor.biostat.wisc.edu>
Message-ID: <14463.40879.393227.519674@anthem.cnri.reston.va.us>

>>>>> "WA" == William Annis <annis at biostat.wisc.edu> writes:

    WA>         Well, that's certainly true.  I suppose I should have
    WA> mentioned in passing that I've had trouble with some operating
    WA> systems trying to parse this:

    WA> #!/usr/bin/env python -- # -*- python -*-

You shouldn't need all that gunk.  With most modern Emacsen, there's a
variable called interpreter-mode-alist which (at least with XEmacs)
has a default mapping for Python.  Sticking

#! /usr/bin/env python

at the top of the file is enough to drop you in Python mode.  This is
covered in the Python Mode installation instructions at

    http://www.python.org/emacs/python-mode/installation.html

-Barry



From wclodius at lanl.gov  Thu Jan  6 12:28:32 2000
From: wclodius at lanl.gov (William B. Clodius)
Date: Thu, 06 Jan 2000 10:28:32 -0700
Subject: Python not a Very High-Level Language?
References: <000801bf5802$dcca7d60$0aa0143f@tim>
Message-ID: <3874D0BE.1B1F4D78@lanl.gov>

Another version of SETL, for UNIX systems only, is available at

http://birch.eecs.lehigh.edu/~bacon/download-setl.html

See also

http://birch.eecs.lehigh.edu/~bacon/

and

http://www.CS.NMSU.Edu/~complog/sets/implem.html


From pf at artcom-gmbh.de  Mon Jan 17 05:49:35 2000
From: pf at artcom-gmbh.de (Peter Funk)
Date: Mon, 17 Jan 2000 11:49:35 +0100 (MET)
Subject: Proposal: i-String literals in Future Version of Python?
Message-ID: <m12A9j5-000CxTC@artcom0.artcom-gmbh.de>

If someone is going to internationalize an application in Python, 
the source will become slightly less readable.  
The following example statements
        print "This is just an annoying message"
	b = Tkinter.Button(text="Foo Bar", ...)
will turn into
        print _("This is just an annoying message")
        b = Tkinter.Button(text=_("Foo Bar"), ...) 
or even more ugly:
        print gettext("This is just an annoying message")
        b = Tkinter.Button(text=gettext("Foo Bar"), ...)
or
        print i18n("This is just an annoying message")
        b = Tkinter.Button(text=i18n("Foo Bar"), ...)

Here the *gettext* method (function) has to return a string, which
is possibly translated into a foreign language.  Since Python already
knows raw string literals, I thought that it would be nice, if a
similar notation could be used to mark i18n string literals.
So that some magic in the python interpreter expands
	print i"This is just a message"
into a call to a gettext function internally, if such a function is
available.  May be there could be a sys.i18n function which could be 
initially defined as a identity function:
	def i18n(string):
		return(string)
This way preparing applications for i18n would become even easier.
What do you think?

Regards, Peter
-- 
Peter Funk, Oldenburger Str.86, D-27777 Ganderkesee, Germany, Fax:+49 4222950260
office: +49 421 20419-0 (ArtCom GmbH, Grazer Str.8, D-28359 Bremen)



From a.eyre at optichrome.com  Tue Jan 25 05:55:24 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 25 Jan 2000 10:55:24 -0000
Subject: Dictionary from a file
In-Reply-To: <200001242113.WAA11012@dionysus.fw.cuci.nl>
Message-ID: <000001bf6722$aee3b1d0$3acbd9c2@peridot.optichrome.com>

> # writeabledict.py
> """A writeable dictionary."""
> [snip]

Isn't this exactly what the 'shelve' module provides?

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From ngps at post1.com  Tue Jan 25 09:50:23 2000
From: ngps at post1.com (Ng Pheng Siong)
Date: Tue, 25 Jan 2000 14:50:23 GMT
Subject: Need XML-RPC help
References: <B4B0ED62.316E%tony.mcdonald@ncl.ac.uk>
Message-ID: <86kd7e$doo$1@nnrp1.deja.com>

In article <B4B0ED62.316E%tony.mcdonald at ncl.ac.uk>,
  Tony McDonald <tony.mcdonald at ncl.ac.uk> wrote:
> Problem is, I can't get XML-RPC to work through our proxy.

[Apologies if you see this more than once. My ISP sucks.]

Try this patch:

*** xmlrpclib.py.orig   Sun Sep 19 16:00:47 1999
--- xmlrpclib.py        Sun Sep 19 16:01:02 1999
***************
*** 553,564 ****
      # client identifier (may be overridden)
      user_agent = "xmlrpclib.py/%s (by www.pythonware.com)" %
__version__

      def request(self, host, handler, request_body):
          # issue XML-RPC request

          import httplib
!         h = httplib.HTTP(host)
!         h.putrequest("POST", handler)

          # required by HTTP/1.1          h.putheader("Host", host)
--- 553,576 ----
      # client identifier (may be overridden)
      user_agent = "xmlrpclib.py/%s (by www.pythonware.com)" %
__version__

+     def __init__(self, proxy_host=None, proxy_port=8080):
+         self.use_proxy=0
+         if proxy_host is not None:
+             self.proxy_host=proxy_host
+             self.proxy_port=proxy_port
+             self.use_proxy=1
+
      def request(self, host, handler, request_body):
          # issue XML-RPC request

          import httplib
!         req='http://%s%s' % (host, handler)
!
!         if self.use_proxy:
!             h = httplib.HTTP(self.proxy_host, self.proxy_port)
!         else:
!             h = httplib.HTTP(host)
!         h.putrequest("POST", req)

          # required by HTTP/1.1
          h.putheader("Host", host)

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



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


From thomas at xs4all.net  Fri Jan 21 11:36:51 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 21 Jan 2000 17:36:51 +0100
Subject: SUGGEST: os.getuser()
In-Reply-To: <20000121170432.B843@stopcontact.palga.uucp>; from gerrit.holl@pobox.com on Fri, Jan 21, 2000 at 05:04:32PM +0100
References: <20000116174004.A5995@stopcontact.palga.uucp> <000b01bf6049$36532140$f29b12c2@secret.pythonware.com> <20000116195710.A7700@stopcontact.palga.uucp> <Jylh4.3912$Ef6.1036319@news.shore.net> <20000121170432.B843@stopcontact.palga.uucp>
Message-ID: <20000121173651.C10685@xs4all.nl>

On Fri, Jan 21, 2000 at 05:04:32PM +0100, Gerrit Holl wrote:

> So getpwuid() is never safe? But in my program, I *also* need the
> comment field. So if there are multiple uids the same, I can't get
> the full name?

If you're paranoid about the USER environment variable, but you do not want
trouble with multiple usernames with the same userid, do it the other way
'round: get the pwentry for USER, check the uid, and if they're the same,
use the GECOS field of that entry. If it's not the same, get the first entry
for the uid, and use that GECOS field. 

Or use filter(lambda x: x[3] == uid, pwd.getpwall()) or some such to get a
list of alternatives and give the user a choice ;)

> a-sysop-letting-multiple-identical-uids-be-in-/etc/passwd-
>     -is-very-paranoid-anyway-ly yours,

It doesn't have much to do with being paranoid. It doesn't provide more
security (actually, it provides _less_ security. The same uid, but an extra
password, an extra homedirectory to put .rhosts and .netrc's in, etc.) But it
definately has it's uses -- when you trust someone enough to use a specific
account, but you do not want to give the person the password for other
reasons, for instance.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From hniksic at iskon.hr  Tue Jan  4 12:47:02 2000
From: hniksic at iskon.hr (Hrvoje Niksic)
Date: 04 Jan 2000 18:47:02 +0100
Subject: exception arg
References: <slrn8749le.d0.mn@banger.iol.it>
Message-ID: <9t9puvhohk9.fsf@mraz.iskon.hr>

mn at altern.org (Mirko Nasato) writes:

> I can't understand the following:
> 
> >>> 1/0
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ZeroDivisionError: integer division or modulo
> >>> try:
> ...     1/0
> ... except ZeroDivisionError, arg:
> ...     if arg == "integer division or modulo":
> ...             print "This is what I expect."
> ...     else:
> ...             print 'Unexpected. "arg" is "%s".' % arg
> ... 
> Unexpected. "arg" is "integer division or modulo".
> >>>
> 
> The same seems to occur with every kind of exception, so maybe
> there's an explanation...

`arg' is an exception object, not a string.  Try this:

... ...
... except ZeroDivisionError, arg:
...     pass
...
>>> arg
<exceptions.ZeroDivisionError instance at 804b568>

The source of your confusion is that `arg' *prints* as a mere string:

>>> print "%s" % arg
integer division or modulo


From malcolmt at smart.net.au  Fri Jan 21 05:41:10 2000
From: malcolmt at smart.net.au (Malcolm Tredinnick)
Date: Fri, 21 Jan 2000 21:41:10 +1100
Subject: regex question
In-Reply-To: <01eb01bf63ec$532be300$f29b12c2@secret.pythonware.com>; from Fredrik Lundh on Fri, Jan 21, 2000 at 09:48:41AM +0100
References: <roy-835F11.21172620012000@netnews.nyu.edu> <01eb01bf63ec$532be300$f29b12c2@secret.pythonware.com>
Message-ID: <20000121214110.A1880@Ridcully.home>

On Fri, Jan 21, 2000 at 09:48:41AM +0100, Fredrik Lundh wrote:
[...snip...]
> umm.  speaking as the "official 1.6 regex implementor", I'd
> go for a simple expression, and filter out any hits that are
> not 12 digits.

umm .. speaking as an "official person who has never seen the 1.6 regex
implementation", could you explain what you are talking about? It sounds
interesting, but I cannot find anything on the Python website about what's on
the hotlist for 1.6 (am I going blind?).

> haven't tried it, but I have a gut feeling that the resulting
> code will be faster than any possible alternative...
> 
> (I'm happy to be proven wrong, of course)

Cheers,
Malcolm Tredinnick

--
You have the right to remain silent. Anything you say will be misquoted,
then used against you.




From pbible at home.com  Mon Jan 17 19:22:20 2000
From: pbible at home.com (Paul E. Bible)
Date: Tue, 18 Jan 2000 00:22:20 GMT
Subject: MSSQL database access module?
References: <85vmo5$kia$1@nnrp1.deja.com>
Message-ID: <0nOg4.838$iL4.7059@news.rdc1.tx.home.com>

You can use mxODBC and access MS SQL via ODBC.  See the following link:
http://starship.python.net/crew/lemburg/mxODBC.html

Good Luck,
Paul


<tiddlerdeja at my-deja.com> wrote in message
news:85vmo5$kia$1 at nnrp1.deja.com...
> Could someone recommend a module to use to access a MS-SQL database
> from python?
>
> Any help appreciated.
>
> Thanks.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.




From thomas.heller at ion-tof.com  Mon Jan 31 13:59:35 2000
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Mon, 31 Jan 2000 19:59:35 +0100
Subject: DL_EXPORT (Extending on Windows question)
Message-ID: <038b01bf6c1d$5aec9a10$4500a8c0@thomasnotebook>

How do I compile my C-extensions so that
the usual 
  DL_EXPORT (void) initmodule (void)
is expanded to
  __declspec (dllexport) void initmodule (void)
?

Cannot figure out the right defines...

Thomas Heller
ION-TOF GmbH




From gmcm at hypernet.com  Mon Jan 31 14:30:39 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 31 Jan 2000 14:30:39 -0500
Subject: performance question when using embedding/extending
In-Reply-To: <1237B133D9B4D311947400005A40FDD4639E3D@knk51.kst.siemens.de>
Message-ID: <1262775503-10797476@hypernet.com>

Eibl Christian writes:
> 
> i need to call a python function from my c++ app and this python function in
> turn
> accesses an instanciated c++ object (i pass the pointer to the object as a
> parameter).
[snip]
> i "quantified" the application an see that almost all the cpu-time is used
> in "PyEval_CallObject".
> the application i'm building needs at least 10-times the performance it has
> now...
> 
> my question:
>  - is this just the (slow) way it is ?

Python functions have a relatively high overhead.

>  - is there a faster way to use embedding/extending (details see below) ?

Getting rid of SWIG and doing a bare-metal C layer yourself 
should (at a guess) meet your performance goals. That 
probably means associating a your C++ object with an 
extension type (see the extending docs) so the Python -> 
C++ stuff needs no shadow class or string / pointer conversion.

You could look at CXX (from the LLNL download - link on the 
Python home page) which is a full-featured Python / C++ API. 
I have a lighter-weight one (scxx - see my starship pages); for 
an example of a truly perverse way of making an extension 
type and a C++ class almost the same thing, see how scxx is 
used in the Python bindings of MetaKit (you'll find that on 
SourceForge).


- Gordon



From samschul at pacbell.net  Thu Jan 27 14:23:49 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Thu, 27 Jan 2000 19:23:49 GMT
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com>
Message-ID: <86q5vv$mci$1@nnrp1.deja.com>

In article <2eYj4.1931$UH3.230869 at news4.usenetserver.com>,
  "Joel Lucsy" <jjlucsy at concentric.net> wrote:
> It's at http://www.concentric.net/~jjlucsy/python15.dll
>
> "Paul Robinson" <paul.robinson at quantisci.co.uk> wrote in message
> news:38900E46.909421F8 at quantisci.co.uk...
> >
> > Joel Lucsy wrote:
> > >
> > > I have a version of python15.dll that I've compiled using the
delayload
> > > mechanism. Basically this means that it will only load a dll when
it
> needs
> > > it.

Could you please explain the delayload mechanism, as I could not find a
mention on it in the MSVC Development environment.

Sam Schulenburg


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


From dworkin at ccs.neu.edu  Fri Jan 21 10:26:58 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 21 Jan 2000 10:26:58 -0500
Subject: Using tabs effectively. Was: Why not use tabs?
In-Reply-To: "Gordon McMillan"'s message of "Fri, 21 Jan 2000 08:49:06 -0500"
References: <1263658531-14676592@hypernet.com>
Message-ID: <vndk8l3zbr1.fsf@betelgeuse.ccs.neu.edu>

"Gordon McMillan" <gmcm at hypernet.com> writes:

> Um, there's no tabsetting that will produce that result, but with 
> the wonders of email, maybe that's not what you wrote.

I believe that James was thinking in terms of typewriter-style "tab
settings", where a tab is not a fixed width, but rather moves to the
next column that is defined as a tab stop.  Such things are common in
word processors, but absent in most text editors.

However, I have no idea how such things are relevant to Python.  :-)

-Justin

 



From PClaerhout at CREO.BE  Sat Jan 29 05:09:32 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Sat, 29 Jan 2000 11:09:32 +0100
Subject: http://chill.webhostme.com/python/
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA709C@MSGEURO1>

Hello everybody,

I'm starting a little website with some usefull stuff on it about Python.

You can check it out at http://chill.webhostme.com/python/.

Kind regards,


Pieter



From amused at webamused.com  Fri Jan 21 10:53:53 2000
From: amused at webamused.com (Joshua Macy)
Date: Fri, 21 Jan 2000 15:53:53 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu>
Message-ID: <38887EF0.C7DCBDA6@webamused.com>

Roy Smith wrote:
> I'm in love with python, but this idea of white space as significant
> syntax is the biggest wart on the language by far.

 As far as I'm aware, white space has been significant in syntax in
every major language since early FORTRAN.  The days when FORK=ITOJ and
FOR K = I TO J are equivalent are long gone.  The white space that
indicates block levels in Python is no more invisible than the white
space that delimits keywords and identifiers in other languages.

  Joshua


From skaller at maxtal.com.au  Fri Jan  7 12:48:45 2000
From: skaller at maxtal.com.au (skaller)
Date: Sat, 08 Jan 2000 04:48:45 +1100
Subject: "sins" (aka, acknowledged language problems)
References: <6D8A17398E28D3119F860090274DD7DB4B3D51@pces.cadlab.it> <38656610.E4DA287B@maxtal.com.au> <006c01bf4f82$a39bc220$bf2b2bc1@martelli>
Message-ID: <387626FD.9572471C@maxtal.com.au>

Alex Martelli wrote:

> > I have introduced ONE entirely
> > new statement into Viper (my version of Python):
> >
> > with: ...
> > do: ...

	.. and now, in alpha 2 (released!), there is another
new keyword for the module statement:

	module X:
		x = 1
		y =2

which creates a nested module. 

> > in which the variables introduced in the with part
> > are available in the do part, and then forgotten.
> > This adds to the lexical scoping Viper provides,
> > to correct Python's lousy scope control.
> 
> Now that sounds extremely interesting.  Python's
> scopes are extremely simple, but perhaps this IS
> too simple, as you claim.

	I wouldn't say they're simple. I spend ages
puzzling over the weird two scopes system.
 
> Perhaps a solution might hinge around extending
> the "exec" statement, which already provides
> for explicitly supplied local and global dictionaries,
> but currently only executes a string or a code
> object.  If "exec" would let me express the "code
> object to be executed" as a plain suite of normal
> Python statements, we'd be there, wouldn't we?
> 
> I.e., rather than your:
>     with:
>         x=23
>         y=45
>     do:
>         z=x+y
> 
> we might instead write:
> 
>     exec:
>         z=x+y
>     in {'x':23, 'y':45}
> 
> or some similar variation thereof.  This might fit
> better with existing constructs, requiring no new
> keywords, and perhaps providing better control
> thanks to the explicit supplying of the dictionaries
> being used/affected...?

	It is not equivalent. The 'with' block can contain
arbitrary code. .. I have no qualms about using new keywords
if they're appropriate.

> What are the drawbacks of supplying the needed
> wrappers, rather than adding new syntax...?

	Performance, readability.
 
> What about the "for x in y" syntax -- that seems neat
> to me, as long as we can supply the y appropriately:-).

	I implemented:

	for k,v in dict: ...

No new keywords. Also, no wrappers. So in the end,
the real solution was the simplest: it just took me a while
to figure out that 'for' runs through the items of a container:
for sequences, the index is not an item -- even though
sequences and dictionaries are both (python) maps.

Works for sets too - Viper 2.0 alpha 2 supports a set type:

	s = {1,2,3}
	for mem in s: print mem

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From gerrit.holl at pobox.com  Mon Jan 31 11:09:08 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 31 Jan 2000 17:09:08 +0100
Subject: Namespace pollution
In-Reply-To: <949332864.553076@news.adtranzsig.de>; from hei@adtranzsig.de on Mon, Jan 31, 2000 at 03:32:19PM +0100
References: <949332864.553076@news.adtranzsig.de>
Message-ID: <20000131170908.A831@stopcontact.palga.uucp>

Dirk-Ulrich Heise wrote on 949329139:
> In the recent whitespace war,
> some people said "Python is such an expressive
> language, you need only short functions.", "don't indent
> too much, that's bad style", even Linus Thorvalds was
> quoted (who likes 1 TAB = 8 spaces indentation to not
> lose control)...
> 
> I've got one problem with all this stuff:
> I end up with dozens of helper functions that are
> called exactly one time. They pollute the namespaces
> of the module or class they're defined in. I hate polluted
> namespaces.
> 
> How do others cope with this problem (that does not only
> apply to Python)?

Nest functions.

If you have a function X only used in _one_ other function Y, move the
function definition X to Y.

regards,
Gerrit.

-- 
Please correct any bad Swahili you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From news at titanic.co.uk  Fri Jan 14 14:04:52 2000
From: news at titanic.co.uk (Simon Faulkner)
Date: Fri, 14 Jan 2000 19:04:52 GMT
Subject: Database to Web
Message-ID: <jmsu7sk17q2hg0dbrf3qi8lobednf72m4a@4ax.com>

Has anyone got a good sugestion for the easiest way to display the
results of an sql query on a web page?

This works but it's not pretty!

TIA

def listdetails():
  db =MySQLdb.connect(host="localhost",db='BTC',user="root",passwd="")
  cursor = db.cursor()
  cursor.execute("SELECT * FROM tblDetail WHERE lngSalesPersonID=30")
  tuple = cursor.fetchall()
  total = len(tuple)
  if total < 1:
    print "No entries"
  else:
    print "Content-type: text/html\n"
    print "<H3>DETAILS</H3>"
    print "<P>"
    print "<table border cellspacing=0 cellpadding=5>"
    for record in range(total):
      Name =  tuple[record]
      print "<tr>"
      print Name
      print "</tr>"
    print "</table>"
  db.close()

Simon


From aahz at netcom.com  Fri Jan  7 19:03:15 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Jan 2000 00:03:15 GMT
Subject: urllib.urlopen returns 'connection refused' on Solaris
References: <3875CBCF.72E26CE7@ina.fr>
Message-ID: <855us3$lae$1@nntp6.atl.mindspring.net>

In article <3875CBCF.72E26CE7 at ina.fr>, Richard Gruet  <rgruet at ina.fr> wrote:
>
>On Solaris 2.6 (Sun OS 5.6), the fct urllib.urlopen() returns an error
>'Connection refused' for all URLs:

Python 1.5.2 works fine for me on Solaris 2.5.1.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From robin at jessikat.demon.co.uk  Wed Jan  5 12:47:24 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Wed, 5 Jan 2000 17:47:24 +0000
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
References: <20000105082703.A9441@vislab.epa.gov>
 <1265040530-3892638@hypernet.com>
Message-ID: <ONAe+FAsO4c4Ew5h@jessikat.demon.co.uk>

In article <1265040530-3892638 at hypernet.com>, Gordon McMillan
<gmcm at hypernet.com> writes
>Robin Becker wrote:
> > > |Also under
>> certain extreme circumstances my code would like to report a >
>> |total failure eg attempt to divide by error. I would like to ask
>> more > |experienced pythoneers how arithmetic errors are handled
>> using built in > |IEEE exceptions and how does one gracefully
>> crash a python script? (eg > |an abend/abort). 
>
>> >raise
>> ArithmeticError, "Oops" >raise ZeroDivisionError, "Ouch" > >? >
>
>> yes that's python, but what's the best way to do it in an
>> existing C library? -- Robin Becker
>
>PyErr_SetString(PyExc_ZeroDivisionError, "Ouch");
>return NULL;
>
>(see bltinmodule.c for the complete list).
> 
>> -- 
>> http://www.python.org/mailman/listinfo/python-list
>
>
>
>- Gordon
>

I guess I could have made my query a bit clearer. When I call my
extension in C I can pass inputs which may cause unexpected arithmetic
exceptions. That's an efficiency saving. I could check every arithmetic
op. Assuming I want speed is there a standard pythonic way eg
signal+setjmp/longjmp to trap these errors?

 In the VBA interface I have implemented exception trapping using VC++
__try blocks (I seem to remember this was because of Excel somehow).

The current C uses a pair of global functions to handle I/O to
stdout/err. What's a good way to provide pythonic output for a C
extension module?

One other thing I am a bit uncertain about is threading. If I call
python inside my C extension must I worry about threading issues for the
extension?
-- 
Robin Becker


From hartmut at oberon.noris.de  Mon Jan  3 13:48:05 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Mon, 03 Jan 2000 19:48:05 +0100
Subject: Pseudo-SQL: searching for Parser/Rewriter
References: <16600378@oberon.noris.de> <87vh5c35xh.fsf@baal.winnegan.de>
Message-ID: <16600381@oberon.noris.de>

Siggy Brentrup (bsb at winnegan.de) schrieb:

 >As soon as the search engine on www.python.org is up again, look out
 >for Gadfly by Aaron Watters, it's an SQL database engine written in
 >Python.

As far as I understood, Gadfly is it's own DBMS. I want to use postgresql for
speed reasons.

 >/siggy


+++hartmut


From wware at world.std.com  Thu Jan  6 17:11:50 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 6 Jan 2000 22:11:50 GMT
Subject: newbie embedding questions
References: <kX6d4.41059$Ke.26557@news21.bellglobal.com>
Message-ID: <Fnxpnq.JrD@world.std.com>

Jason Maskell (backov at nospam.csolve.net) wrote:
> Well, I've finally gotten around to embedding Python in an app...
> The main app is essentially a skeleton, that will be driven by Python
> scripts.. It does however hold a bunch of C++ objects that it owns and
> instantiates and that the Python scripts need access to. So basically I need
> "glue" code to make sure the Python stuff has access to those instances'
> methods, and to set callbacks in the python code for generated events.

You might want to look at Swig (www.swig.org) which makes it really
painless to extend Python with C++ objects. I think the preferred
approach with C++ would be more extending than embedding, but from
your description of what you're doing, it sounds like this might
work for you. The only difference is that instead of writing a main()
function that calls the Python interpreter, you'll write a Python
script that imports the module produced by Swig.

I've done a few C and C++ extensions and had great luck. I have much
less experience with embedding and can't offer much insight about it.
If you're dead set on embedding, I would imagine Swig could still help
a lot, but I don't know the details of how you'd use it.
-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From akuchlin at mems-exchange.org  Fri Jan  7 11:55:14 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 07 Jan 2000 11:55:14 -0500
Subject: Reading tar/gz-files
References: <3874DDAA.E14CC23B@bibsyst.no> <p6qzoujouig.fsf@informatik.hu-berlin.de>
Message-ID: <3dhfgpn7nx.fsf@amarok.cnri.reston.va.us>

Martin von Loewis <loewis at informatik.hu-berlin.de> writes:
> provides you with an uncompressed stream. All you'd need to do is to
> interpret the format of tar yourself. See /usr/include/tar.h for a
> starting point.

Clement Hintze wrote a tarlib.py module for reading tar files; it's at
ftp://ftp.python.org/pub/python/contrib-09-Dec-1999/System/tarlib.py .

(Let us have a moment of silence for the FTP section on www.python.org.
 So long and thanks for all the fish^H^H^H^Hfiles...)

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
We mistrust anything that too strongly challenges our ideal of mediocrity.
    -- Robertson Davies, "The Noble Greeks"



From pinard at iro.umontreal.ca  Mon Jan 17 14:02:02 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 17 Jan 2000 14:02:02 -0500
Subject: Python DIFF to track changes in HTML pages anyone?
In-Reply-To: janhansen@my-deja.com's message of "Mon, 17 Jan 2000 11:04:32 GMT"
References: <85usvr$1ps$1@nnrp1.deja.com>
Message-ID: <oqu2kc4j4k.fsf@titan.progiciels-bpi.ca>

janhansen at my-deja.com writes:

> I need to track changes in HTML pages and to return meaningful (to humans)
> pieces of information.  In html pages there is often added a new piece
> (or packet) of information but the "string" concept of common diff sort of
> has no meaning.  crlf' s in html is not logical connected with "packets"
> of info in html.  anyone now of python code for this?

It is not Python code, and not especially tied to HTML, but maybe my
`wdiff' utility might help you.  Let me see...  The latest pretest is at:

   http://www.iro.umontreal.ca/contrib/~pinard/wdiff/wdiff-0.5g.tar.gz

Just remove the `g' if you prefer a "stable" version. :-)

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





From mross at rochester.rr.com  Sat Jan 29 17:57:57 2000
From: mross at rochester.rr.com (Michael Ross)
Date: Sat, 29 Jan 2000 17:57:57 -0500
Subject: Creating a local variable with a parameterized name
Message-ID: <38937075.18B75753@rochester.rr.com>

I'm new to this, so I hope I'm not missing something obvious...

I want to create a local variable with a passed in name.

I realize I can do the following:

>>> def f(varname):
             cmd = '%s = 123' % varname
             exec(cmd)
             cmd = 'print %s' % varname
             exec(cmd)

>>> f('mike')
123
>>>

But exec needs to invoke the parser, which seems to be an extremely
costly thing to do.

It seems local variables are stored in a special, inaccesible (from
python) place in the frame object (I've been looking at
PyFrame_FastToLocals and PyFrame_LocalsToFast). Do I need to write C
code to do this?  A builtin to do this would be nice...

Thanks in advance!
Mike Ross



From emile at fenx.com  Thu Jan  6 17:43:13 2000
From: emile at fenx.com (Emile van Sebille)
Date: Thu, 6 Jan 2000 14:43:13 -0800
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com>
Message-ID: <03cd01bf5897$6d61dfa0$01ffffc0@worldnet.att.net>

FYI,

I get a significant improvement running Darrell's sample 
with string based exceptions (python -X...):

Time: 7.36000001431
Time: 0.5

w/-X
Time: 2.86000001431
Time: 0.490000009537


but running Skip's there is no difference.

I-don't-know-what-it-means-ly y'rs

Emile van Sebille
emile at fenx.com
-------------------






From wware at world.std.com  Thu Jan 20 12:32:24 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 20 Jan 2000 17:32:24 GMT
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu>
Message-ID: <FonA21.B26@world.std.com>

Paul M (paul.magwene at yale.edu) wrote:
: Dear Pythoneers...

A great evil has been perpetuated in the design of the Python
language. It grieves me to be the one to bring this to the world's
attention but it must not continue to go unacknowledged. Persons of
right indentation have once again been marginalized and excluded
because _every_editor_ever_written_ indents from the LEFT SIDE OF THE
PAGE. The brutal, systematic, uniform, codified, societally
legitimized vicious assault on right indentation has gone unnoticed,
despite my many pleas to Greenpeace and Amnesty International. I am
forced to bring this issue to the very feet of the Python community
itself, and beg, nay, demand, that the humanity of right-indenters be
recognized and respected. Let the bloodshed end here.

Right-indenters are not a vindictive lot. We do not insist that the
last fifty years of oppression be compensated overnight. We do not
require that CPython's left indentation be removed and replaced with
right indentation. We are a peaceful people, eager for understanding
and amicable coexistence. In the near term, we can get by with the
following script, which translates left indentation to right
indentation.

                            # to-right.py
                   import sys, string, re

              before = re.compile('^\ +')

                               maxlen = 0
                              lines = [ ]
          for L in sys.stdin.readlines():
                           L = L[:-1]
                  if len(L) > maxlen:
                  maxlen = len(L)
                      lines.append(L)

                          for L in lines:
                 s = before.search(L)
                        if s != None:
                 n = s.regs[0][1]
                        L = L[n:]
                                else:
                            n = 0
print (maxlen - n - len(L)) * ' ' + L

As a temporary retrofit, I also offer the left-indented version:

# to-right.py
import sys, string, re

before = re.compile('^\ +')

maxlen = 0
lines = [ ]
for L in sys.stdin.readlines():
    L = L[:-1]
    if len(L) > maxlen:
        maxlen = len(L)
    lines.append(L)

for L in lines:
    s = before.search(L)
    if s != None:
        n = s.regs[0][1]
        L = L[n:]
    else:
        n = 0
    print (maxlen - n - len(L)) * ' ' + L

Right-indenters are a forebearing bunch, and do not instantly
expect a command-line option allowing for right indentation.
We can be gracious enough to wait to see it in Python 2.0.
-- 
   People say that everyone has a few skeletons in their closet. Not me.
   Well, not yet anyway. I mean, the bodies are still decomposing.
Will Ware		email: wware[at]world[dot]std[dot]com


From cwr at crash.cts.com  Sat Jan 29 02:30:24 2000
From: cwr at crash.cts.com (Will Rose)
Date: 29 Jan 2000 07:30:24 GMT
Subject: Case Sensitivity and Learnability
References: <B4191059635E81B1.8AFDD386EFACF4A3.6FF161EA0151B5EB@lp.airnews.net>
Message-ID: <86u4uf$26lq$1@thoth.cts.com>

Tres Seaver <tseaver at starbase.neosoft.com> wrote:
: Thinking about the notion of case sensitivity as a barrier to CP4E, as
: advocated quite glibly at IPC8, I typed up this module on the plane to
: explore the consequences of allowing/encouraging users to enter case-
: mangled code, especially in non-interactive uses (I think that distinc-
: tion may be important, because readability is almost insignificatn in
: an interactive session).


Your arguments are reasonable, but they are arguments against systems
which are both case-insensitive _and_ case-preserving.  The trouble with
CI/CP is that the compiler sees a different set of tokens from those seen
by the programmer, and it's sometimes difficult for the programmer to
realise this.  If unskilled programmers find case too difficult or
distracting, the cure is to force a single case, either upper or lower. 
This could be done by an (optional?) compiler switch.


Will
cwr at cts.com



From paul at prescod.net  Sun Jan 23 17:48:59 2000
From: paul at prescod.net (Paul Prescod)
Date: Sun, 23 Jan 2000 16:48:59 -0600
Subject: Lexing in Python 2
References: <001501bf63f3$3c018ac0$ec2d153f@tim>
Message-ID: <388B855A.78D6E46B@prescod.net>

Tim Peters wrote:
> 
> Yes, but that's really a different topic.  The Python world has no good
> approach to that now, paying attention to the "fast" part, and where "good"
> means "enough like Flex and Bison that you don't feel you've been stranded
> on some strange alien planet" <wink>.

At the last XML conference I told someone that the reason that re
doesn't take a stream instead of string parameter was because anyone
sane working on a large file would use a proper tokenizer. Shouldn't
such a tokenizer come with Python? With all due respect, what the hell
is shlex and how did it get into the standard distribution? 

I mean the standard distribution alone must contain half a dozen
hand-coded lexers and in a few places, the weirdness you need to apply
regular expressions to streams. Surely we can do better for Python 2?

It is my unconsidered, uneducated opinion that lexers do not vary as
widely as parsers (LL(1), LR(1), LR(N) etc.) so we could just choose one
at random and start building modules around it.

All in favor? Opposed? Carried.
-- 
 Paul Prescod  - ISOGEN Consulting Engineer speaking for himself
Earth will soon support only survivor species -- dandelions, roaches, 
lizards, thistles, crows, rats. Not to mention 10 billion humans.
	- Planet of the Weeds, Harper's Magazine, October 1998



From morse at harborcom.net  Wed Jan 12 01:50:28 2000
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Wed, 12 Jan 00 06:50:28 GMT
Subject: pyhints - Python function signatures in VIM
Message-ID: <mt2.0-27553-947659828@news.informatik.uni-bonn.de>

Pyhints scans one or more Python source files and produces a VIM 'hint' script.
 When sourced, this script displays the function signature in the status line
after you type the function name. This idea is from Dr. Cambell's flist program 
(http://users.erols.com/astronaut/vim/) , which does the same for C/C++.

http://www.geocities.com/SiliconValley/Peaks/4673/python.html#pyhints

<P><A HREF="http://www.geocities.com/SiliconValley/Peaks/4673/python.html#pyhints">pyhints
for VIM</A> - scans one or more Python source files and produces a 'hint'
script describing function signatures; for the VIM editor.  (21-Dec-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 s.schwarzer at ndh.net  Wed Jan 26 04:12:58 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Wed, 26 Jan 2000 10:12:58 +0100
Subject: vi and python?
References: <Pine.LNX.4.20.0001252214240.523-100000@marvin.squad51.net>
Message-ID: <388EBA9A.ECBDEC04@ndh.net>

Hello Ben,

Ben Beuchler schrieb:
> Sounds good.  This may sound like a stupid question, but is there a way to
> have it only use these options with .py files?  I haven't done too much
> yet with customizing vim...

Read the help text on autocommands:
:help autocmd.txt

One thing to consider is to undo the changes you made on (e. g.)
BufEnter for your *.py files. "Undo" is probably difficult or at least
tedious if you want exactly the settings that you had before, not
only a kind of default set.

Stefan


From guido at CNRI.Reston.VA.US  Tue Jan 18 19:40:18 2000
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Tue, 18 Jan 2000 19:40:18 -0500
Subject: Suggestions for python 2
In-Reply-To: Your message of "Wed, 19 Jan 2000 00:28:47 GMT."
             <m12AizP-00006lC@vortigen.demon.co.uk> 
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us> <200001171159.LAA17397@lsls4p.lsl> <200001171403.JAA10668@eric.cnri.reston.va.us>  
            <m12AizP-00006lC@vortigen.demon.co.uk> 
Message-ID: <200001190040.TAA18277@eric.cnri.reston.va.us>

Eddy,

Have you read my (very old) metaclasses article?

It seems to me that where you write

> Body Sun:
>     """The Sun: the star at the heart of the solar system."""
>     type = 'G2 V'
>     magnitude = 4.79
>     aliases = 'Sol',

I would write

class Sun(Body):
    """The Sun: the star at the heart of the solar system."""
    type = 'G2 V'
    magnitude = 4.79
    aliases = 'Sol',

I.e. only the syntax at the top is slightly different, the rest is the
same.  The metaclass is defined in a similar way too; where you define
__call__(*bases, **what), I define __init__(name, bases, what).

There are some details that are different, but in general it seems
that your proposal is of similar power.  Have you considered this?

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



From a.eyre at optichrome.com  Mon Jan 31 05:23:06 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 10:23:06 -0000
Subject: GetObject instead of CreateObject
In-Reply-To: <871sqv$kp$1@nnrp1.deja.com>
Message-ID: <000401bf6bd5$2a25a720$3acbd9c2@peridot.optichrome.com>

> Can someone tell me what function I should call in Python to perform a
> similar task to VB's GetObject?

Maybe it would help (not being a VB programmer) to explain what VB's
GetObject actually does before looking for an analogue in Python.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From gmcm at hypernet.com  Sun Jan 16 13:21:41 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 16 Jan 2000 13:21:41 -0500
Subject: Stopping a thread from the calling process
In-Reply-To: <FoFsDH.8EL@boss.cs.ohiou.edu>
Message-ID: <1264075595-4560668@hypernet.com>

Scott Barron asks:

> is there an easy and fairly portable way of doing this?

If you mean "interrupting" or "killing", no. WIth good reason. 
OS level threads are often "light weight" processes. Generally 
what's been trimmed is all the stuff dealing with reclaming 
resources, releasing locks etc. on thread death. So if you 
"kill" a thread, you're likely to be leaving your process in an 
indeterminate state.

The normal idiom is to ask politely - have the thread 
periodically check some variable to see if it should stop.



- Gordon



From hanche at math.ntnu.no  Sun Jan 23 20:32:03 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 23 Jan 2000 20:32:03 -0500
Subject: Tkinter, threads, mainloops, etc
References: <pcobt6c635a.fsf@math.ntnu.no>
Message-ID: <pco901g5k6k.fsf@math.ntnu.no>

I am following up to my own posting here, since everybody else is
probably at the python conference anyway, and/or getting drunk at the
bar.  Some reckless experimentation shows that, indeed, I can do
graphs in Tkinter without ever calling any mainloop method - however,
I have to throw in the occasional call to an update method, or nothing
is updated until I get the command prompt back.

I'll most likely be doing this live in a seminar in a couple of weeks,
so if there are any gotchas I'd like to know about them before it's
too late.

I'll-probably-stumble-across-them-while-fumbling-in-the-dark-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From PClaerhout at CREO.BE  Fri Jan 21 03:31:43 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Fri, 21 Jan 2000 09:31:43 +0100
Subject: Screen depth
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA7075@MSGEURO1>

Hello,

how can I determine how many colors a screen can show?

Kind regards,


Pieter



From fxml at excite.com  Mon Jan 10 22:17:50 2000
From: fxml at excite.com (fxml at excite.com)
Date: Tue, 11 Jan 2000 03:17:50 GMT
Subject: Zope and NT databases
References: <387956BD.CFDF17DD@bby.com.au> <3879CE9B.BE77787C@bam.com> <3879E4E4.B4BD96FD@ipro.de> <387A10CE.E8CD408F@bellatlantic.net> <85dimq$pi7$1@nnrp1.deja.com>
Message-ID: <85e7cj$91s$1@nnrp1.deja.com>


Hi everyone,
  I am the author of ODBC Socket Server.  If someone writes a Python
client for ODBC Socket Server, please feel free to e-mail it to me and I
will include it in the base distribution.  I will also give you full
credit for the client :)
  Thanks for trying ODBC Socket Server, and if you have any specific
questions please let me know by e-mail or in this newsgroup.  Thanks to
Marc and Jochen for making me aware of this thread.

Thanks,

Tyler

In article <85dimq$pi7$1 at nnrp1.deja.com>,
  marc_risney at my-deja.com wrote:
> Hmmm, yes Zope is one way to do it, ODBC Socket Server,
> is actually rather elegant, here is sample code that you could use
> provided that you download the ODBC Socket Server
> from http://odbc.linuxbox.com/products.html
>
> Python 1.5.2 (#1, Jan  6 2000, 15:54:56)  [GCC egcs-2.91.66 19990314
> (egcs-1.1.2
>   on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>>from socket import *
> >>>import select
>
> >>>FORWARD = 'ntbox.domain.com'
> >>>PORT = 9628
> >>>s = socket(AF_INET, SOCK_STREAM)
> >>>s.connect(FORWARD, PORT)
>
> >>>conn_str = "DSN=ODBCSS;UID=;PWD=;"
> >>>sql_str = "insert into Table1 values ('test1','test2')"
>
> >>>xml_str = """<?xml version=\"1.0\"?
> >\r\n<request>\r\n<connectionstring>""" + conn_str
> + "</connectionstring>\r\n<sql>" + >>>sql_str
> + "</sql>\r\n</request>\r\n"
> >>>s.send(xml_str)
> >>>msg = s.recv(1024)
> >>>s.close()
> >>>print str(msg)
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From news at dorb.com  Sat Jan  8 13:17:03 2000
From: news at dorb.com (Darrell)
Date: Sat, 8 Jan 2000 13:17:03 -0500
Subject: calling C functions from Python
References: <852ck5$7da$1@noao.edu> <s%Kd4.2786$Ef6.704898@news.shore.net>
Message-ID: <%7Ld4.787$mj7.20244@typhoon.nyroc.rr.com>

Don't forget http://starship.python.net/crew/gmcm/scxx.html

--
--Darrell




From gmcm at hypernet.com  Mon Jan  3 21:22:27 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 3 Jan 2000 21:22:27 -0500
Subject: Python not a Very High-Level Language?
In-Reply-To: <slrn872iqi.gie.neelk@brick.cswv.com>
Message-ID: <1265170846-519622@hypernet.com>

Neel Krishnaswami wrote:

> Tres Seaver <tseaver at aftershock.neosoft.com> wrote:
> > 
> > What reasonable use can you propose for an associative
> > container which allows mutable keys?  You might as well revert
> > to a list of (key,value) pairs, since you're going to do a
> > linear search on each lookup anyway if the keys are allowed to
> > change.
> 
> You can compute the hash function using the object's identity,
> rather than its value. IOW, hash on the object's address. Then
> you can mutate the contents of the object as needed and still
> access the hash. In fact, I believe this is the default behavior
> for Common Lisp hash tables (though of course you can create hash
> tables with different hash functions).
> 
> This should be even easier in Python than in Lisp, since objects
> are guaranteed not to be relocated, so a garbage collection can't
> force rehashing.

Well, you could, but then you have to pass around the one-
and-only key to access the value. Seems more sensible, as 
Tres said, to pass around (key, value) instead and just forgo 
the hash.

make-mine-spam-eggs-and-spam-without-the-eggs-ly y'rs

- Gordon



From neelk at brick.cswv.com  Mon Jan  3 20:22:15 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 4 Jan 2000 01:22:15 GMT
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <386E7BD4.514764EA@ndh.net> <84nv0u$bu9$1@nntp9.atl.mindspring.net> <3870DB37.761034EA@roguewave.com>  <75B5A3A6EB0EDE04.FCE8CED8D9B414D5.95006D8FA0ECEB31@lp.airnews.net>
Message-ID: <slrn872iqi.gie.neelk@brick.cswv.com>

Tres Seaver <tseaver at aftershock.neosoft.com> wrote:
> 
> What reasonable use can you propose for an associative container
> which allows mutable keys?  You might as well revert to a list of
> (key,value) pairs, since you're going to do a linear search on each
> lookup anyway if the keys are allowed to change.

You can compute the hash function using the object's identity, rather
than its value. IOW, hash on the object's address. Then you can mutate
the contents of the object as needed and still access the hash. In
fact, I believe this is the default behavior for Common Lisp hash
tables (though of course you can create hash tables with different
hash functions).

This should be even easier in Python than in Lisp, since objects are
guaranteed not to be relocated, so a garbage collection can't force
rehashing.


Neel


From wware at world.std.com  Thu Jan 13 16:04:46 2000
From: wware at world.std.com (Will Ware)
Date: Thu, 13 Jan 2000 21:04:46 GMT
Subject: Desperate: Linux Sound Module
References: <20000113210535.A2015@stopcontact.palga.uucp>
Message-ID: <FoAL7y.50@world.std.com>

Gerrit Holl (gerrit.holl at pobox.com) wrote:
: I'm really desperate: I can't find any module for playing sounds on a
: Linux system! I already searched everywhere: google, python.org, parnassus.
: I just want to play .au or .wav, but even edna can't *play* *play* *PLAY*
: an mp3!

I did this once, let me try to remember how. It wasn't intrinsically
Pythonic, it involved some other thing (sox?) which could be easily
invoked with an os.system() call, and you could pass it command line
arguments such as sampling rate. Here we go, a URL:

	http://surfin.spies.com/Sox/

Nifty-looking things of related interest here:

	http://www.links2go.com/more/www.spies.com/Sox

I get by with a pretty primitive Soundblaster card. Sampled music
sounds just awful, but it plays MIDI ok (different executable:
playmidi, xplaymidi). My interest in this stuff isn't strong enough
to merit pursuing better results.
-- 
   People say that everyone has a few skeletons in their closet. Not me.
   Well, not yet anyway. I mean, the bodies are still decomposing.
Will Ware		email: wware[at]world[dot]std[dot]com


From jblazi at netsurf.de  Fri Jan 28 15:49:58 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Fri, 28 Jan 2000 21:49:58 +0100
Subject: newbie question on modules
References: <38872a70_4@goliath.newsfeeds.com> <Ww8k4.487$4f5.132763@tw12.nn.bcandid.com>
Message-ID: <3891ff16_4@goliath.newsfeeds.com>

Thx.
the problem was that I was not familiar even with the most simple facts.

What I wanted was "from XXX import *" and it seems I was not even able to
ask the right question.
Thx for the answer anyway.

Janos Blazi


Max M. Stalnaker <stalnaker at acm.org> schrieb in im Newsbeitrag:
Ww8k4.487$4f5.132763 at tw12.nn.bcandid.com...
> Would it be adequate for module b to inherit module a?
> Then in module b:
>
> class b(a):
>
> It seems possible that this issue is flowing not from python newness, but
> from OO newness.  Taking a shot at rethinking the structure to account for
> OO possibilities might be useful.  I had been going through that process
on
> a program that started out modular and is ending up OO and the program is
a
> structured a bit different than when it started.
>
> "Janos Blazi" <jblazi at netsurf.de> wrote in message
> news:38872a70_4 at goliath.newsfeeds.com...
> > I have a module a in which a function m_add is defined. Now I want (as
the
> > last statement of module a) load and execute a module b. Now it seems
that
> b
> > does not know the function m_add. How can I change that?
> >
> > I'd like to access everything defined in a in the module b that is the
> > (variable) second part of a.
> >
> > What can I do?
> >
> > Janos Blazi
> >
> >
> >
> >
> >   -----------== Posted via Newsfeeds.Com, Uncensored Usenet News
> ==----------
> >    http://www.newsfeeds.com       The Largest Usenet Servers in the
World!
> > ------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers
> ==-----
>
>




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From drj at mail.bulkley.net  Mon Jan 24 01:27:17 2000
From: drj at mail.bulkley.net (Peter Schopfer (aka Dr. J. - magician))
Date: Sun, 23 Jan 2000 22:27:17 -0800
Subject: Dictionary from a file
Message-ID: <388BF0C4.F0E08005@mail.bulkley.net>

Hello

I have a dictionary that is too big to fit right in my program, so i was
going to have it as a seperat file, labeled dic.py. how would i be able
to retrieve this dictionary. wuld this work:

from dic.py import dictionary
mydic=dic.dictionary

i am a newbie, so any help would be appreciated
thanks

keagan



From donn at u.washington.edu  Fri Jan 28 11:49:14 2000
From: donn at u.washington.edu (Donn Cave)
Date: 28 Jan 2000 16:49:14 GMT
Subject: problems with blocking pipes, threading
References: <86rqr7$gn2$1@news.go.dlr.de>
Message-ID: <86shaa$15g2$1@nntp6.u.washington.edu>

Quoth "Sven Drescher" <Sven.Drescher at dlr.de>:
| I'm on the way to implement a process manager to start and finish specific
| processes. I do that with fork and execv under Solaris and with Mark
| Hammond's win32 extensions under Windows NT. All starts and finishs correct.
| But if I try to redirect the childs stdout with a pipe and read it out
| parallel from parent, the read-function blocks until the child is killed.
| Here is a short example of my test programm. I don't know what's wrong. I
| tried two threads, but still blocking.

I'm on boggy ground here if we're talking about Windows NT.  One of
these days I'm going to have to wade in there.  But at least on UNIX,
this can be the fault of the other process.

Most programs write their output through a C library function like
printf().  Those functions can buffer output, for efficiency, and
unless the output device is a terminal or the programmer specifically
directs otherwise, they will.  So according to its logic, the program
may have written output, but some or all of that output may still be
in that process buffer, waiting to be flushed out by more writes or
by program exit.

If you want to see whether this is really the problem, you might
write a sample program like  sys.stdout.write('Hello\n'); time.sleep(1000.0)
and verify that its pipe blocks this way.  Then add sys.stdout.flush()
after the write, and see if anything is different.

Don't know anything about Solaris, either!  Threads are hairy.

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


From skip at mojam.com  Thu Jan 20 15:06:18 2000
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 20 Jan 2000 14:06:18 -0600 (CST)
Subject: [PSA MEMBERS] getpass .. bug ?
In-Reply-To: <14471.25610.767260.153081@bitdiddle.cnri.reston.va.us>
References: <1000120110301.ZM48480@noah.scripps.edu>
	<14471.25610.767260.153081@bitdiddle.cnri.reston.va.us>
Message-ID: <14471.27322.699533.506747@beluga.mojam.com>

    MFS> Hi, just waned to mention the when the optional argument |prompt|
    MFS> is used in getpass from the getpass module, the passowrd is echo'ed
    MFS> to the terminal.

    Jeremy> This is strange.  I do not see the reported behavior on my
    Jeremy> Sparc. 

Perhaps an environmental thing, IDLE vs non-IDLE?  I keep seeing notes about 
problems with interactive prompting type thingies in an IDLE context.
Perhaps this is another case?

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098



From JoeSmith at bogusaddress.com  Wed Jan 19 05:13:43 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Wed, 19 Jan 2000 10:13:43 GMT
Subject: Python Rocks! - ASP problem?
References: <860ftk$bas@news.or.intel.com>
Message-ID: <38858F25.84946F4B@bogusaddress.com>

I am a newbie as well.  I don't know whether it is better or worse only time
will tell.  The one place where I believe it causes a problem is with ASP.  One
can do the following thing in vbscript
<%
    if some_condition then
%>
        <html stuff here>
<%
    end if
%>

This conditionally put the html code into the page.  If I remember right,
python can't do this in ASP because an end marker is needed.  I would say that
python probably was not designed with ASP in mind.


tye4 wrote:

> I'm a python newbie.. but I like Python better than Perl (no ugly @$#@.. in
> syntax).
>
> Python does have one quirk: no end if or end loops
>
> if x == 10:
>     print 'x is 10'
> print 'Hello world'
>
> # easy to get confused whether this line is in 'if' block or not
> Having an 'end' block would help tremendously.
>
> if x == 10:
>     print 'x is 10'
> end if
> print 'Hello World'
>
> or
> for x in range(1, 10):
>     print x,
> end loop
>
> Indentation is a confusing and lame way to separate what is inside a block
> and what's outside it.
>
> tye4



From gmcm at hypernet.com  Mon Jan 17 19:24:21 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 17 Jan 2000 19:24:21 -0500
Subject: DBI spec and parameter passing
In-Reply-To: <14467.42391.905771.673058@beluga.mojam.com>
References: <3883A0CA.AB0D4854@exceptionalminds.com>
Message-ID: <1263967432-11066499@hypernet.com>

Skip wrote:

> Instead of doing the parameter interpolation yourself, try letting your
> database module do it for you, e.g.:
> 
>     c = db.cursor()
>     c.execute("""select id from city where"""
>               """ city = %s and"""
>               """ state = %s and"""
>               """ country = %s""", (city, state, country))

but Skip _meant_:

c.execute(""" select id from city where city = ? and state = ? 
and country = ?""", (city, state, country))


- Gordon



From pierre at saiph.com  Thu Jan 27 18:34:29 2000
From: pierre at saiph.com (Pierre Imbaud)
Date: Fri, 28 Jan 2000 00:34:29 +0100
Subject: lib reference in Unix man format
References: <200001271548.KAA04215@trm.bmo.com>
Message-ID: <3890D605.BD0F07BD@saiph.com>

Tao Lu wrote:
> 
> Where can I find the lib reference in Unix man format? Forgive me if this has
> been asked before.
> 
> Have been programming in Perl for years, I found those perlfunc/perlop/perlfaq
> are very convenient. Any similar work done for Python developers/newbies?
> 
> Thanks,
> --
> Tao

Unix man format is pretty poor. U can find the lib reference as html,
not bad,
or as info file, browsable under emacs, xemacs or a specific gnu info
browser.

-- 
Pierre Imbaud <pierre at saiph.com>
12 Rue des Bosquets 91480 Quincy Sous S?nart France
Tel:  01 69 00 94 57 Fax 79 65



From parkw at better.net  Wed Jan 26 09:24:30 2000
From: parkw at better.net (William Park)
Date: Wed, 26 Jan 2000 09:24:30 -0500
Subject: vi and python?
In-Reply-To: <Pine.LNX.4.20.0001252215590.523-100000@marvin.squad51.net>
References: <20000125185932.A19619@better.net> <Pine.LNX.4.20.0001252215590.523-100000@marvin.squad51.net>
Message-ID: <20000126092430.A20974@better.net>

On Tue, Jan 25, 2000 at 10:17:17PM -0600, Ben Beuchler wrote:
> On Tue, 25 Jan 2000, William Park wrote:
> 
> > This is what I have in ~/.vimrc relevant to *.py.  You can move the
> > global settings to 'au BufRead *.py set ...' if you like.
> 
> It looks like a useful set of settings, from what I've been able to
> interpret...  What is accomplished by the section you mention above?
> 
> Thanks,
> Ben
> -- 
> "There is no spoon"
> 	-- The Matrix

'autocmd' executes its commands when certain condition happens.  Here,
it executes 'set ...' when entering a new buffer (BufRead) with '*.py'
suffix.

If you're interested, I got whole bunch of stuffs for *.html and *.tex.

William



From bitbucket at isomedia.com  Fri Jan  7 21:33:24 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Sat, 08 Jan 2000 02:33:24 GMT
Subject: Newbie Win32 Questions
References: <855mj9$ckc$1@nnrp1.deja.com>
Message-ID: <3876a1f3.532831280@news.isomedia.com>

There is something I've heard of called "freeze" that I believe makes
an executable that is a combination the python interpreter, your
script, and anything else required, all bundled together.  While this
might solve the problem of installing python on all the client
machines, I'd guess it comes with a file size cost that might mean you
wouldn't want to do this with a lot of scripts.

Plus, virtually every computer should have Python installed on it,
anyway.

Preventing prying eyes is altogether different.  Why don't you want
people to see your code?  Are you surreptitiously collecting data on
your users' music listening habits and sending it to yourself?

Anyway, when you import python libraries (i.e. import foo), Python
compiles the imported file (foo.pyc).  Successive imports will load
from the .pyc version until changes to the .py version require a
recompile.  So you can write a dummy script that imports your "real"
script and fires it up, then distribute just the compiled version of
the "real" script.

Or, you could make your script self-compiling, like the following
(self_compiler.py).  Run "self_compiler.py COMPILE" to make it compile
itself into self._compiler.pyc.  You can distribute self_compiler.pyc
to the people with the crowbars in their faces (provided they have
Python installed, of course), and they can run that.  I don't know how
hard it would be to de-compile your .pyc file, but I'm guessing it's
not impossible, since Python is a powerful, interactive, interpreted
language.

self_compiler.py:
->>>
import sys

class fooClass:
	def __init__ (self):
		print 'Initializing instance variables and stuff.'
		print 'Finished initializing.'
	def run (self):
		print 'Running...'
		print 'Finished.'

if len (sys.argv) > 1:
	if sys.argv[1] == 'COMPILE':
		if __name__ == '__main__':
			print 'Compiling myself.'
			import self_compiler
		sys.exit ()
oObject = fooClass ()
oObject.run ()
-<<<

    -Eugene

On Fri, 07 Jan 2000 21:42:04 GMT, tuttled at my-deja.com wrote:

>
>
>Help please. I've got a couple of questions I
>haven't been able to find any answers to.
>
>1) Can you make a stand alone app? Meaning is
>there a way to compile and distribute one or more
>files without having to install Python on every
>machine you want to run the app on?
>
>2) If not, how do protect source code from prying
>eyes?
>
>TIA,
>Don
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.

import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From gmcm at hypernet.com  Thu Jan  6 14:38:58 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 6 Jan 2000 14:38:58 -0500
Subject: Naming conventions
In-Reply-To: <3874EBF7.8317091D@ndh.net>
Message-ID: <1264934956-6218325@hypernet.com>

Stefan Schwarzer asks:
> 
> are there common naming conventions for Python (functions,
> classes, variables ...)?
> 
> From the code I have seen, I suppose the following naming styles:
> 
> variable   this_is_a_variable
> class      ThisIsAClass (or This_is_a_class?)
> function   this_is_a_funtion
> module     ThisIsAModule, thisisamodule (or what else?)
> 
> Have I forgotten something; do you have more remarks or hints?

The set of naming conventions has a cardinality equal to the 
number of Python users.

Generally agreed:
 class names are Capitalized
 "constants" are ALL_UPPER_CASE

You'll make Windows users lives easier if you use lower case 
for module names.

Any more forceful or extended statement would undoubtedly 
start a religeous war...

only-an-idiot-woud-spell-doThis()-as-do_this()-ly y'rs

- Gordon



From hei at adtranzsig.de  Thu Jan 20 07:09:54 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 20 Jan 2000 13:09:54 +0100
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
References: <Pine.SOL.3.91.1000119233305.25615A-100000@sunse> <vndr9fd1ibd.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <948373510.257625@news.adtranzsig.de>

Justin Sheehy schrieb in Nachricht ...
>Evguenii Smogailov <sei at sunse.jinr.ru> writes:
>
>> Generally for improving readability,
>> (and Perl compatibility).
>
>That's not an and, that's an xor.

You mean a "ne" (see subject).

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





From fredrik at pythonware.com  Fri Jan 14 10:54:54 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 14 Jan 2000 16:54:54 +0100
Subject: XML-RPC Server
References: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de><slrn87ne3i.qla.kc5tja@garnet.armored.net> <xqxr9fkk94u.fsf@skewer.ca.boeing.com>
Message-ID: <008701bf5ea8$d51d7f20$f29b12c2@secret.pythonware.com>

Harry G. George <hgg9140 at skewer.ca.boeing.com> wrote:
> What is the python community's position re SOAP and XML-RPC?  As I
> understand it, XML-RPC is standards-base, platform-independent, while
> SOAP uses MS-sponsored extensions (including, e.g., XML Data).  Can I
> assume that *both* will be supported in the future?

only speaking for myself here (as the xmlrpclib and
soaplib implementer):

- XML-RPC is simple, platform-independent, and already
available (http://www.pythonware.com/products/xmlrpc).
unfortunately, XML-RPC is a bit too simple: it cannot
marshal all common Python data structures (in other
words, it cannot replace pickle).

- SOAP is much more complete, and afaik, is exactly as
platform independent as XML-RPC.  SOAP can marshal
everything that pickle can.

- the goal is to have full implementations of both XML-RPC
and SOAP available in a not too distant future. (and with a
little luck, we might even be able to persuade GvR to ship
them with Python 2.0 or so...)

</F>




From mcalla at home.com  Sun Jan 23 21:14:29 2000
From: mcalla at home.com (Mike Callahan)
Date: Mon, 24 Jan 2000 02:14:29 GMT
Subject: ConfigParser and .ini files
References: <86a1s2$7ha$1@nnrp1.deja.com>	<14472.39533.223900.313932@weyr.cnri.reston.va.us>	<38889FB1.F692A45D@callware.com> <14472.44829.865615.286152@weyr.cnri.reston.va.us> <3888B360.EAE0A804@callware.com>
Message-ID: <9AOi4.71$mF3.1661@news.rdc1.tn.home.com>

On another note, while using ConfigParser I discovered that if you call the
read method with a file that does not exist, no error or exception is
returned. Is this an intended result? If so, is this a good idea? I know I
can get around this easily, I am just curious why no exception is raised.

for example:

>>> from ConfigParser import ConfigParser
>>> test = ConfigParser()
>>> test.read(['test.ini'])         # test.ini does not exist
>>>                                       # no error or traceback

Mike Callahan





From jsight at mindspring.com  Fri Jan  7 22:29:21 2000
From: jsight at mindspring.com (Jesse D. Sightler)
Date: Fri, 07 Jan 2000 22:29:21 -0500
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <3876AF11.DE0201F7@mindspring.com>

http://www.pair.com/

They are quite good.

Chuck Esterbrook wrote:
> 
> Seems like many web host providers, such as MindSpring, offer CGI with
> Perl, but not other languages, notably Python.
> 
> Does anyone know who would host a CGI Python site?
> 
> -Chuck

-- 
---------------
Jesse D. Sightler
http://www.activated.com/


From loewis at informatik.hu-berlin.de  Wed Jan 12 14:41:02 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 12 Jan 2000 20:41:02 +0100
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <000801bf5b43$69184d40$532d153f@tim>
 <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov>
 <20000111114923.B42270@vislab.epa.gov> <387B6733.2C49FB8A@home.com>
Message-ID: <p6qd7r7hycx.fsf@informatik.hu-berlin.de>

Doug Hellmann <doughellmann at home.com> writes:

> How is this different from function overloading, as implemented in C++? 
> I'm not a C++  user, but my impression is the effect would be the same. 
> Is that the point?

No. Function overloading in C++ is a static thing. You look at the
static types, and at all statically visible methods, and then chose
the "best" one (in C++, this considers conversions from argument types
to parameter types as well).

In multi-methods, lookup is dynamic. You don't do anything statically,
but defer all decisions to run-time. Then, you care only about the
dynamic types of the arguments.

Regards,
Martin


From arcege at shore.net  Tue Jan 18 20:14:43 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 19 Jan 2000 01:14:43 GMT
Subject: Pardon my ignorance, data conversion
References: <38829357.ED633932@home.com> <85u6cs$85d$1@nntp9.atl.mindspring.net> <oqpuv04ixm.fsf@titan.progiciels-bpi.ca>
Message-ID: <7e8h4.3861$Ef6.1020706@news.shore.net>

Fran?ois Pinard <pinard at iro.umontreal.ca> wrote:
: aahz at netcom.com (Aahz Maruch) writes:

:> Robert W. McGwier <rwmcgwier at home.com> wrote:

:> >I need to do the simple task of taking a nonnegative floating point
:> >variable and taking its integer part and storing it in an integer.
:> >Damned if I can find how to do it.

:> import math
:> i = int(math.floor(real))

: Would not a mere:

:    integer_value = int(real_value)

: be sufficient?

Floor does funny things to negative numbers.

Python 1.5.2 (#2, May 11 1999, 17:14:37)  [GCC 2.7.2.1] on freebsd3
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> int(4.5)
4
>>> int(-3.4)
-3
>>> int(-4.5)
-4
>>> from math import floor
>>> int(floor(-3.4))
-4
>>> int(floor(-4.5))
-5
>>>

  -Arcege



From hjwidmai at foxboro.com  Tue Jan 18 07:48:21 2000
From: hjwidmai at foxboro.com (Hans-Joachim Widmaier)
Date: Tue, 18 Jan 2000 13:48:21 +0100
Subject: Tkinter hangs again under Windows
In-Reply-To: <38839ED7.7FAB2941@weihenstephan.org>
References: <85v092$3n4$1@nnrp1.deja.com> <38839ED7.7FAB2941@weihenstephan.org>
Message-ID: <20000118134821.A30730@foxboro.com>

Peter:
> Hi Hans-Joachim,
> 
> how about posting the script or the relevant part of it?

Also it's not really long, I see the need to apologize for 200 lines
of not-so-elagant python programming. Well, you asked for it. ;-)

----- cut here -----
#!/usr/bin/env python

from Tkinter import *
from Tkconstants import *
import tkMessageBox, os, sys, string, tkFileDialog, Dialog

__version__ = "0.2"

class CDDat:
    def __init__(self):
        self.artist = ""
        self.title  = ""
        self.data   = []
    
    def append(self, dat):
        c = dat[:1]
        # Die ersten A:- und T:-Eintr?ge extra merken
        if c == "A" and not self.artist:
            self.artist = string.strip(dat[2:])
        elif c == "T" and not self.title:
            self.title = string.strip(dat[2:])
        self.data.append(dat)
    
    def getdata(self):
        return self.data

# Fredrik Lundhs AutoScrollbar

class AutoScrollbar(Scrollbar):
    # a scrollbar that hides itself if it's not needed.  only
    # works if you use the grid geometry manager.
    def set(self, lo, hi):
        flo = float(lo)
        fhi = float(hi)
        self.ll = flo
        self.rr = fhi
        if flo <= 0.0 and fhi >= 1.0:
            # grid_remove is currently missing from Tkinter!
            self.tk.call("grid", "remove", self)
        else:
            self.grid()
        Scrollbar.set(self, lo, hi)
    def pack(self, **kw):
        raise TclError, "cannot use pack with this widget"
    def place(self, **kw):
        raise TclError, "cannot use place with this widget"

class GetProject(Frame):
    def __init__(self, master, db):
        Frame.__init__(self, master)

        self.root = master
        self.file = None
        self.data = []
        self.dbname = None
        self.rtitle = "Read Feurio! DB V " + __version__
        self.root.title(self.rtitle)
        self.root.iconname('readfdb')

        # Hauptfenster mit beiden Pfaden und den Kn?pfen
        # f?r "Anzeigen" und "Beenden"

        # Die beiden Frames (0,0) f?r's Men?, (1,0) f?r Listbox, (2,0) f?r OK-Knopf
        self.menuFrame = Frame(self.root, relief=RAISED, bd=2)
        self.win       = Frame(self.root)
        self.but       = Frame(self.root)

        # Men? erstellen
        self.filebtn      = Menubutton(self.menuFrame,text ="Datei", justify=LEFT)
        self.filebtn.menu = Menu(self.filebtn)
        self.filebtn.menu.add_command(label="DB ?ffnen ...", command=self.openDB)
        self.filebtn.menu.add_separator()
        self.filebtn.menu.add_command(label="Beenden", command=self.quit)
        self.filebtn['menu'] = self.filebtn.menu
        self.filebtn.grid(row=0, column=0)

        # Ok-Knopf im untersten Frame
        self.but.ok = Button(self.but, text="Ok", state=DISABLED, command=self.save)
        self.but.ok.grid(row=0, column=0)

        # Frames layouten
        self.menuFrame.grid(row=0, column=0, sticky=E+W)
        self.win.grid(row=1, column=0, sticky=E+W+N+S)
        self.but.grid(row=2, column=0, sticky=E+W+N+S)

        self.win.grid_rowconfigure(0, weight=1)
        self.win.grid_columnconfigure(0, weight=1)
        self.but.grid_rowconfigure(0, weight=1)
        self.but.grid_columnconfigure(0, weight=1)
        self.menuFrame.grid_columnconfigure(3, weight=1)
        self.root.grid_rowconfigure(1, weight=1)
        self.root.grid_columnconfigure(0, weight=1)

        # Datenbank lesen
        if not db:
            db = self.askOpen()
        if db:
            self.readDB(db)
        return

    def readDB(self, dbname):
        try:
            f = open(dbname, "rt")
            self.data = []
            while 1:
                l = f.readline()
                if l == "":                     # EOF
                    break
                c = l[:1]
                if c == "#":                    # Hier beginnt ein neuer Datensatz
                    cdd = CDDat()
                elif c in "AT0123456789":       # Nur diese interessieren uns
                    cdd.append(l)
                elif c == "$":                  # Und hier endet ein Datensatz
                    self.data.append(cdd)
            f.close()
        except IOError, why:
            tkMessageBox.showwarning("Fehler",
                'Die Feurio!-Datenbank "%s" kann nicht ge?ffnet werden - %s' % (dbname, why[1]),
                parent=self)
            return

        self.root.title(self.rtitle + " - " + os.path.basename(dbname))
        if len(self.data) == 0:
            tkMessageBox.showwarning("Warnung",
                'Die Datei enth?lt keine g?ltigen CD-Beschreibungen',
                parent=self)
            return
        
        # Auswahlliste zusammenbasteln
        self.win.xs = AutoScrollbar(self.win, orient=HORIZONTAL)
        self.win.xs.grid(row=1, column=0, sticky=E+W)
        self.win.ys = AutoScrollbar(self.win, orient=VERTICAL)
        self.win.ys.grid(row=0, column=1, sticky=N+S)
        self.win.box = Listbox(self.win, width=60, height=20, selectmode=BROWSE,
                        yscrollcommand=self.win.ys.set, xscrollcommand=self.win.xs.set)
        self.win.xs.config(command=self.win.box.xview)
        self.win.ys.config(command=self.win.box.yview)
        self.win.box.grid(row=0, column=0, sticky=N+S+E+W)
        self.win.grid_rowconfigure(0, weight=1)
        self.win.grid_columnconfigure(0, weight=1)

        for cdd in self.data:
            self.win.box.insert(END, cdd.artist + '  "' + cdd.title + '"')
        self.win.box.insert(END, "")    # Leere Zeile am Ende
        self.win.grid(row=1, column=0, sticky=N+S+E+W)
        self.win.box.bind("<Double-Button-1>", self.save)

        # Letzten Eintrag anzeigen - er ist der "prime candidate"
        self.win.box.see(len(self.data) - 1)
        self.win.box.select_set(len(self.data) - 1)
        
        self.but.ok['state'] = NORMAL

    def save(self, event=None):
        try:
            i = string.atoi(self.win.box.curselection()[0])
            dat = self.data[i].getdata()
        except IndexError:
            tkMessageBox.showwarning("Fehler", 'Kein Projekt ausgew?hlt', parent=self)
            return
        file = tkFileDialog.asksaveasfilename(title="Speichern als ...",
                        initialdir=".", defaultextension=".cdb",
                        filetypes=[("Feurio!-DB", "*.cdb"), ("Alle Dateien", "*.*")])
        if not file:
            return
        try:
            f = open(file, "wt")
            for l in dat:
                f.write(l)
            f.close()
        except IOError, why:
            tkMessageBox.showwarning("Fehler",
                'Kann "%s" nicht schreiben - %s' % (file, why[1]), parent=self)

    def openDB(self):
        dbname = self.askOpen()
        if dbname:
            self.readDB(dbname)

    def askOpen(self):
        if not self.dbname:
            idir = ""
        else:
            idir = os.path.split(self.dbname)[0]
        return tkFileDialog.askopenfilename(title="Datenbank ?ffnen ...",
                        initialdir=idir, defaultextension=".cdb",
                        filetypes=[("Feurio!-DB", "*.cdb"), ("Alle Dateien", "*.*")])

def main():
    db = None
    if len(sys.argv) > 1:
        db = sys.argv[1]
    tk = Tk()
    GetProject(tk, db)
    tk.mainloop()

if __name__ == '__main__':
    main()

------- cut -------

[This time I'm posting through the mailing list interface; Deja is cumbersome.]
-- 
Hans-Joachim Widmaier         E-Mail:  hjwidmai at foxboro.com
Foxboro Eckardt GmbH          Telefon: +49 711 502 - 397
Pragstrasse 82                Fax:     +49 711 502 - 300
D-70376 Stuttgart 



From guido at CNRI.Reston.VA.US  Sun Jan 16 21:21:35 2000
From: guido at CNRI.Reston.VA.US (Guido van Rossum)
Date: Sun, 16 Jan 2000 21:21:35 -0500
Subject: [Types-sig] Suggestions for python 2
In-Reply-To: Your message of "Mon, 17 Jan 2000 00:15:23 GMT."
             <m129zpL-00006JC@vortigen.demon.co.uk> 
References: <m129zpL-00006JC@vortigen.demon.co.uk> 
Message-ID: <200001170221.VAA10441@eric.cnri.reston.va.us>

> I intend to push this proposal during IPC8.  I would greatly appreciate
> constructive comments in advance, if only to temper my enthusiasm ;^}

Eddy, it's too late in Python's life for radical proposals.  My head
hasn't exploded yet, but it sure hurts whenever I try to read your
proposal carefully.  I'm not a functional programmer, and Python is
not a functional programming language; and as far as I'm concerned it
never will be one.

Maybe you'll make the case better in person -- I'll be there to listen!

(Practical comment: the syntax <generator> <name> [<bases>]:
<suite> won't fly; you seem to want <generator> to be any expression,
and an expression followed by a name is too fragile a construct to be
comfortably parseable (even if it might be unambiguous).)

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



From mgushee at havenrock.com  Sat Jan  8 10:39:29 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 08 Jan 2000 10:39:29 -0500
Subject: Error using string.strip
References: <38775227.2CC0D685@worldnet.att.net>
Message-ID: <m33ds8wp1q.fsf@havenrock.com>

Eric Hausig <codeslave at worldnet.att.net> writes:

> I'm very new to Python.  (In fact, I'm walking through the tuturial
> right now!)

Welcome!

> >>> string.strip('str') + 'ing'
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> NameError: string

This usually happens when you've forgotten an import statement. Just
do this first:

>>> import string

... and you should be fine. I think the concept you need to keep in
mind here is that Python is very modular. In other words, a great many 
commonly used functions/classes/methods are not built in but are in
the standard libraries. Most Python programs of any consequence make
generous use of 'import' statements.

Hope this helps.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From a.eyre at optichrome.com  Mon Jan 31 08:39:28 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 13:39:28 -0000
Subject: sending keystroke to a dos application
In-Reply-To: <slrn89b11p.ool.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <000601bf6bf0$987e2470$3acbd9c2@peridot.optichrome.com>

> I believe that's not what he meant. He has an existing application
> (probably non-Python) that runs in a DOS box. He wants to script it
> with Python by sending keystrokes to it.

Ahhh. Completely misread that one. :)

It depends how the DOS app gets the keystrokes. If it uses the normal
method (i.e. from stdin), then you should be able to do it with
os.popen(). e.g.

# Warning: Untested code
import os

class DOSApp:

	def __init__(self, filename):
		self.progpipe = os.popen(filename)

	def __del__(self):
		self.progpipe.close()

	def sendKeyPress(self, char):
		self.progpipe.write(char)
		self.progpipe.flush()
		
myApp = DOSApp("c:\\someapp.exe")
myApp.sendKeyPress("a")
myApp.sendKeyPress("b")
myApp.sendKeyPress("c")

If it goes straight to the BIOS, it's a lot more tricky, and is
pretty much outside the scope of Python's abilities (unless you're
prepared to write an extension module to do it.)

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From slinkp23 at yahoo.com  Wed Jan 19 04:14:12 2000
From: slinkp23 at yahoo.com (Paul Winkler)
Date: Wed, 19 Jan 2000 04:14:12 -0500
Subject: Realtime scheduling??
Message-ID: <38858064.CDA4866C@yahoo.com>

Hi,

I'm working on a module I eventually hope to use to generate musical
events in realtime. (These events will be sent to MIDI devices or to
software synthesizers.)

I've found that on my Linux system (pretty much stock RedHat 5.2),
scheduling actions with sched.scheduler is not as reliable as I'd
like. Events are anywhere from 2 to 20 milliseconds late. I'd like
to improve the reliability, which brings me to my next question:

Is there any way to run a python script, or rather the python
interpreter itself, with SCHED_FIFO priority on a linux system?
Or any other way to guarantee that something will happen when you
really want it to happen??


-- 
................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
A member of ARMS    ----->    http://www.reacharms.com
or http://www.mp3.com/arms or http://www.amp3.net/arms
personal page   ---->    http://www.ulster.net/~abigoo


From warlock at eskimo.com  Mon Jan 17 12:14:15 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Mon, 17 Jan 2000 09:14:15 -0800
Subject: Any Newbies?
References: <s2Wf4.974$sv3.3886@newsfeed.slurp.net>  <38816C3B.984A3941@mail.mc.maricopa.edu>
Message-ID: <slrn886jf6.2ke.warlock@raven.myth>

On Sun, 16 Jan 2000 07:01:00 GMT, 
 Pablo J. Nevares, in the persona of <pnevares at mail.mc.maricopa.edu>,
 brought forth the following words...:

>Yes. =)
>
>I'm waiting til I sign up with my web host (around 2 weeks) so I can develop
>Python on a live server (as opposed to just running it on the console on my
>local system.
>
>--pn
>

I don't know what OS you are running, but a small dedicated
partition on a harddrive running freebsd or linux would get you a
webserver to play with for little or no cash outlay, and  since
you have access to the logs and the ports (IP ports that is)
that the server is using, you have a great deal more info for
debugging.

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From kc5tja at garnet.armored.net  Sat Jan  8 20:15:06 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 9 Jan 2000 01:15:06 GMT
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu>
Message-ID: <slrn87fol9.15t.kc5tja@garnet.armored.net>

In article <38771491.BA08010 at iqsoft.hu>, Nemeth Miklos wrote:
>(2) No more COM or any Microsoft-only garbage. The only reason I am

COM is not garbage.  Coming from a die-hard anti-Windows proponent, you may
be suprised to hear me say that.  But COM is, in my opinion, vastly superior
to anything CORBA has produced, or ever WILL produce.  I find it interesting
that even Zope succeeds in many areas where CORBA fails.

Furthermore, COM is not Microsoft only.  They may have invented the
technology, and therefore the largest user of it.  But it's now an open
standard -- in the hands of DCE, the same people who manage the RPC
standards.  As an example, Mozilla is using COM to implement all of its
component architecture, because CORBA just plain ain't up to snuff.  This
includes its Linux version.  And then there's the COM implementation I'm
currently working with that I wrote myself.

How many people do you know of can take a CORBA spec and write a crude
implementation in a day?

Before making wide, sweeping generalizations about whose technology "sucks"
or not, please be sure to do the research first (I'm not saying you didn't
-- we're all entitled to our opinions).  Tons of things from Microsoft
"sucks" for various technical reasons.  I can also think of tons of things
in Linux which have almost made me switch back to Windows, come to think of
it. And, in fact, it's those issues that are making me write my own
operating system too.

>(1) Grail was not mentioned as an alternative. I am a bit sad that this
>project was a failure.

Indeed.  I tried Grail once, and it seemed to work well enough, but it was
...sssslllloooowwww....  Perhaps with its core functionality rewritten in C,
it would make a fine and extensible browser.  Perhaps even suitable enough
to replace "man" and "info".  Now THERE'S a set of commands that need to die
a slow and painful death...

>PyQt, etc. communicating via HTTP, XML-RPC, CORBA (all supported by

Bingo -- XML-RPC.  This is perhaps far, far more suitable for your
applications than any plug-in architecture could ever afford.  My
user-management system for Armored Internet will be using XML-RPC
pervasively to support distributed processing.  And yes, it'll use COM on
both Linux and NT boxes as well (sorry -- shameless plug for COM... ;) ).

>Python) with my server. If HTML is acceptable for my application I
>should use Zope with any browser.

And Zope supports XML-RPC to boot.  Combine that with External Methods and
ZClasses, and you'll have one kick-butt system to work with.  It can be
scripted using something like Python, AND it can have a front-end user
interface based on HTML.

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From ivanlan at callware.com  Mon Jan 24 18:05:31 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Mon, 24 Jan 2000 16:05:31 -0700
Subject: Python Rocks!
References: <000d01bf6312$77edd660$31a2143f@tim> <388BB4E1.974E63B7@prescod.net> <slrn895cqh.rar.kc5tja@garnet.armored.net>
Message-ID: <388CDABB.EF07BA23@callware.com>

Hi All--

"Samuel A. Falvo II" wrote:
> 
> In article <388BB4E1.974E63B7 at prescod.net>, Paul Prescod wrote:
> >Wouldn't it be easier if we just added an optional, (but
> >compiler-recognized if present) block delimiter to Python 2? Then people
> 
> Sometimes you just have to stick up for what you believe in.  I say, "No,
> let's not do this."
> 

I don't think we get a vote.  After all, we have Guido the Benevolent
Dictator, don't we?  And we all know how he's going to vote (if you can
call it vote--it's more like fiat).

But for what it's worth, I agree.  I like Python because it's *clean*
and powerful.  This suggested change probably wouldn't affect the power,
but it would certainly increase the level of possible clutter.  Somebody
would find a way to enforce it, too, even if only for their own code. 
That doesn't strike me as The Python Way.

<so-if-you-want-clutter-visit-your-aunt>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From akuchlin at mems-exchange.org  Sat Jan 22 15:58:09 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 22 Jan 2000 15:58:09 -0500
Subject: anydbm - a simple question
References: <20000121110715.A184793@vislab.epa.gov>
Message-ID: <3diu0lbz8e.fsf@amarok.cnri.reston.va.us>

Randall Hopper <aa8vb at yahoo.com> writes:
>     File "/home/rhh/software/python-1.5.2/lib/python1.5/dumbdbm.py", line 87, \
>       in _addval
>       f.write('\0'*(npos-pos))
>   TypeError: can't multiply sequence with non-int

I'll bet you're on Solaris and are picking up the large file support,
which makes the .tell() method of file objects return a long integer.
But you can't multiply a sequence by a long integer, only a regular
integer:

>>> 2*'a'
'aa'
>>> 2L * 'a'
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: can't multiply sequence with non-int

Short-term, you can patch dumbdbm.py to use int(npos-pos), but this is
really an interpreter bug; there's no reason 2L * 'a' should be an
error.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
All the evils of publishing can be traced to one source -- copyright.
    -- Stefan Stykolt, quoted by Kildare Dobbs in _The Living Name_ (1964)



From mwh21 at cam.ac.uk  Mon Jan 24 11:50:31 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 24 Jan 2000 16:50:31 +0000
Subject: anydbm - a simple question
References: <20000121110715.A184793@vislab.epa.gov> <3diu0lbz8e.fsf@amarok.cnri.reston.va.us> <20000124105627.A440284@vislab.epa.gov>
Message-ID: <m3ln5fquqw.fsf@atrus.jesus.cam.ac.uk>

Randall Hopper <aa8vb at yahoo.com> writes:

> On a related not, another rough spot in non-uniform long int support
> cropped up in another thread just recently:
> 
>     >>> str( 0x80000000L )
>     '2147483648L'
>     >>> "%ld" % 0x80000000L
>     OverflowError: long int too long to convert
> 
> Hopefully long int support is on the Python v2 list of wrinkles to smooth.

1.6, I'd hope.

Cheers,
Michael


From mprakash at inetport.com  Fri Jan  7 12:12:56 2000
From: mprakash at inetport.com (Praki Prakash)
Date: Fri, 07 Jan 2000 17:12:56 GMT
Subject: Cookie.load method problem
Message-ID: <38761F38.808AC09@inetport.com>

This is on a NT 4 Python 1.52 version. I am using Cookie.py to store
cookies received from websites. The set-cookie header looks like:
"Set-Cookie: B=a95d7i92qlv3a; expires=Thu, 15 Apr 2010 0:00:00 GMT;
path=/ ; domain=.yahoo.com;" and when I invoke load on this string, it
tells me I am trying to set a reserved attribute. Am I using this
incorrectly? (This is my first attempt at writing Python programs.)

TIA Praki



From backov at nospam.csolve.net  Thu Jan  6 19:23:33 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Fri, 07 Jan 2000 00:23:33 GMT
Subject: Getting SWIG to work with C++/Python
Message-ID: <9mad4.42353$Ke.27158@news21.bellglobal.com>

Well, I bit the bullet and reinstalled VC5, and got SWIG to compile. Now, I
try "swig -c++ -python interf~1.cpp" and it chokes, hard. I try it with the
h, same thing.. I've tried variations of that, and I keep getting loads of
syntax errors, etc. Did it compile wrong? Is there some hidden syntax I
should be using? The other possibility is that since the files I'm swigging
are Borland C++ and SWIG itself was compiled with VC++.. But I can't see
that being a problem.

Any help?

Cheers,
Jason





From gerrit.holl at pobox.com  Tue Jan 25 10:27:35 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 25 Jan 2000 16:27:35 +0100
Subject: Python Rocks!
In-Reply-To: <000f01bf66cb$6861d720$a42d153f@tim>; from tim_one@email.msn.com on Mon, Jan 24, 2000 at 07:30:40PM -0500
References: <388BB4E1.974E63B7@prescod.net> <000f01bf66cb$6861d720$a42d153f@tim>
Message-ID: <20000125162735.B2497@stopcontact.palga.uucp>

Tim Peters wrote on 948738640:
> The option was offered to Guido years ago, via a patch by Donald Beaudry to
> do just that.  He declined then, and I doubt he'd change his mind now.
> 
> > Then people would have to stop complaining that Python is
> > broken
> 
> No, they wouldn't.  That's the whole problem here:  people with a bug up
> their butt about indentation are no different than people with a bug up
> their butt about integer division (like me!), or people who hate reference
> counting, or those who hate the lack of an address-of operator, or lack of
> lexical closures, or lack of an ISO std, or ... I once drew up a list of
> about 30 of these "crucial show-stoppers" culled from semi-hysterical c.l.py
> threads.

%{$answer: @($thats, $why, $I, $havent, $learnt, $perl)}

> When people say that Python's indentation drives away thousands of
> programmers, they're almost certainly correct!  And so do each of dozens of
> other things.  The same is also true of every other language on earth.
> Guido is on the receiving end of *all* these gripes, so has a sobering
> global perspective the proponents of individual features X, Y and Z lack in
> their single-minded zeal.  Over the years, and only after a great deal of
> thought and care, he occasionally "yields" on the few he judges most likely
> to add more power than cause harm.

Guido won't have World Dominatior. Neither will Larry.

> I don't know whether he keeps a formal priorty list of these things, but, if
> he does, optional block delimiters must be scraping the bottom of it (adds
> no power, at best leads to the usual flamewars about where the delimiters
> "should be" placed, at worst ruins the easy visual uniformity of Python code
> that so many old-timers (incl. esp. Guido!) actively love).
> 
> > and instead admit that Python programmers choose to
> > program without delimiters.
> 
> The existence and utter non-use of Guido's Tools/Scripts/pindent.py is
> sufficient proof of that.
> 
> that-was-indeed-his-last-word-on-the-subject-ly y'rs  - tim

but-the-last-word-in-c.l.py-will-never-be-ly y'rs
    -- Gerrit Holl

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From News at Titanic.co.uk  Thu Jan 13 02:21:26 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Thu, 13 Jan 2000 07:21:26 GMT
Subject: Text File
References: <cc0q7s8ggiqj62j27mcrtgg5a374qbm94h@4ax.com> <iacaemalvck.fsf@vier.idi.ntnu.no>
Message-ID: <r5vq7soihk782qms7cg1afemvo1f3ni95t@4ax.com>

Magic,

I am sure this will get me finished.

Now,  if you will come over and predict next weeks Lotto numbers my
life will be perfect!

Cheers

Simon


On 13 Jan 2000 00:32:27 +0100, mlh at vier.idi.ntnu.no (Magnus L.
Hetland) wrote:

>Simon Faulkner <news at titanic.co.uk> writes:
>
>> Hi All,
>> 
>> I would like to write a CGI program that diskplays a small (<1k) text
>> file on a web page and saves any changes made by the browser back to
>> the same file.
>> 
>> Any sugestions to help me get started?
>> 
>> Specifically
>> 
>> Which data type should I use and how do I read the text  file into
>> it?
>
>Just do:
>
>  file = open("filename.txt")
>  text = file.read()
>  file.close()
>
>And there it is - as a string. Put that into a text-area in your HTML
>output for the user to edit. When you get the result, you'll probably
>get it in a manner similar to this:
>
>  data = FieldStorage()
>  text = data["text"].value()
>
>  file = open("filename.txt","w")
>  file.write(text)
>  file.close()
>
>And there you are. Now all you need to do is build the CGI stuff and
>HTML around it :)
>
>> 
>> Cheers  
>> 
>> 
>> Simon
>> 

Simon Faulkner


From rupole at compaq.net  Sun Jan 16 21:44:29 2000
From: rupole at compaq.net (Roger Upole)
Date: Mon, 17 Jan 2000 02:44:29 GMT
Subject: win32pipe missing WaitNamedPiped ?
Message-ID: <gmvg4.6200$kv.435556@news.easynews.com>

I can't seem to find the WaitNamedPipe routine anywhere in the
win32pipe module.  Am I misspelling it or just looking in all the
wrong places ?  Any help greatly appreciated.




From dworkin at ccs.neu.edu  Fri Jan 14 11:25:31 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 14 Jan 2000 11:25:31 -0500
Subject: A Python bug? (core dump)
In-Reply-To: huaiyu_zhu@moon.localdomain's message of "13 Jan 2000 17:52:52 -0600"
References: <slrn87spbq.16i.huaiyu_zhu@localhost.localdomain>
Message-ID: <vndbt6oboxw.fsf@betelgeuse.ccs.neu.edu>

huaiyu_zhu at moon.localdomain (Huaiyu Zhu) writes:

> The following script dumps core

I can't get it to dump core under any of:

Python 1.5.2 (#1, Oct 22 1999, 16:54:21)  [GCC 2.7.2.3] on linux2
Python 1.5.2 (#3, Aug 24 1999, 15:55:20)  [GCC 2.7.2.1] on freebsd3
Python 1.5.2 (#3, Sep 30 1999, 10:35:10)  [GCC 2.8.1] on sunos5

On all of those, it produces the correct exception.

-Justin

 



From thomas at bibsyst.no  Thu Jan  6 13:23:38 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Thu, 06 Jan 2000 19:23:38 +0100
Subject: Reading tar/gz-files
Message-ID: <3874DDAA.E14CC23B@bibsyst.no>

hi,

Would it be possible to read out the contents of tar.gz/tgz files using
Python? Could I, using Linux, use catch the output of a os.exec('tar tvf
' + filename) for instance, in a list??

Tudelidu.

Thomas


From tim_one at email.msn.com  Tue Jan 25 20:55:17 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jan 2000 20:55:17 -0500
Subject: Python compiler?
In-Reply-To: <20000125163234.C2497@stopcontact.palga.uucp>
Message-ID: <001601bf67a0$651b9d00$862d153f@tim>

[J Dubal]
> Does [Python] have a compiler for Linux and/or windows that can
> produce native executables?

[Gerrit points to freeze, Tim explains that freeze inserts a copy
 of the interpreter]

[Gerrit Holl]
> Ah, so *that's* why it's so big!

In large part, yes.  If we could assume that "the normal" Python shared
libraries were available (as, e.g., compiled C programs routinely assume for
*their* libraries), freeze-like executables could be much smaller.  But
people asking for "native executables" are almost always really asking for
single-file solutions they can ship as-is to machines they know nothing
about.  After Python is installed everywhere (don't expect that before March
<wink>), we can skip stuffing in a copy of the Python .so/.dll core (which
consumes the bulk of the space).

> But isn't it possible to write a *compiler* for Python?

Looks like others have covered that.  They still need to access copies of
the normal Python runtime routines, though (e.g., for dict access and list
growing and method lookup and ...).

>> the-natives-are-restless-tonight-ly y'rs  - tim

> The natives?
> Native Pythoneers?

Had more in mind the "native executables".  Native Pythoneers are a very
calm lot <wink>.





From fredrik at pythonware.com  Thu Jan 20 07:26:09 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 20 Jan 2000 13:26:09 +0100
Subject: problems setting up PIL under Redhat 6.0??
References: <8658kv$nf3$1@nnrp1.deja.com>
Message-ID: <00af01bf6341$88a1ea80$f29b12c2@secret.pythonware.com>

charliehubbard at my-deja.com wrote:
> I'm trying to use PIL under redhat 6.0, but I can't seem to get it to
> work.  I go through the instructions for building everything, but when
> I try to import _imaging I get this error.
> 
> >>> import _imaging
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: /usr/lib/libtk8.0.so: undefined symbol: XFreePixmap
> 
> Any ideas?

the answer is to be found in the Image SIG archives:
http://www.python.org/pipermail/image-sig/1999-December/001852.html

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! &smiley; -->




From alex at somewhere.round.here  Fri Jan  7 21:25:20 2000
From: alex at somewhere.round.here (Alex)
Date: 07 Jan 2000 21:25:20 -0500
Subject: Python friendly emacs
References: <3876056F.36E4335D@spamfree.clacknet.com>
Message-ID: <etd7lhl47vz.fsf@w20-575-109.mit.edu>

> I know that python comes with a .el file for emacs but I am unable to
> make emacs understand it.

What error messages is it giving?  Have you turned font-locking on?

> Also, I'd like emacs to default to python mode when it edits a file
> with the extension of .cgi

Put the following in your ~/.emacs file:

(setq auto-mode-alist
      (cons '("\\.cgi$" . python-mode) auto-mode-alist)) 

(autoload 'python-mode "python-mode")
(autoload 'py-shell    "python-mode")

Also, make sure that Emacs knows where to find python-mode:

(setq load-path (append '("directory containing python-mode")
                        load-path))

Alex.


From rickhigh at aol.com  Tue Jan  4 04:36:40 2000
From: rickhigh at aol.com (RICKHIGH)
Date: 04 Jan 2000 09:36:40 GMT
Subject: Java Applet woes and JPythonC...When did JPythonC last work?
Message-ID: <20000104043640.13218.00000284@ng-fk1.aol.com>

Subj: Applet woes and JPythonC...When did JPythonC last work?
First of all, it is 2:00 AM and I am sorry if at times I don't make sense.

Assertion: JPythonC does not work for JPython version 1.1.beta4.

I checked the jitterbug and did not see this bug listed. But I vaguely remember
someone mentioning something like this before.

If there is a patch, please let me know.
If there is a way to get a hold of an older version of JPython and JPythonC
that works, please let me know.

Here is the problem.
I am having trouble running an Applet after I compile. I am using
JPython1.1beta4. 

I was having troubles porting an JPython application to a Java Applet....SO...I
compiled a real simple Applet called MyApplet with the following options.

jpythonc --deep --package com.mycompany.mypackage --jar myapplet.jar
MyApplet.py
(BTW I tried --core but it would not work.)

The above compiles the Applet okay.

However, the Applet will not run in the AppletViewer. The AppletViewer said it
could not find some class files in jpython.jar (though jpython.jar is in my
class file and the archive APPLET param-[Yes I know that the AppletViewer
ignores the archive parameter
]).
I tried putting jpython.jar in runtime ext directory. Problems changed but did
not go away---I got a NullPointer Exception (more on this later.)

Then, I tried running the Applet in Internet Explorer (IE).
IE showed that there was a security violation as follows:

at java/io/File.exists (File.java)
	at org/python/core/PySystemState.addRegistryFile (PySystemState.java:255)
	at org/python/core/PySystemState.initRegistry (PySystemState.java:235)
	at org/python/core/PySystemState.initialize (PySystemState.java:295)
	at org/python/core/Py.initProperties (Py.java:564)
	at org/python/core/Py.initProxy (Py.java:596)
	at com/mycompany/mypackage/MyApplet.<init> (MyApplet.java:167)

It looks like the class generated is trying to read the JPython registry. Which
is a bad thing for an Applet to do. :o) (Because of the security sandbox.)

Well, I figured I will run the Applet as a standalone application until there
is a patch or the final release of JPython is released end of Jan.

But I am not out of the woods yet!

So I created a real simple Java class, just to see if I could instantiate an
instance of com.mycompany.mypackage.MyApplet.

import com.mycompany.mypackage.MyApplet;

public class Test{

public static void main(String args[]){
	MyApplet applet = new MyApplet();
	System.out.println(applet.getClass().getName());

}
}

When I compile and run the above, I get the following error:

Exception in thread "main" java.lang.NullPointerException
        at java.util.Hashtable.put(Compiled Code)
        at org.python.core.Py.initProperties(Compiled Code)
        at org.python.core.Py.initProxy(Compiled Code)
        at com.mycompany.mypackage.MyApplet.<init>(MyApplet.java:170)
        at Test.main(Test.java:6)

Now it appears that the MyApplet will not even run as a standalone class. Thus,
it appears that JPythonC is not compiling classes.

I know I have compiled classes with a previous version of JPython.
And I know I have compiled Applets with a previous version of JPython.
I don't remember which version.

I checked the jitterbug and did not see this bug listed. But I vaguely remember
someone mentioning something like this before.

If there is a patch, please let me know.
If there is a way to get a hold of an older version of JPython and JPythonC
that works, please let me know.

Here is the listing for MyApplet.py module.

from java.applet import Applet
from java.awt import List

	# Define a Python class that subclasses the Applet class 
	# (java.applet.Applet)
class MyApplet (Applet):
	def __init__(self):
		self.list = List()
		self.add(self.list)
	def init(self):
		self.list.add("Init called")
	def destroy(self):
		self.list.add("Destroy called")
	def start(self):
		self.list.add("Start called")
	def stop(self):
		self.list.add("Stop called")

if __name__ == "__main__":
	from FakeBrowser import FakeBrowser
	fakeBrowser = FakeBrowser()
	fakeBrowser.addApplet(MyApplet())

FakeBrowser is just a frame that treats MyApplet like a Panel.
It's just a way to show an Applet running standalone.

--Rick Hightower

BTW I wrote a really interesting utility to run JPython applets for debugging.
The utility implements an AppletContext and an AppletStub. It builds a list of
URLs based on the archive tag and the codebase tag. Then it uses
java.net.URLClassLoader to load classes across the wire--just like a browser. 
You can specify all of the props in a properties file and then just add an
Applet to the utility--just like FakeBrowser above, but unlike fakeBrowser it
passes an AppletContext and an AppletStub to the applet. It even generates an
HTML file based on the params that you specify. It is great for debugging an
Applet written in Jpython. Of course the utility is written in JPython. Let me
know if you are interested and I will send you the utility. It is all in one
module. Again, it is great for debugging a JPython Applet or just deploying an
Applet as a standalone application.



From che at debian.org  Mon Jan 10 12:51:24 2000
From: che at debian.org (Ben Gertzfield)
Date: 10 Jan 2000 09:51:24 -0800
Subject: Installing .py modules with Setup.in and Makefile.pre.in
Message-ID: <yttwvphizmr.fsf@gilgamesh.cse.ucsc.edu>

Hey, folks. I'm trying to create a module that will install both
a shared library Python module built from C sources *and* a few
pure-Python modules at once.

The documentation doesn't seem to say how, and I can't find any
examples of what to put into Setup.in to have it install both C-built
modules *and* pure-Python modules. Here's what I've got now, for just
the C modules:

# Setup.in file for the CDDB Python module
# Please uncomment the lines that pertain to your OS!

*shared*

# For Win32, you'll just need to copy the files 'win32/mci.dll' and
# 'win32/cdrom.py' to your Python module path by hand, or re-build
# the mci.dll file using the mci.dsp project file.

# Linux module
cdrom linux/cdrommodule.c -DLINUX

# Solaris module
# cdrom linux/cdrommodule.c -DSOLARIS

*snip*

I've tried just putting the names of my python modules in on separate
lines, but then the Makefile tries to build them as .so shared
libraries containing no object files. :)

Is there any way to do this? Do I have to just tell folks to copy the
resultant .so library and all the .py files to their site-python
directory?

For what it's worth, my module's at:

http://csl.cse.ucsc.edu/~ben/python/cddb.html

I'm also interested in knowing if the "guess-and-uncomment" method I'm
using above in the Setup.in is the best way to detect different
operating systems. The code that I'm running really needs to know what
OS it's on, but I'm not sure how best to detect that, so I just went
with the good old "make the user uncomment their own darned lines"
method. :)

And finally, what to do about the win32 users? I've provided a
'makefile' of sorts (a Microsoft Developer Studio project file) but I
doubt this will play well with the Makefile.pre.in style of
building. For now, I'm just figuring I'll include both the binary and
source of the win32 version of the module, and let the users copy
them by hand. (yuck!)

Thanks,

Ben

-- 
Brought to you by the letters H and G and the number 4.
"A squib is a firecracker."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/


From family.cavaille at libertysurf.fr  Tue Jan 25 15:15:49 2000
From: family.cavaille at libertysurf.fr (family cavaille)
Date: Tue, 25 Jan 2000 21:15:49 +0100
Subject: Hi, a bug python , is it possible ? I've found a bug with the syntaxic analyser.
Message-ID: <388E0475.312F79BB@libertysurf.fr>

Hello, I have a trouble with this code:

------------
class Bug:
  def __init__(self):
    print "My beautiful class don't  bug"

print "Hello"
del Bug()
------------

It's all,(i use python 1.5.1 on Linux, AIX, & NT) and I've an error
But I use a temporal variable, (I  replace  'del Bug()' by 'x=Bug();del
x') It's Ok.
For my buggous code, I have a strange comportement, because my code
don't print Hello, and it's anormal  (print it's before the del
statement).
=> My Conclusion:
del Bug() breaks the syntax analyse, -> the code is invalid -> I've not
the print statement
Imho del "thinks" its argument is a function (because, the () ) and not
an instance

Have you an another interpretation ?

JCC

Please don't flam me, I'm an VIP  ;-)
a Very Innocent Person, and it's not my fault, ...




From R.Brodie at rl.ac.uk  Tue Jan 25 06:38:41 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Tue, 25 Jan 2000 11:38:41 -0000
Subject: Whitespace delimiters suck - Seen Occam?
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <86cmom$5pp$1@news-int.gatech.edu> <86h7jm$o7i@newton.cc.rl.ac.uk> <388C760A.1E66C20@gtri.gatech.edu>
Message-ID: <86k8km$11p6@newton.cc.rl.ac.uk>

"Andrew Henshaw" <andrew.henshaw at gtri.gatech.edu> wrote in message
news:388C760A.1E66C20 at gtri.gatech.edu...

> That's interesting.  Everyone here, who has used it for any
> reasonably-sized project, has really liked it.  I've seen opposition
> from programmer's who try it for a day and then immediately
> abandon it, however.

You are probably right, in that the projects were relatively small,
and programmers didn't stick to it for long enough.

> Have you tried the Windows version of Origami from Elcom (it adds several
> niceties, but the default color scheme is horrible)?

Windows? Colours? Linux? We stopped using Occam in about '92,
so the state of the art might have improved since then ;) I can see
a folding editor would be better if you had a GUI to drive it from.





From akuchlin at mems-exchange.org  Sat Jan 22 16:23:23 2000
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 22 Jan 2000 16:23:23 -0500
Subject: Python ODP category
Message-ID: <3dg0vpby2c.fsf@amarok.cnri.reston.va.us>

Looking at the Open Directory Project's category for Python, it could
use some tidying, and no one has edited it in a while.  Take a look
at:

http://www.dmoz.org/Computers/Programming/Languages/Python/

I'm listed as an editor, but only really joined to get info about the
ODP's RDF dumps, and have only done 1 or 2 edits to the category since
joining.  Looking at the logs, "cyberjet", the other editor, hasn't
made any edits in some months; other people responsible for higher
level categories such as 'Languages' have made a few edits, but it's
really not being maintained.

So, what should be done with the category to improve it?  I don't
think anyone is available to shoulder the maintenance burden of
keeping a very complete resource; for example, 'Python/Modules' could
duplicate much of Parnassus, but what would be the point?

Suggestions?  (I have a list of suggestions which I won't post at the
moment; first see what you all think...)

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
That jackanapes! All he ever does is cause trouble.
    -- The Doctor talking about the Master, in "Terror of the Autons"




From gmcm at hypernet.com  Sun Jan  9 19:18:00 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 9 Jan 2000 19:18:00 -0500
Subject: no return values for __init__ ??
In-Reply-To: <3879074F.1A4EDED3@mdlink.de>
Message-ID: <1264659013-22815850@hypernet.com>

Helge Hess wrote:
> Aahz Maruch wrote:
> > >Hm, I wonder why it is simpler to have two separate kinds of
> > >methods instead of a consistent behaviour for all methods ?
> > 
> > Because we're looking at it from two different perspectives. 
> > You're looking at the process from the bottom and I'm looking
> > at it from the top.  Providing your capability would make it
> > too easy for an inexperienced programmer to screw things up
> > royally.
> 
> I can't see on what you base this statement on ?! This capability
> generalizes, not specializes, the environment - which leads me to
> the assumption that the *current* state will make programmers
> wonder.
> 
> Until now I haven't heard a reason against this 'feature' which I
> can follow. I guess that it is just a matter of taste.

Is inheritance a matter of taste?
 
You claim to have patched your interpreter. Try this:

class A:
  def __init__(self):
    pass

class B:
  def __init__(self):
    return A()

class C(B):
  def __init__(self):
    self = B.__init__(self)
  def method(self):
    print "C.method called"

c = C()
c.method()

Produces an AttributeError, doesn't it?

- Gordon



From culliton at clark.net  Wed Jan 12 09:29:41 2000
From: culliton at clark.net (Tom Culliton)
Date: Wed, 12 Jan 2000 14:29:41 GMT
Subject: Perl speed vs. Python speed
References: <85hjfa$mlp$1@nnrp1.deja.com>
Message-ID: <pd0f4.26480$W2.358425@iad-read.news.verio.net>

In article <85hjfa$mlp$1 at nnrp1.deja.com>,  <rwgk at my-deja.com> wrote:
>I just wrote a tiny Python program to do some text processing.
>When I ran it I was very surprised how long it took to finish.
>My first guess is that Perl would be a lot faster, maybe even a
>factor 10, but I did not do a direct comparison.
>
>Could someone with more experience comment on Perl speed vs.
>Python speed?

Naively written Python is often not as fast as Perl for processing
large quantities of text.  On the other hand you'll still be be able
to read and understand the Python 6 months from now, and by using some
common optimizaions it you can be speed up considerably while still
remaining readable.

Things to look out for:

Reading a single line at a time in Python is not as fast as it could
be.  Using readlines with or without a buffer size argument to limit
memory usage can make a huge difference.

Function calls and name look up outside the local scope are relatively
expensive.  Using the localization trick can make an improvement of
several percent per localized variable, as can moving function code
in line.

The "new" Python re module needs to be optimized and possibly pushed
into C code, the older regex module is about twice as fast.

Precomputing values like bound regex search or match functions can
also help quite a bit.

Python can spend a lot of time allocating and freeing memory if you
get sloppy with your string manipulations.

There are other tricks but this is just supposed to be a quick answer.
Using the profiler is often very instructive.  Posting some code might
also get you more examples.

Bottom line, Perl does nasty incestuous things with i/o and regular
expressions to make them very fast.  Python does less magic, which
means that it leaves more room for you to do things like provide
pseudo file objects with special behaviour (like the StringIO module),
or replace the regular expression module (regex vs. re).


From cjw at connection.com  Sun Jan 23 08:37:14 2000
From: cjw at connection.com (Colin J. Williams)
Date: Sun, 23 Jan 2000 08:37:14 -0500
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de> <869uel$suj$1@nntp1.atl.mindspring.net> <slrn88hnpb.brp.quinn@auto.ugcs.caltech.edu>
Message-ID: <388B040A.F7A15B9C@connection.com>

I recognize that most of this thread deals with deeper issues,
see the later posting by Tim Peters, but I would like to focus
on the list and one of its operations.

Quinn Dunkan wrote:
> 
> A curried function always takes one argument.  If it needs more values to
> 'finish', it returns a closure which will accept the next arg.  In python:
> 
> >>> def add(a):
> ...    return lambda b, a=a: a + b
> ...
> >>> add(4)
> <function <lambda> at 100b41f8>
> >>> add(4)(5)
> 9
> 
> A simple example of how this can be useful:
> 
> >>> map(add(5), range(5))
> [5, 6, 7, 8, 9]
> 
Wouldn't it be nice if we could write this as:

              5 + range(5) or 
              range(5) + 5?

Are we not complicating things by using '+' to mean addition in
most cases but catenation in the case of a list?

Colin W.

> Many functional languages such as the ml family and haskell have all functions
> curried by default, including +, *, etc. (there are no two-argument
> functions... of course, you could pass a _single_ tuple of two elements :))
> 
> Hugs> map (5+) [0..4]
> [5,6,7,8,9]
> 
> Whenever you know some args for a function now, and the rest later, you can
> partially apply it and toss the closure around.  It's very useful if your
> brain is working in the appropriate mode.




From doe at nowhere.com  Tue Jan 18 20:44:41 2000
From: doe at nowhere.com (Ralf Canis)
Date: Wed, 19 Jan 2000 02:44:41 +0100
Subject: Win32 BSDDB
References: <20000118120921.020261@artemis.dnastar.com>
Message-ID: <gk5a8ssjrp8cg7qctbgd8m5smhree7kk66@4ax.com>

On Tue, 18 Jan 2000 12:09:21 -0600, chriss at dnastar.com wrote:

>
>Perhaps i'm missing something, but where dose the source for Python-Win32's
>BSD-DB come from? That is, the data base itsself, not python's adaptors.

Berkeley DB library: http://www.sleepycat.com



From billtut at microsoft.com  Thu Jan 13 15:37:27 2000
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 13 Jan 2000 12:37:27 -0800
Subject: Win32: using PySECURITY_DESCRIPTOR Object
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101D8BCC03@RED-MSG-50>

Well, I know how to deal with a SECURITY_DESCRIPTOR in C++, but I'm afraid
win32security doesn't expose enough Win32 APIs to help you out in this
regard. Surprisingly enough I don't have the time atm to go about fixing
this limitation. :(

I can provide help in explaining how they work if you have time to make the
changes to win32security and can't fathom the MSDN docs on how this
complicated structure works.

Bill

> -----Original Message-----
> From: Christian Schaller [mailto:Christian.Schaller at mchp.siemens.de]
> Sent: Thursday, January 13, 2000 9:25 AM
> To: python-list at python.org
> Subject: Win32: using PySECURITY_DESCRIPTOR Object
> 
> 
> Hello...
> 
>   Does anyone know how to use a PySECURITY_DESCRIPTOR object 
> in Python? 
> I was reading file security settings via:
> >>> import win32security
> >>> res = win32security.GetFileSecurity('c:\', \
>     win32security.DACL_SECURITY_INFORMATION)
> 
>   How can I print all values contained in res?  Please help me!
> 
>   Thanks a lot in advance!
> 
> bye
>   Chris...
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From hildeb at www.stahl.bau.tu-bs.de  Fri Jan 21 09:51:17 2000
From: hildeb at www.stahl.bau.tu-bs.de (Ralf Hildebrandt)
Date: 21 Jan 2000 14:51:17 GMT
Subject: os._exit(), os.kill(os.getpid(), 9)
References: <20000121145429.A4320@stopcontact.palga.uucp>
Message-ID: <slrn88gsj5.77g.hildeb@stahlw06.stahl.bau.tu-bs.de>

On Fri, 21 Jan 2000 14:54:29 +0100, Gerrit Holl <gerrit.holl at pobox.com> wrote:

>os._exit(127)

This exits with exit code 127 and doesn't clean up...

>os.kill(os.getpid(), 9)

Rather make that
os.kill(os.getpid(), signal.SIGKILL)

This makes the interpreter kill itself with SIGKILL. The exit code is
undefined.

Ergo those two are NOT equivalent.


From erno at iki.fi  Sun Jan 23 22:06:17 2000
From: erno at iki.fi (Erno Kuusela)
Date: 24 Jan 2000 05:06:17 +0200
Subject: Using glade to build guis
References: <38845051.5A9A1A49@bibsyst.no> <p6q3drsdi4j.fsf@informatik.hu-berlin.de> <388714F6.3B0BE7B4@bibsyst.no>
Message-ID: <87iu0ki2xi.fsf@fun77.kivikko.hoas.net>

there is a short tutorial written by Deirdre Saoirse at
<URL:http://www.baypiggies.org/10mintig.html>.

the mailing list archive can be found at
<URL:http://www.mail-archive.com/pygtk at daa.com.au/>

  -- erno


From schliep at Octopussy.MI.Uni-Koeln.DE  Thu Jan 27 11:09:29 2000
From: schliep at Octopussy.MI.Uni-Koeln.DE (Alexander Schliep)
Date: 27 Jan 2000 17:09:29 +0100
Subject: Any graph editors?
References: <86nigr$724$1@pollux.ip-plus.net>
Message-ID: <tq9oga7cx8m.fsf@Twister.MI.Uni-Koeln.DE>

lin li <linli at computer.org> writes:
> scratch. So here is the question: is there any implementation of
> some graph editors in Python, JPython (or maybe Java) that I can
> copy and modify?

Dear Lin,

Gato <http://www.zpr.uni-koeln.de/~gato> contains Gred -- the GRaph EDitor
-- which is written in Python/Tkinter. The whole thing is released under
the LGPL.

Yours,
Alexander

-- 
Alexander Schliep                    schliep at zpr.uni-koeln.de
ZPR/ZAIK                             Tel: +49-221-470-6040 (w) 
University of Cologne                FAX: +49-221-470-5160
Weyertal 80                          http://www.zpr.uni-koeln.de/~schliep
50931 Cologne, Germany               


From breiter at usf.Uni-Osnabrueck.DE  Wed Jan 12 07:26:02 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 12 Jan 2000 12:26:02 GMT
Subject: XML-RPC Server
References: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de>
    <slrn87ne3i.qla.kc5tja@garnet.armored.net>
Message-ID: <85hrsq$4t9$1@newsserver.rrzn.uni-hannover.de>

In article <slrn87ne3i.qla.kc5tja at garnet.armored.net>,
	kc5tja at garnet.armored.net (Samuel A. Falvo II) writes:
>>I know it should be hard to do, someone must have done it.
>>Just a couple of simple objects answering towards XML-RPC requests.
> 
> In fact, it's insanely hard to do.  I can't tell you the tears I had to go
> through just to download the dang thing.  ;-)

A type on my part of course. It should have read "not hard to do".  :)

> Look for the "xmlrpclib" package for Python.  
Found it in there minutes after posting.  Thanks for the resonse.

	Bernhard
-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From mlh at idi.ntnu.no  Sun Jan 23 11:04:20 2000
From: mlh at idi.ntnu.no (Magnus Lie Hetland)
Date: Sun, 23 Jan 2000 17:04:20 +0100
Subject: Stackless Python 1.0 + Continuations 0.6
References: <mt2.0-27365-948450827@news.informatik.uni-bonn.de>
Message-ID: <86f8uu$3ds$1@kopp.stud.ntnu.no>

Cool!

:)

Without *really* having read all the documentation, I venture a question...
By using S.P. and continuations etc. -
would it be possible to stop and pickle a thread? This might be very
interesting for Mobile Agent Technology
etc. (I.e. a running program can migrate over a network and resume its
execution etc...)

--

  Magnus
  Lie
  Hetland                                mailto:<anything>@hetland.org





From thantos at chancel.org  Wed Jan 26 18:15:05 2000
From: thantos at chancel.org (Alexander Williams)
Date: Wed, 26 Jan 2000 23:15:05 GMT
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com>
Message-ID: <slrn88uvvp.935.thantos@brimstone.mecha>

Personally, I found Sweeny's article to be interesting in content but,
in the end analysis, blind to the facts.  Interestingly, he discards
the entire class of functional programming languages within the first
two paragraphs, then summarises his article with the 'needed
facilities' the industry will need in the next decade ... which,
perversely enough, /already exist and have been used successfully for
years/ in the first group of languages he threw out.

I'm tempted to create a new term: Ivory Tower Commercialia, meaning
the tendency for Corp programmers to believe /everything/ must be like
Java to be worthwhile ...

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From thor at localhost.localdomain  Wed Jan 26 18:36:14 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 26 Jan 2000 23:36:14 GMT
Subject: Interesting article on programming languages
References: <388E7D5B.7840983F@erols.com> <slrn88uvvp.935.thantos@brimstone.mecha>
Message-ID: <slrn88v4ou.93j.thor@localhost.localdomain>

On Wed, 26 Jan Alexander Williams <thantos at chancel.org> wrote:
>Personally, I found Sweeny's article to be interesting in content but,
>in the end analysis, blind to the facts.  Interestingly, he discards
>the entire class of functional programming languages within the first
>two paragraphs, then summarises his article with the 'needed
>facilities' the industry will need in the next decade ... which,
>perversely enough, /already exist and have been used successfully for
>years/ in the first group of languages he threw out.
>
>I'm tempted to create a new term: Ivory Tower Commercialia, meaning
>the tendency for Corp programmers to believe /everything/ must be like
>Java to be worthwhile ...
>

I started to read that article, but I got bored too soon. It's 
a bit amazing how succesful programming people ( usually in USA )
have so poor visions of Information technologies. Anyone with 
a bit of common sense realize that in fact there're two main
broad ways of handling things:

- the rigorous/mathematical one : This is the way of Haskell or
Design by contract of Eiffel. This one is like wearing a XV century
armor, you walk very safely but in a very tiring way. Anyway
it's the best way because it's the safest one. This way will tend
to be more and more used because it's systematic and proffessional
and low level things will be handled by C libraries.

- the scripting way: python as the maximum representative. Fast,
very fast design, but not very clearly defined in terms of types
and if every function what is intended to do.  This way has 
all the benefit the other lacks: fast and clear vision (high ) of the
algorithms and flows of data . 

Anyone should respect both ways because both are good and both
are necessary, first python, then Haskell to put in a rigorous
way. Functional programming is not "theorical" , it's just 
the __ultimate__ way of expressing things, once that things
have been sketched with python.


Using Java or C++ is just to mix both approaches and getting no
benefit of any.


-- 
MGA


From tseaver at aftershock.neosoft.com  Mon Jan  3 13:09:55 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 3 Jan 2000 12:09:55 -0600
Subject: Python on NT systems (help a Unix guy, please)
References: <19991218113757.A637@Ridcully.home> <3870CA2B.DD09F50E@bellatlantic.net>
Message-ID: <779C8E5E83B84998.EEDA50CBCDE89256.7932A76752E8633A@lp.airnews.net>

In article <3870CA2B.DD09F50E at bellatlantic.net>,
Steve Holden  <sholden at BellAtlantic.net> wrote:
>Don't suppose anyone has implemented freeze() on NT, have they?
>I only have the binary 1.5.2 on my NT box, and I don't see it
>there :-(

You want Gordon McMillan's installer for this:

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

Found it via a "parnassus" search:  http://www.vex.net/~x/parnassus

Parnassus-is-your-friend'ly

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From gherman at darwin.in-berlin.de  Fri Jan 28 08:25:37 2000
From: gherman at darwin.in-berlin.de (Dinu C. Gherman)
Date: Fri, 28 Jan 2000 14:25:37 +0100
Subject: Zope 2 on Starship, when?!
Message-ID: <389198D1.8C0A0BCB@darwin.in-berlin.de>

At IPC8 I've asked several people when we'll finally have some
working version of Zope 2 running on the starship with a tiny
bit of information telling people what to do in order to use 
it?

As I didn't get much of a satisfactory answer I'll repeat the
question again. Is anybody capable of answering it (maybe some
Starship webmaster or some Digital Creation people)?

Or maybe I got something wrong and people are supposed to use
a public Zope 2 installation on the Zope.org site (if availa-
ble)? 

Regards,

Dinu

-- 
Dinu C. Gherman
................................................................
Je me revolte, donc je suis. (Albert Camus)


From gerrit.holl at pobox.com  Sat Jan 29 08:34:47 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sat, 29 Jan 2000 14:34:47 +0100
Subject: statements and blocks [LONG]
In-Reply-To: <86tt56$d4a$1@nnrp1.deja.com>; from rdudfield@my-deja.com on Sat, Jan 29, 2000 at 05:17:29AM +0000
References: <86t0ov$4en$1@news.psc.edu> <86tt56$d4a$1@nnrp1.deja.com>
Message-ID: <20000129143447.B2146@stopcontact.palga.uucp>

rdudfield at my-deja.com wrote on 949119449:
> Speaking of blocks, is there a way to show where the block ends?
> 
> I have been caught a few times doing this:
> class b():
>   def asdf():

This is invalid, BTW...

class b:
  def asdf(self):
...

>     print 'a'
>     if bla:
>       print 'a'
>       print 'a'
>       ... more than a page of stuff.
> 
>    # method is supposed to be in the classes scope.
>     def qwer():
>       print 'a'
>       print 'a'
>       print 'a'
> 
> 
> Basically because the function asdf is more than a page long.
> 
> I find indentation is better in most cases except this.  This isn't too
> much trouble, as I usually find the problem pretty quickly.  However
> occasionally I waste a bit too much time on it than I'd like to :)

Indent more:

class Henk:
        def walk(self, step):
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...
                ...something very complicated...

        def sleep(self):
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...
                ...something less complicated...

That's a lot more clear than:

class Henk:
  def walk(self, step):
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...
    ...something very complicated...

  def sleep(self):
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...
    ...something less complicated...

If your lines become > 80 characters long, split up your code.

Linus torvalds tells about kernel coding style, but this also counts
in general.

		Linux kernel coding style 

This is a short document describing the preferred coding style for the
linux kernel.  Coding style is very personal, and I won't _force_ my
views on anybody, but this is what goes for anything that I have to be
able to maintain, and I'd prefer it for most other things too.  Please
at least consider the points made here. 

First off, I'd suggest printing out a copy of the GNU coding standards,
and NOT read it.  Burn them, it's a great symbolic gesture. 

Anyway, here goes:


	 	Chapter 1: Indentation

Tabs are 8 characters, and thus indentations are also 8 characters. 
There are heretic movements that try to make indentations 4 (or even 2!)
characters deep, and that is akin to trying to define the value of PI to
be 3. 

Rationale: The whole idea behind indentation is to clearly define where
a block of control starts and ends.  Especially when you've been looking
at your screen for 20 straight hours, you'll find it a lot easier to see
how the indentation works if you have large indentations. 

Now, some people will claim that having 8-character indentations makes
the code move too far to the right, and makes it hard to read on a
80-character terminal screen.  The answer to that is that if you need
more than 3 levels of indentation, you're screwed anyway, and should fix
your program. 

In short, 8-char indents make things easier to read, and have the added
benefit of warning you when you're nesting your functions too deep. 
Heed that warning. 


		Chapter 2: Placing Braces

[offtopic] :-)

		Chapter 3: Naming

C is a Spartan language, and so should your naming be.  Unlike Modula-2
and Pascal programmers, C programmers do not use cute names like
ThisVariableIsATemporaryCounter.  A C programmer would call that
variable "tmp", which is much easier to write, and not the least more
difficult to understand. 

HOWEVER, while mixed-case names are frowned upon, descriptive names for
global variables are a must.  To call a global function "foo" is a
shooting offense. 

GLOBAL variables (to be used only if you _really_ need them) need to
have descriptive names, as do global functions.  If you have a function
that counts the number of active users, you should call that
"count_active_users()" or similar, you should _not_ call it "cntusr()". 

Encoding the type of a function into the name (so-called Hungarian
notation) is brain damaged - the compiler knows the types anyway and can
check those, and it only confuses the programmer.  No wonder MicroSoft
makes buggy programs. 

LOCAL variable names should be short, and to the point.  If you have
some random integer loop counter, it should probably be called "i". 
Calling it "loop_counter" is non-productive, if there is no chance of it
being mis-understood.  Similarly, "tmp" can be just about any type of
variable that is used to hold a temporary value. 

If you are afraid to mix up your local variable names, you have another
problem, which is called the function-growth-hormone-imbalance syndrome. 
See next chapter. 

		
		Chapter 4: Functions

Functions should be short and sweet, and do just one thing.  They should
fit on one or two screenfuls of text (the ISO/ANSI screen size is 80x24,
as we all know), and do one thing and do that well. 

The maximum length of a function is inversely proportional to the
complexity and indentation level of that function.  So, if you have a
conceptually simple function that is just one long (but simple)
case-statement, where you have to do lots of small things for a lot of
different cases, it's OK to have a longer function. 

However, if you have a complex function, and you suspect that a
less-than-gifted first-year high-school student might not even
understand what the function is all about, you should adhere to the
maximum limits all the more closely.  Use helper functions with
descriptive names (you can ask the compiler to in-line them if you think
it's performance-critical, and it will probably do a better job of it
that you would have done). 

Another measure of the function is the number of local variables.  They
shouldn't exceed 5-10, or you're doing something wrong.  Re-think the
function, and split it into smaller pieces.  A human brain can
generally easily keep track of about 7 different things, anything more
and it gets confused.  You know you're brilliant, but maybe you'd like
to understand what you did 2 weeks from now. 


		Chapter 5: Commenting

Comments are good, but there is also a danger of over-commenting.  NEVER
try to explain HOW your code works in a comment: it's much better to
write the code so that the _working_ is obvious, and it's a waste of
time to explain badly written code. 

Generally, you want your comments to tell WHAT your code does, not HOW. 
Also, try to avoid putting comments inside a function body: if the
function is so complex that you need to separately comment parts of it,
you should probably go back to chapter 4 for a while.  You can make
small comments to note or warn about something particularly clever (or
ugly), but try to avoid excess.  Instead, put the comments at the head
of the function, telling people what it does, and possibly WHY it does
it. 


		Chapter 6: You've made a mess of it

That's OK, we all do.  You've probably been told by your long-time Unix
user helper that "GNU emacs" automatically formats the C sources for
you, and you've noticed that yes, it does do that, but the defaults it
uses are less than desirable (in fact, they are worse than random
typing - a infinite number of monkeys typing into GNU emacs would never
make a good program). 

So, you can either get rid of GNU emacs, or change it to use saner
values.  To do the latter, you can stick the following in your .emacs file:

(defun linux-c-mode ()
  "C mode with adjusted defaults for use with the Linux kernel."
  (interactive)
  (c-mode)
  (c-set-style "K&R")
  (setq c-basic-offset 8))

This will define the M-x linux-c-mode command.  When hacking on a
module, if you put the string -*- linux-c -*- somewhere on the first
two lines, this mode will be automatically invoked. Also, you may want
to add

(setq auto-mode-alist (cons '("/usr/src/linux.*/.*\\.[ch]$" . linux-c-mode)
                       auto-mode-alist))

to your .emacs file if you want to have linux-c-mode switched on
automagically when you edit source files under /usr/src/linux.

But even if you fail in getting emacs to do sane formatting, not
everything is lost: use "indent".

Now, again, GNU indent has the same brain dead settings that GNU emacs
has, which is why you need to give it a few command line options. 
However, that's not too bad, because even the makers of GNU indent
recognize the authority of K&R (the GNU people aren't evil, they are
just severely misguided in this matter), so you just give indent the
options "-kr -i8" (stands for "K&R, 8 character indents"). 

"indent" has a lot of options, and especially when it comes to comment
re-formatting you may want to take a look at the manual page.  But
remember: "indent" is not a fix for bad programming. 

### CUT HERE ###

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From janssen at parc.xerox.com  Fri Jan  7 19:44:55 2000
From: janssen at parc.xerox.com (Bill Janssen)
Date: Fri, 7 Jan 2000 16:44:55 PST
Subject: Python friendly emacs 
In-Reply-To: Your message of "Fri, 07 Jan 2000 15:21:47 PST."
             <3876056F.36E4335D@spamfree.clacknet.com> 
Message-ID: <00Jan7.164342pst."3446"@watson.parc.xerox.com>

Daniel, I run GNU Emacs 20.4 under X, and have no problem with using
Python.  I use python-mode.el, which you can get from www.python.org.
You'll need to put the file in a directory which is on your Emacs
load-path, and add

  (autoload 'python-mode "python-mode" nil t)
  (setq auto-mode-alist (cons '("\\.py" . python-mode) auto-mode-alist))

to your ~/.emacs file.

To get Emacs to default to python mode when it edits a file with the
extension of .cgi, just add

  (setq auto-mode-alist (cons '("\\.cgi" . python-mode) auto-mode-alist))

to your ~/.emacs file.

Bill



From winikoff at cs.mu.oz.au  Thu Jan  6 21:35:13 2000
From: winikoff at cs.mu.oz.au (Michael David WINIKOFF)
Date: 7 Jan 2000 02:35:13 GMT
Subject: Help with sockets!
Message-ID: <853jd1$etf$1@mulga.cs.mu.OZ.AU>

Hi!

I've started tinkering with sockets in Python and have the following
modified version of the example program in the library docs.

This is intended to forward everything from a port on one machine to
a telnet port on another machine.

The program works fine except that when I terminate the connection the
socket remains in use:

prompt% ./redir
Listening for connections on X port 60016
Connected by ('X.X.X.X', 62268)
Forwarding to 23 on X
Closing connections. Bye.
prompt%
prompt% ./redir
Listening for connections on X port 60016
Traceback (innermost last):
  File "./redir", line 16, in ?
    s.bind(FORWARD, PORT)
socket.error: (125, 'Address already in use')

Any ideas why this isn't working? The sockets are being closed before
the program terminates.

Thanks,

Michael


--- CUT HERE ---
#!/usr/local/bin/python
# Echo server (p143 Python lib man)
from socket import *
import select
# from select import *

def sock_avail(x):
        """Test whether data is available on the socket."""
        return select.select([x], [], [], 0) == ([x], [], [])

FORWARD = 'host1'
DEST = 'host2'
PORT = 60016
print 'Listening for connections on', FORWARD, 'port', PORT
s = socket(AF_INET, SOCK_STREAM)
s.bind(FORWARD, PORT)
s.listen(1)
conn, addr = s.accept()
print 'Connected by', addr
try:
        s1 = socket(AF_INET, SOCK_STREAM)
        s1.connect(DEST,23) # telnet port
        print 'Forwarding to 23 on', DEST
        while 1:
                if sock_avail(conn):
                        data = conn.recv(1024)
                        s1.send(data)
                if sock_avail(s1):
                        data = s1.recv(1024)
                        conn.send(data)
except error:
        s.close()
        conn.close()
        s1.close()
        print 'Closing connections. Bye.'






From markus.oberhumer at jk.uni-linz.ac.at  Tue Jan 18 12:50:29 2000
From: markus.oberhumer at jk.uni-linz.ac.at (Markus F.X.J. Oberhumer)
Date: Tue, 18 Jan 2000 18:50:29 +0100
Subject: [ANNOUNCE] PySol 3.20 - a solitaire game collection
Message-ID: <20000118185029.A7620@laetitia.oberhumer.com>

-----BEGIN PGP SIGNED MESSAGE-----


                    PySol - a Solitaire Game Collection
                              Version 3.20

                          http://pysol.tsx.org

           Copyright (C) 1998, 1999, 2000 Markus F.X.J. Oberhumer
                   <markus.oberhumer at jk.uni-linz.ac.at>


What is PySol ?
===============

PySol is an exciting collection of 151 solitaire card games.

Among the supported games are classics like Aces Up, Baker's Game,
Canfield, FreeCell, Forty Thieves, Golf, Klondike, Monte Carlo,
Osmosis, Pyramid, Scorpion, Spider, Yukon, and many more...

PySol may be freely distributed under the terms of the GNU GPL.

PySol aims to be Commercial Quality Freeware.


Why yet another solitaire game ?
================================

Here are some highlights of PySol:

  - freely available
  - distributed under the GNU GPL with full source code
  - currently supports 151 (!) distinct solitaire variants
  - based upon an extensible solitaire engine
  - very nice look and feel including multiple cardsets
    and background table tiles
  - unlimited undo & redo
  - load & save games
  - player statistics and log files
  - hint system
  - demo games
  - support for user written plug-ins - add your own solitaire variants
  - integrated HTML help browser
  - lots of documentation
  - fully portable across Unix/X11, Windows 95/98/NT and MacOS
  - written in 100% pure Python
  - just run it - no need to compile anything


Nice. And what's new ?
=======================

  * Added 8 Hanafuda type games (Oonsoo, Pagoda, MatsuKiri, ...)
    Many thanks to T. Kirk.

  * Added 2 Poker type games (Poker Shuffle, Poker Square)

  * Added 13 other new games (Der Katzenschwanz, Perpetual Motion,
    Die Schlange, Three Shuffles and a Draw, Vegas Klondike, ...)

  * Background music now is supported under Win32 as well.

  * Added 2 cardsets to the PySol-Cardsets collection.


Cool. Where can I get it ?
==========================

Point your browser to http://pysol.tsx.org
The PySol Gallery is awaiting your visit as well.


What do I need to start playing ?
=================================

PySol requires Python 1.5.2 and Tcl/Tk 8.0.5. Both packages are
freely available for Unix, Windows 95/98/NT and Macintosh.

The Windows version ships as a completely self-contained setup file.

BTW, there is no need to compile anything since the whole program is just
a Python script. Just run it, and that's all.


License terms
=============

PySol is Copyright (C) 1998, 1999, 2000 Markus Franz Xaver Johannes Oberhumer

PySol is distributed under the terms of the GNU General Public License (GPL).
See the file COPYING.


Have fun,
Markus

http://pysol.tsx.org



-----BEGIN PGP SIGNATURE-----
Version: 2.6.3ia
Charset: noconv

iQCVAwUBOISZnm10fyLu8beJAQFb9AP8Dy5ejjFXjb6rxVuke4a1xdslmnG62kGx
uckyKJJ1UHPxACj6esVE+GDASz/X5SH8fs0Rg8GkiLIKfE6VBoF8r7mfVJJ2dedm
zoWJxlZDtT3435oCI7UQ4PrzLVszOIYbZ+RXJqX+AUAFoZsbOAvtjKws+HDqzCek
9ojKr6x+FOk=
=ILMM
-----END PGP SIGNATURE-----



From mmiller3 at iupui.edu  Fri Jan 14 09:57:35 2000
From: mmiller3 at iupui.edu (M.Miller)
Date: 14 Jan 2000 09:57:35 -0500
Subject: file width and height
References: <Pine.WNT.4.21.0001131745340.84-100000@wks_3.e-shopwork.com>
Message-ID: <87ya9s1z1c.fsf@lumen.med.iupui.edu>

>>>>> "Sposhua" == Sposhua  <sposhua at my.pc> writes:

    > Is there a module to find gif/jpeg width and height?

Use PIL:

import Image
im = Image.open(filename)
print im.size


From leonar_n at epita.fr  Thu Jan 13 07:33:01 2000
From: leonar_n at epita.fr (Nicolas Léonard)
Date: Thu, 13 Jan 2000 13:33:01 +0100
Subject: os.popen under Win32 Using Apache
Message-ID: <85kgd6$3032$1@news6.isdnet.net>

Hit there


I have a problem when I try to use os.popen() in a cgi with Apache,
under Win32.

It works in command-line mode, but don't work through Apache.

Any idea ?

Thanks.

--
Nicolas L?onard
T?l : (33) 06-61-40-30-31
E-Mail : leonar_n at epita.fr




From warlock at eskimo.com  Fri Jan 21 00:34:03 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Thu, 20 Jan 2000 21:34:03 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net>  <862jqd$l9s@news.or.intel.com>
Message-ID: <slrn88frub.2d0.warlock@raven.myth>

On Tue, 18 Jan 2000 12:50:55 -0800, 
 tye4, in the persona of <tye4 at yahoo.com>,
 brought forth the following words...:

>
>Tom Culliton <culliton at clark.net> wrote in message
>news:YZRg4.1292$Ve7.28796 at iad-read.news.verio.net...
>> In article <860ftk$bas at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>> Have you got extensive experience at designing and implementing
>> computer lanaguages like Guido?  Did you spend years working on a
>> major language design/development project (ABC) including lots of
>> usability studies like Guido?  Have you had the benefit of Tim Peters,
>> a veritable font of language implementation experience and good taste,
>> whispering (or occasionally shouting) in your ear?  I rather doubt it.
>>
>Are you insinuating that this is not a problem since I'm not as experienced
>as Guido? I don't need to design a language to detect flaws in it.
>This problem needs to be fixed. That's because indentation is
>highly dependent on what text editor you use. Some editors convert
>tabs to spaces some don't. Only way around that would be not use
>tabs... which is quite ridiculous (I don't want to type 4 spaces for every
>line.)
>Besides, some editors auto-indent which is a very nice feature.. but one
>that might cause problems with Python.

well, for me, I use vim, and :set tabstop=4 works just jimdandy
(yes, vim also auto indents nicely thank you.)

Frankly, I have grown to prefer the whitespace syntactic style. 

>
>> Part of the reason that Python is so clean and elegant is because
>> Guido, with that background, discarded everything that didn't improve
>> clarity, conprehensibility, and ease of learning, and only kept things
>> that helped.
>
>Python is clean and elegant.. yes but it could me much better.
>Unfortunately, it's hard to change basic syntax of a language.
>
>-tye4
>
>

The source is there, you are free to make any changes you see
fit, and others are free to adapt them if they wish. I'll keep
whitespace myself though thank-you-very-much.

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From a.eyre at optichrome.com  Fri Jan 21 09:08:20 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Fri, 21 Jan 2000 14:08:20 -0000
Subject: Changing the passed arguments value after returning from a module
In-Reply-To: <867ebt$d5i2@tosca.frcl.bull.fr>
Message-ID: <002c01bf6418$f8b99d90$3acbd9c2@peridot.optichrome.com>

> I'm trying to write a C module that changes the values of a passed
> argument, which I am sure is a classic problem. The Python script
> would look as:
> 
>     import mymodule
>     i=0
>     mymodule.test( i )
>     print i

You have to really mess with Python's internals to do that. I'm
assuming you only want to change integers here.

/* N.B.: Not tested */
static PyObject* sct_tt( PyObject* zself, PyObject* args)
{
    PyObject* o;    
    if (!PyArg_ParseTuple(args, "O", o))
         return NULL;
    ((PyIntObject*) o)->ob_ival = 101;
    Py_INCREF(Py_None);
    return Py_None;
}

Be *very* careful if you intend to do this though:

1. You are dealing with undocumented intefaces, and cannot rely on
   this being the same in later releases.

2. Python caches several small integers (not sure how many), so this
   code would actually change all integer variables with the value 0
   to the value 101.

>>> import mymodule
>>> i = 0
>>> j = 0
>>> mymodule.test(i)
>>> print i
101
>>> print j
101

Also... the reason why your version does not work is that you are
changing the tuple used when the function is called. This goes out
of scope as soon as the function exits. It would work if you did:

>>> import mymodule
>>> i = (0,)
>>> apply(mymodule.test, i)
>>> print i
101

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com  Tue Jan  4 12:40:53 2000
From: coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com (Stephen Coursen)
Date: 4 Jan 2000 17:40:53 GMT
Subject: CPAN for Python?
References: <38720F70.A5DC6F33@ttm.com> <84t52c$1gg$1@nntp9.atl.mindspring.net> <C28D1A44F2165E48.B5C4E34C4CD957A5.44F995461BA2F6ED@lp.airnews.net>
Message-ID: <slrn874buo.2do.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>

On 4 Jan 2000 11:30:50 -0600, Tres Seaver <tseaver at aftershock.neosoft.com>
wrote:
>In article <84t52c$1gg$1 at nntp9.atl.mindspring.net>,
>Aahz Maruch <aahz at netcom.com> wrote:
>>In article <38720F70.A5DC6F33 at ttm.com>,
>>Scott Anderson  <sanderson at ttm.com> wrote:
>>>
>>>Is there an equivalent of Perl's CPAN for Python?
>>
>>Well, there's Parnassus (http://www.vex.net/parnassus/), but there's not
>>yet an overarching authority.
>
>www.python.org blesses it, which is probably about as close as we'll see.
>I may misrecall (blessedly little Perl experience) but doesn't CPAN allow
>for browsing from the interpreter prompt, and downloading?  That would be
>a cool extension of a wonderful resource.
>
>>OTOH, Python's "built-in" libraries are
>>more extensive than Perl's; see the Library Reference for details.
>
>parnassus.py,-anyone?'ly
>
That would be a useful module to have, but for it to work, there would have
to be a standard installation mechanism.  Perl uses:
     perl Makefile.pl
     make
     make install
     
If python had a standard way of installing modules, writing a parnassus.py
script would be easy.  Unfortunately, not every module installs the same way
(although there seems to be only a few different methodologies).

Steve

>Tres.
>-- 
>---------------------------------------------------------------
>Tres Seaver           tseaver at palladion.com       713-523-6582
>Palladion Software    http://www.palladion.com


From colin_kirkhamNOcoSPAM at inmarsat.org.invalid  Tue Jan 18 10:54:00 2000
From: colin_kirkhamNOcoSPAM at inmarsat.org.invalid (cthulhu)
Date: Tue, 18 Jan 2000 07:54:00 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <17599f0b.ab824846@usw-ex0105-038.remarq.com>

Just my ha'penny worth, but I don't mind the indentation at all, apart
from the fact that I use Remarq.com to read the newsgroups at work, and
that removes all indentation from code anyway.

This means that the logic of any nested indentation is hidden from me
in any posted code.

Now if only deja was faster...




* 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 bovinesoft at my-deja.com  Sun Jan 30 23:42:25 2000
From: bovinesoft at my-deja.com (bovinesoft at my-deja.com)
Date: Mon, 31 Jan 2000 04:42:25 GMT
Subject: Can someone fix my code?
Message-ID: <8733rg$ral$1@nnrp1.deja.com>

I am a complete Python newbie and am trying to learn it as my first real
programming language (I have dabbled in C, Java, and Assembly, but never
really used them).  I am trying to write a program to get some practice,
but am getting errors when I try to run it.  If someone could take a
look at my code and tell me what is wrong with it, I would be greatly
appreciative.

import sys  # to use the exit() function

def ssi_or_exit():
   sorx = raw_input("Would you like to generate HTML <g> or exit <x>? ")
   if sorx == "g":  # if they want to generate a web page
      ssi()
   elif sorx == "x":  # if they want to exit
      sys.exit("Thank you for using Bovine No Server SSI!")
   else:
      ssi_or_exit() # try it again

def ssi():
   body_path = raw_input("Please type the path of the body section: ")
   body = open(body_path, "w")  # open the text of the web page body
   temp_path = raw_input("Please type the path of the HTML template: ")
   temp = open(temp_path, "r")  # open the HTML template
   temp2 = temp.readlines()
   where_temp = raw_input("Put the template before <b> or after <a> the
body? ")
   if where_temp == "b":  #around here is where my trouble starts
      for line in temp2:
         body.write(line)
      for line in body:
         body.write(line)
   else:
      for line in body:
         body.write(line)
      for line in temp2:
         body.write(line)
   body.close()
   temp.close()
   ssi_or_exit()

ssi_or_exit()


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


From flounder_pounder at telebot.com  Sat Jan  1 22:20:07 2000
From: flounder_pounder at telebot.com (flounder)
Date: Sun, 02 Jan 2000 03:20:07 +0000
Subject: Need help with canvas widget in Tkinter please
Message-ID: <386EC3E6.FAC5BC0@telebot.com>

I cannot seem to figure out how to add images to a canvas with python/Tkinter

I can do it with Tcl/Tk and Perl/Tk but not with python i have tried this but

it did not work at all. Here is the code i used

-----------CODE--------------

#!/usr/bin/python

from Tkinter import *

mw = Tk()
mw.title("Map Editor")
photo = PhotoImage(file="tile0.gif")

display = Canvas(mw, width=640, height=480).pack()

item = display.create_image(0, 0, anchor=NW, image=photo)

image = display.addImage
mainloop()

--------------END OF CODE---------------

when i run this it gives me this error

-----------ERROR MESSAGE--------------

Traceback (innermost last):
  File ".//test.py", line 11, in ?
    item = display.create_image(0, 0, anchor=NW, image=photo)
AttributeError: 'None' object has no attribute 'create_image'

-----------END OF ERROR---------------

can anyone show me how this should be done I have looked at the

documentation and can't figure it out

Thanks in advance

--
Flounder





From mgushee at havenrock.com  Sun Jan  9 10:01:36 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 09 Jan 2000 10:01:36 -0500
Subject: Question about dictionaries (Question about is ?)
References: <3.0.5.32.20000105215635.0091d420@mail.dicksonstreet.com> <3.0.5.32.20000108113031.007bccc0@mail.dicksonstreet.com>
Message-ID: <m3iu13uw4v.fsf@havenrock.com>

Felix Thibault <felixt at dicksonstreet.com> writes:

> >> or something, or is this just a Stupid Programming
> >> Trick?
> >
> >Hah! Your next challenge is to find a use for this interesting
> >technique. If you can do that, it won't be just a Stupid Programming
> >Trick ;-)
> 
> Well the use I have is building a nested dictionary out of 
> some HTML...near the beginning of the program I make 
> a dictionary with the tags in the document for keys, so 
> it's easy to make everything downhill from that point to the
> same 20 or so slices of the document instead of pointing
> to hundreds ... Which is simpler for me to think about, but
> I'm getting the idea that the computer doesn't care.

Well, it's at least a fascinating experiment. And without knowing in
more detail what you're trying to accomplish, it's hard to say for
sure, but my impression is that perhaps you're being a little too
clever.  If you can keep straight which of these indirect object
references is referring to what (as apparently you can) you're smarter
than I am. But when you start trying to process real HTML documents
with all their idiosyncracies, you might have an enormous mess on your
hands.

In summary: if you're just playing around to see how things work,
don't let me or anybody else stop you. But if you are trying to
produce some useful code in a reasonable amount of time, you might
start by looking at how other people have dealt with HTML. If you look 
around The Vaults of Parnassus (http://www.vex.net/parnassus/), I'm
sure you'll find something to do with HTML. The XML-SIG might also
give you some ideas.

Best of luck!

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From backov at nospam.csolve.net  Thu Jan  6 18:21:44 2000
From: backov at nospam.csolve.net (Jason Maskell)
Date: Thu, 06 Jan 2000 23:21:44 GMT
Subject: newbie embedding questions
References: <kX6d4.41059$Ke.26557@news21.bellglobal.com> <Fnxpnq.JrD@world.std.com>
Message-ID: <cs9d4.42068$Ke.26436@news21.bellglobal.com>

Will Ware <wware at world.std.com> wrote in message
news:Fnxpnq.JrD at world.std.com...
> Jason Maskell (backov at nospam.csolve.net) wrote:
> > Well, I've finally gotten around to embedding Python in an app...
> > The main app is essentially a skeleton, that will be driven by Python
> > scripts.. It does however hold a bunch of C++ objects that it owns and
> > instantiates and that the Python scripts need access to. So basically I
need
> > "glue" code to make sure the Python stuff has access to those instances'
> > methods, and to set callbacks in the python code for generated events.
>
> You might want to look at Swig (www.swig.org) which makes it really
> painless to extend Python with C++ objects. I think the preferred
[snip]

Actually, I was looking at that this afternoon, and right now.
Unfortunately, there don't seem to be any binaries available, and compiling
it in BCB is looking to be a huge chore. I suppose I could install VC again,
but that seems like a big hassle just to compile this tool and uninstall it.

Cheers,
Jason





From apederse at siving.hia.no  Fri Jan 21 08:02:13 2000
From: apederse at siving.hia.no (Asle Pedersen)
Date: Fri, 21 Jan 2000 14:02:13 +0100
Subject: HTMLparser
Message-ID: <869lbo$r3$1@snipp.uninett.no>

I'm a beginner Python user and experimenting with the HTMLparser. I want to
convert all relative urls to absolute urls without thouching the rest of the
file containment.

this is what I have so far. but somehow it is throwing away all tags but the
anchor tags. (which is not what I want)

class minparser(htmllib.HTMLParser):

    def __init__(self, formatter, verbose=0):
 htmllib.HTMLParser.__init__(self, formatter, verbose)

    def anchor_bgn(self, href, name, type):
        self.anchor = urlparse.urljoin("baseurl",href)
        if self.anchor:
     self.save_bgn()

    def anchor_end(self):
        if self.anchor:
     text = self.save_end()
    #need to do something here
     self.handle_data("%s <%s>"%(text,self.anchor))
     self.anchor = None


-Asle




From hei at adtranzsig.de  Wed Jan 26 07:43:34 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Wed, 26 Jan 2000 13:43:34 +0100
Subject: Wanted: A python icon for X11
References: <pcosnzl4bpf.fsf@math.ntnu.no> <00012610342500.00739@fishology>
Message-ID: <948894359.212742@news.adtranzsig.de>

http://www.cs.ubc.ca/spider/mjmccut/python.gif

A Python and a can of spam, impersonated by actor
Nastassja Kinski. Does she count as an icon?

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


Robert Longstaff schrieb in Nachricht <00012610342500.00739 at fishology>...
> Would anybody perhaps have some pixmap lying about that might be (or
> become) a suitable icon for IDLE windows under X11?  A picture of a
> python (the snake) perhaps, or of a can of spam, or something like
> that?  Right now, I only have the generic X icon, and that is of
> course intolerable.





From mal at lemburg.com  Tue Jan 18 13:01:08 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 18 Jan 2000 19:01:08 +0100
Subject: DBI spec and parameter passing
References: <3883A0CA.AB0D4854@exceptionalminds.com>
		<14467.42391.905771.673058@beluga.mojam.com>
		<1263967432-11066499@hypernet.com>
		<20000118015809.F21652@xs4all.nl>
		<388455BB.154F36EC@lemburg.com> <14468.38220.788564.314607@beluga.mojam.com>
Message-ID: <3884AA64.AF62FB3C@lemburg.com>

Skip Montanaro wrote:
> 
>     Marc> Note that one should always use binding parameters if possible:
>     Marc> the driver or interface usually knows better how to encode the
>     Marc> data according to the DBs specs and it's also much easier to write
>     Marc> and more portable.
> 
> In my local case (MySQLdb), I use "%s" and your indication is that ODBC uses
> "?", which is fine.  Other database modules apparently use named
> interpolation (e.g., "%(city)s").  Is there a database-independent way of
> building the basic strings other than switching on the value of
> db.paramstyle:
> 
>     db = somethingDBIcompliant(...)
>     c = db.cursor()
>     if db.paramstyle == "qmark":
>         c.execute("sql statement with ? embedded", params)
>     elif db.paramstyle == "numeric":
>         c.execute("sql statement with :1 embedded", params)
>     elif db.paramstyle == "format":
>         c.execute("sql statement with %s embedded", params)
>     ...
> 
> or is that much portability worth the effort?  My code stops before this
> point in its attempts to be independent of a particular database module.

I would suggest to add a DB abstraction layer on top of the
DB module interface -- even with ODBC which is supposed to
provide much of this abstraction you often run into situations
where DBs use different SQL syntax, different column type
names etc.

If you really want your code to DB independent,
you should consider coding your particular needs into a
Python abstration layer and then write backends for each DB
you target. 

Not sure whether it's worth it though... some
applications do just fine when coded for a single DB target,
e.g. large accounting systems can well assume to have an
Oracle or IBM DB available, small apps can do well with MySQL
(which is not free in this case, BTW).

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





From neelk at brick.cswv.com  Tue Jan  4 19:56:11 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 5 Jan 2000 00:56:11 GMT
Subject: Simple cartesian product
References: <iacpuvhqwdp.fsf@vier.idi.ntnu.no> <iacn1qlqvw7.fsf@vier.idi.ntnu.no> <iack8lpqvoi.fsf@vier.idi.ntnu.no>  <iachfgtqvj9.fsf@vier.idi.ntnu.no>
Message-ID: <slrn8755l1.gvl.neelk@brick.cswv.com>

Magnus L. Hetland <mlh at vier.idi.ntnu.no> wrote:
> 
> [Cartesian product using list operators]
>  
> >>> def product(a,b):
> ...     a2 = reduce(lambda l, x, n=len(b): l + [x]*n,
> ...                [[]] + a)
> ...     b = b * len(a)
> ...     return map(None, a2, b)                      

Here's my entry:

def product(a, b):
    u = map(lambda i: a[i/len(a)], range(len(a)*len(b)))
    v = map(lambda j: b[j%len(b)], range(len(a)*len(b))) # Same as b*len(a)
    return map(None, u, v)

>>> product([1,2,3], ['a','b'])
[(1, 'x'), (1, 'y'), (1, 'x'), (2, 'y'), (2, 'x'), (2, 'y')]

Obviously, if you want to save your sanity the nested for-loops are
the way to go. :)


Neel


From cfelling at iae.nl  Sat Jan 29 07:34:13 2000
From: cfelling at iae.nl (Carel Fellinger)
Date: 29 Jan 2000 13:34:13 +0100
Subject: Cutting off trailing zeroes?
References: <3892ad81$1_2@goliath.newsfeeds.com>
Message-ID: <86umo5$184$1@vvs.superst.iae.nl>

Janos Blazi <jblazi at netsurf.de> wrote:
> Id like to format a string like this:
>   s = "(%f)" % x

> Now if x=1.5 then s="1.50000" or something like this. Is there a tirivial
> way of cutting of the zeroes, i.e. getting s="1.5" instead of s="1.50000"?

try the g and G specifiers like:

>>> "%g" % 1.5000
1.5
>>> "%g" % 1.50002
1.50002
>>> "%1.3g" % 1.50002
1.5

the default precision for g is, I think, .8. But it doesn't quit work like
advertised (on my system at least), somehow it influences the field width
as wel:( So you get 8 chars atmost, including the point and the numbers
before the point. So be carefull to avoid the following:

>>> "%1.3g" % 111.50002
112
-- 
groetjes, carel


From aahz at netcom.com  Mon Jan 24 19:06:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 25 Jan 2000 00:06:31 GMT
Subject: timeout for urlopen
References: <8561e2$kah$1@nnrp1.deja.com>
Message-ID: <86ipe7$l04$1@nntp2.atl.mindspring.net>

In article <8561e2$kah$1 at nnrp1.deja.com>,  <ryuch at my-deja.com> wrote:
>
>How can I set the timeout for urlopen?

You can't.  You have to use httplib directly.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From tim_one at email.msn.com  Wed Jan  5 20:30:23 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 5 Jan 2000 20:30:23 -0500
Subject: lazy (?) evaluation, request for advice
In-Reply-To: <014701bf57d2$f9f69a20$c02b2bc1@martelli>
Message-ID: <000701bf57e5$9aa20560$b1a0143f@tim>

[Alex Martelli]
> I'm planning to prototype in Python a bunch of bridge-
> related functionality, including selective dealing.  In
> selective dealing, the program must repeatedly generate
> random deals, and apply on each generated deal a user-
> supplied function, which can accept or reject the deal ...
> ... I'd much rather let the user write the deal-selection
> function in Python, than in either Tcl, or any custom,
> ad-hoc "little language".

Against that, you invented ad-hoc semantics as you went along; the only
connection to Python appeared to be syntactic.  Let the user write real
Python!  Then as the program gets fancier they'll have the full language to
exploit; with "spades" etc as magical global names, what chance is there to
extend the scheme to a user-defined function that e.g. wants to examine
*two* hands at a time?  (I know! "spades1" and "spades2" <wink>)

> ...
> Further, since the typical user will not be a programmer,
> I would like to make the user's life as easy as possible.
> I.e., despite the important mantra about explicit being
> better than implicit, I think the user would much rather
> be able to write, say:
>     (spades>=6 or hearts>=6) and (3.0<losers<4.0)
> rather than have to mention "current_hand.spades" or
> other such explicit notation.

Here you fall head first into the "ad hoc little language" trap, as if to
say "my users are so stupid there's no chance of getting them to learn
*anything* -- so I'll give them something so flat there's no chance either
it can be extended cleanly later".

def userfunc(x):  # they get to make up the argument name
    ... (x.spades >= 6 or x.hearts >= 6) ... etc

Provided your users can pass a mirror test (i.e., you hold a mirror under
their nose, and they pass if it fogs up <wink>), they can learn this.  After
which, playing __getattr__ tricks with the x object gives you the "lazy"
behavior you want.

> ...
> I've thought about doing the eval repeatedly, within
> a try/except block to catch the NameError's, inserting
> the missing names in the dictionary one by one, but
> I'm suspicious about the performance characteristics
> of such an approach.

You should be more suspicious of the sheer convolution <0.5 wink>.

Anyway, what's expensive there isn't so much the try/except, but the
repeated eval *if* the user's expression is carried around as a string (eval
will recompile it to byte code each time).  If you do go this route, see the
builtin "compile" function for a way to do the compilation to byte code just
once (eval accepts compiled code objects too).  Note also that if your users
truly are morons, there's nothing to stop *you* from marching over their
string and replacing e.g. "spades" with "current_hand.spades" before
compilation.

treat-users-like-grownups-and-they-won't-demand-you-give-
    them-an-allowance<wink>-ly y'rs  - tim





From pett at address.is.in.sig  Mon Jan 31 22:01:34 2000
From: pett at address.is.in.sig (Eric Pettersen)
Date: 1 Feb 2000 03:01:34 GMT
Subject: Tkinter and clone widgets
Message-ID: <875iae$4ij9$1@cgl.ucsf.edu>

I came up with a mod to Tkinter.py to handle Tk clone widgets that I
thought I'd run by the wizened sages of the group to see if it seemed
copesetic.

As a quick review, when you use the "menu=" option of a Toplevel to set
the window menubar, Tk clones your menu hierarchy and uses these clones
to form the menubar, rather than using your original menu hierarchy.
Changes to the original hierarchy are reflected in the clone hierarchy,
and choosing menu items in the clone hierarchy cause the appropriate
original callbacks to get fired.  The problem is if you are trying to use
Tkinter to bind other event handlers to these widgets (e.g. balloon help),
since the clone widget name that Tkinter receives from Tk isn't properly
resolved to the Menu instance.

However, the clone menu name can be fairly easily parsed to extract the
name of the corresponding original menu.  A clone widget name looks like
this:

.#5372422832#5374429664.#5372422832#5374429664#5374866736.#5372422832#5374429
664#5374866736#5373451424.#5372422832#5374429664#5374866736#5373451424#537487
1536

Yes, it is indeed butt-ugly.

The rules used to construct clone widget names seem to be undocumented
(please correct me if I'm mistaken), but seem to be:

1) the name is dot-separated like a normal widget name (and starts with
a dot)

2) each name component corresponds to a clone widget, with the first being
the menubar and successive components corresponding to increasingly deeper
nested child clone widgets

3) The name components are each like a normal full widget names with hashes
('#') instead of dots ('.'), and the hash-separated components correspond
to their "real" (non-cloned) widgets.

So, it's possible to find the Tkinter Menu instance from the clone name
by looking at the last dot-separated name component, change the hashes to
dots, and then look that up as if it were a normal widget name.  The change
to Tkinter to do that is appended below.  Essentially, it changes the "w
= w.children[name]" statement of the nametowidget function into a try/except
clause.

*** Tkinter.py  Fri Jan 21 16:03:47 2000
--- Tkinter.py.orig     Fri Jan 21 15:45:02 2000
***************
*** 532,550 ****
                                name, tail = name[:i], name[i+1:]
                        else:
                                tail = ''
!                       try:
!                               w = w.children[name]
!                       except KeyError:
!                               if name[0] == '#':
!                                       # looks like clone widget...
!                                       if not tail:
!                                               tail = name
!                                       lastDot = _string.rfind(tail, '.')
!                                       if lastDot >= 0:
!                                               tail = tail[lastDot+1:]
!                                       clone = _string.replace(tail, '#', 
'.')
!                                       return self.nametowidget(clone)
!                               raise
                        name = tail
                return w
        _nametowidget = nametowidget
--- 532,538 ----
                                name, tail = name[:i], name[i+1:]
                        else:
                                tail = ''
!                       w = w.children[name]
                        name = tail
                return w
        _nametowidget = nametowidget

Opinions?
---
		Eric Pettersen
		pett "at" cgl "dot" ucsf "dot" edu (NeXTmail capable)


From a at b.c  Wed Jan 12 07:51:20 2000
From: a at b.c (DORAEMON.HK)
Date: Wed, 12 Jan 2000 20:51:20 +0800
Subject: Compile snmpy problem
Message-ID: <85ht8l$o4j1@imsp212.netvigator.com>

Dear All,

I try to compile SNMPy in a RedHat 6.1 server.  But I got,

checking whether byte ordering is bigendian... configure: error: can not run
tes
t program while cross compiling


Any help?  :-(


Regards,
DORAEMON.HK




From Robin.Friedrich at pdq.net  Sat Jan 29 16:32:28 2000
From: Robin.Friedrich at pdq.net (Robin Friedrich)
Date: Sat, 29 Jan 2000 21:32:28 GMT
Subject: Pics from IPC8 (SNOWSPAM2000)
Message-ID: <38935C6B.45E4EB4A@pdq.net>

http://freeweb.pdq.net/robinf1/spam8pix/
(FTP to the starship isn't working so I put them up here.)
These are some of the photos taken this week at SNOWSPAM2000
-- 
Robin K. Friedrich     Houston, Texas


From aahz at netcom.com  Mon Jan 31 15:59:47 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Jan 2000 20:59:47 GMT
Subject: nesting class definitions?
References: <20000127155551.A2069@stopcontact.palga.uucp> <20000128182031.C16093@xs4all.nl> <871vqt$a8d$1@nntp8.atl.mindspring.net> <20000131144411.A19725@xs4all.nl>
Message-ID: <874t43$3jq$1@nntp9.atl.mindspring.net>

In article <20000131144411.A19725 at xs4all.nl>,
Thomas Wouters  <thomas at xs4all.net> wrote:
>On Sun, Jan 30, 2000 at 06:27:41PM +0000, Aahz Maruch wrote:
>> Yup.  But try this one on for size:
>>
>> def SpamGen ( foo ):
>> 	class Spam:
>> 		bar = foo
>> class Viking:
>> 	def __init__(self, foo):
>> 		self.spamClass = SpamGen(foo)
>
>Er, correct me if i'm wrong, but shouldn't this generate a NameError (asside
>from SpamGen returning None, of course) ? Or did you intend to show that
>this is very clear and very useful code, but impossible at the moment ?

Oops.  That's what I get for assuming that something will work without
testing.  I've figured out how to make it work using "global" but that's
really ugly.  For the curious:

def SpamGen ( foo ):
	global plugh
	plugh = foo
	class Spam:
		bar = plugh
	return Spam
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From shang.remove_edu at st.jyu.fi.edu  Thu Jan 20 14:23:40 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Thu, 20 Jan 2000 21:23:40 +0200
Subject: How to store a function pointer in class?
References: <867j5p$qvq$1@mordred.cc.jyu.fi> <vndaem01s1r.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <867nbb$1ho$1@mordred.cc.jyu.fi>

Justin Sheehy <dworkin at ccs.neu.edu> wrote in message
news:vndaem01s1r.fsf at betelgeuse.ccs.neu.edu...

> A class or instance variable may be a reference to a function
object, though.

If wish they could! The problem is that a class variable can NOT refer
a function object, only an unbound method object. When a reference to
a function object is stored in a class variable, it is instatly
transformed into an unbound method (or so it seems).

> Since I'm not sure what you want to _do_ with your function
pointers,
> I'm not sure if I am answering your real question.  This may be
helpful:

Sorry, I wasn't being clear enough. This is what I want to do:

>>> Test.class_attribute = some_func
>>> another_func = Test.class_attribute
>>> another_func == some_func  # here's the problem :(
0

So I wan't to store function references into class variables and
retrieve them unchanged.

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi




From mfletch at tpresence.com  Tue Jan 25 05:10:42 2000
From: mfletch at tpresence.com (Mike Fletcher)
Date: Tue, 25 Jan 2000 05:10:42 -0500
Subject: Integer literal overflow error with -2147483648 but not with 0x80
	000000 ?
Message-ID: <CC59868A0A76D311B1F50090277B321C1F30A8@VRTMAIL>

Just wondering if there's a reason for this (doesn't cause a problem for me,
just seems strange)...

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> 0x80000000
-2147483648
>>> -2147483648
Traceback (innermost last):
OverflowError: integer literal too large
>>> 

__________________________________
 Mike C. Fletcher
 Designer, VR Plumber
 http://members.home.com/mcfletch



From tiddlerdeja at my-deja.com  Sun Jan 30 12:36:31 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Sun, 30 Jan 2000 17:36:31 GMT
Subject: GetObject instead of CreateObject
Message-ID: <871sqv$kp$1@nnrp1.deja.com>

Can someone tell me what function I should call in Python to perform a
similar task to VB's GetObject?

I realise that this is probably a lame question. Could someone even tell
me where I might find this information in my Python distribution so that
I dont have to ask it here?

Any help appreciated.

Cheers.


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


From haje01 at netsgo.com  Mon Jan 24 12:21:41 2000
From: haje01 at netsgo.com (Kim Jeong Ju)
Date: Tue, 25 Jan 2000 02:21:41 +0900
Subject: Re-assigning 'stdin' Problem...
Message-ID: <cW%i4.3$R04.297@news.hananet.net>

Hi!
I am a newbie to Python and having some trouble.

What I want to do is re-assigning standard input
to my own version of console system. ( it uses c string
buffer inside. )

I made a console class in order to link Python's stdin
to my c string buffer..

class MyConsole:
    def readline( self ):
        if ( self.BufferHasData )
            return self.GetBufferData()
        else
->        return ""

    def InBufferHasData( self )
            .
            .
    def GetBufferData( self )
            .
            .

sys.stdin = MyConsole()

When stdin is in need, like running 'pdb' procedure,
system hangs out.

I guess that marked line in readline function is problem.
if i replace it with 'return None ', interpreter complains
that it is not string object...
I guess I must stop the Python interpreter from requesting
more readline.

What can I do to fix this problem? You could just drop me a
link to a reference.

Any suggestions, instructions and helps appreciated ahead..

Bye.

from Kim Jeong Ju.





From tim_one at email.msn.com  Wed Jan  5 20:33:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 5 Jan 2000 20:33:15 -0500
Subject: Simple cartesian product
In-Reply-To: <iacembwr7qg.fsf@vier.idi.ntnu.no>
Message-ID: <000a01bf57e6$00b4b280$b1a0143f@tim>

>> c'mon-folks-new-millennium-new-topics<wink>-ly y'rs  - tim

[Magnus L. Hetland]
> Now - you're not going to get that "new millennium"-discussion
> started, are you?

Of course not!  Ivan settled that one last millennium <wink>:

    http://www.douglasadams.com/dna/pedants.html

> We are still in the same millennium, century, and decade as
> last year. The third millennium starts with 2001.01.01.

all-words-mean-what-guido-says-they-mean-ly y'rs  - tim





From zeitlin at seth.lpthe.jussieu.fr  Thu Jan 20 18:10:35 2000
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 20 Jan 2000 23:10:35 GMT
Subject: HTML Widget
References: <38876AB2.58FF3EA5@hks.com>
Message-ID: <slrn88f5n6.q2r.zeitlin@seth.lpthe.jussieu.fr>

On Thu, 20 Jan 2000 15:06:11 -0500, Chris Swepson <swepson at hks.com> wrote:
>    I've been searching far and wide for an HTML widget which can be
>used in Python/Motif.

 Hello,

 I don't know if it works under Motif (it should, but I didn't test it), but
wxPython has interface to wxHTML library which you could check out.

 Regards,
VZ


From AKiss at GEOMETRIA.hu  Tue Jan 11 10:57:00 2000
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Tue, 11 Jan 2000 16:57:00 +0100
Subject: Newbie simple question - can't find answer
Message-ID: <D148AB3A9D8FD311A99A008048C79D11038702@MAIL>


<eleisonme at juno.com> wrote in message news:<85fhha$6js$1 at nnrp1.deja.com>...
> Hello all,
> I am using Python 1.5.2-7 on intel Red Hat Linux 6.1.
> I do not seem to be able to use any of the math or cmath functions
> either in a script or on the python command line.
> 
> For example if I type:
> >>>import math
> >>>exp (4)
> 
> OR
> 
> >>>exp(4)
> 
> I get a traceback error:
> NameError: exp (or any other math function that I try to use).
> 

Try: math.exp(4) 
If you want use exp(4) instead of math.exp(4) then use this format of
import:
from math import *
The documentation describes how you can import and what are the differences
between them.

Arpad



From loewis at informatik.hu-berlin.de  Thu Jan 20 08:26:59 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Jan 2000 14:26:59 +0100
Subject: Suggest: .get() method on lists.
References: <20000118212229.A542@stopcontact.palga.uucp>
Message-ID: <p6qwvp4c1r0.fsf@informatik.hu-berlin.de>

Gerrit Holl <gerrit.holl at pobox.com> writes:

> I thought it would be useful, for example, for sys.argv. At the moment,
> I have this code:
> 
> file = sys.stdin
> if len(sys.argv) > 1:
>     file = sys.argv[1]
> 
> But this, would be much easier to read:
> file = sys.argv.get(1, sys.stdin)

I guess it wouldn't be that useful in this specific case, as
sys.argv[1] if present is a string, so you (probably) write

file = sys.stdin
if len(sys.argv) > 1:
    file = open(sys.argv[1])

right now, to open the file. I can't see how you could get the
equivalent of that code 

Regards,
Martin




From thierry at acm.org  Wed Jan 26 19:14:59 2000
From: thierry at acm.org (Thierry Thelliez)
Date: Wed, 26 Jan 2000 17:14:59 -0700
Subject: Smalltalk blocks in Python
Message-ID: <388F8E03.389BC83A@acm.org>

I come from the Smalltalk world and I am trying to understand Python.

Is there a Smalltalk blocks equivalent in Python ?
Java has a pale imitation with Inner classes.


Thanks
Thierry



From aahz at netcom.com  Thu Jan 20 12:51:14 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Jan 2000 17:51:14 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <slrn886oku.2ac.grant@grante.comtrol.com> <38871EA1.8AF33147@sage.att.com> <slrn88ebbe.67b.grant@grante.comtrol.com>
Message-ID: <867hui$6to$1@nntp3.atl.mindspring.net>

In article <slrn88ebbe.67b.grant at grante.comtrol.com>,
Grant Edwards <grant at nowhere.> wrote:
>
>Then he complained that it was too much work supporting POP/IMAP, and
>he wanted everybody to use Outlook.  People resisted (particularly the
>Unix users, most of whom left shortly thereafter).  The fight went all
>the way to the president of the company, and the decision was made that
>everybody use Outlook.  (Why is management always so big on monotony
>and so down on individuality?)

Everything you've said has made sense except for one thing: why are you
still working there?  I know jobs are relatively easy to find in
Minneapolis, particularly for techies.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From wtanksle at hawking.armored.net  Thu Jan 20 23:10:43 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 21 Jan 2000 04:10:43 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com>
Message-ID: <slrn88fnn3.5fj.wtanksle@hawking.armored.net>

On Thu, 20 Jan 2000 17:04:34 -0800, tye4 wrote:

Hmm...  Interesting subject change.  It's honest -- thanks.  I think ;-).

>Tim Peters <tim_one at email.msn.com> wrote in message

>> I haven't used this script myself (I prefer indentation, for reasons
>> covered earlier), but there's no good reason it *can't* be used for
>> its stated purpose <wink>.

>I don't oppose indentation either. It's good for program readability, and
>readability only. It shouldn't be part of the syntax.

But you forgot something: you forgot to justify your conclusion.  All
you've done is assert it over and over.

>Whitespace end delimiters are invisible and hence confusing and prone to
>error.

Whoops, here's an argument.  Looks like I was wrong about you forgetting
to justify your conclusion.  Unfortunately, it's both invalid and false.

It's invalid because it's arguing beside the point -- Python doesn't use
whiespace as a delimiter.  Python uses indentation to _remove_ delimiters.
There are no more delimiters, invisible or otherwise.

It's false because indentation is NOT invisible -- on the contrary, it's
almost impossible to hide.  All of the other block delimiting schemes have
suffered from illegibility -- Oberon even whent so far as to require its
block delimitation in UPPERCASE to make it more visible.  Unfortunately,
programmers don't want to see block delimiters; they want to see BLOCKS.

>I've used some stupid Unix shell languages which fail if I add whitespace:

There you go.  Fortunately, Python isn't one of them.

>I thought they gotten rid of all such languages until I came across Python.

Have you seen TCL yet?  Shudder.

>Good programming languages should be whitespace tolerant, and most of them
>are.
>(C++, Java, Ada, BASIC, HTML, Eiffel, Perl, Pascal, ML etc... really long
>list.)

>Why is my question about whitespace raise so many eyebrows? This feature is
>commonplace.

tye4, you've just answered your own question.  If you think about it
really hard -- and try to put your prejudice aside -- you might realize
the truth of the matter.

Or I might tell you.

Python uses indentation for the same reason ML uses type inference, Perl
uses multiple namespaces, and Eiffel uses static typing.  Because it's one
the the features of the language.  That's IT!  It's so simple.

If I didn't like indentation, I would walk away and use Ruby or Rebol.

(Psst -- guys?  Shall I tell him?)

Okay, here's the REAL truth.  You said that we were resisting your ideas
because we were stubborn idiots.  Actually, we're almost fiendishly
clever. We use Python this way strictly because it pisses you off so much.
Well, you and Michael (just a guy here at work).

Well, now you know.  And feel free to talk to the press about it.

The van's on its way.  No, not the press van -- the black van.

>-tye4

-- 
-William "Billy" Tanksley, in hoc signo hack


From tothfile at freemail.hu  Fri Jan 28 05:09:13 2000
From: tothfile at freemail.hu (Attila =?iso-8859-1?Q?Filet=F3th?=)
Date: Fri, 28 Jan 2000 11:09:13 +0100
Subject: What is winfo_id() under Win32
References: <3890439A.1C1BA6D2@freemail.hu> <86pt4k$f6h$1@nnrp1.deja.com>
Message-ID: <38916AC9.AD1008A2@freemail.hu>

Hi!

> >         hwnd = self.winfo_id()
> > #       hwnd = win32ui.GetParent(hwnd)  # Attribute error: GetParent
> >         win32gui.ShowWindow(hwnd, SW_SHOWMAXIMIZED)  # It would only
> > work after the GetParent() call
> 
> win32gui _should_ include these functions.  If I remember, I will add
> them.

Only 'GetParent' is the one that cannot be imported in the example.
'ShowWindow' can be called, just doesn't take the expected effect
(because of the wrong hwnd).

> In the meantime, the following _should_ work:
> 
> wnd = win32ui.CreateWindowFromHande()
> parent = wnd.GetParent()

Unfortunately I also get an Attribute error for 'CreateWindowFromHande'.

Actually when I search through win32* files for the string 'GetParent',
it matches in the file Pythonwin\win32ui.pyd (and win32com\win32com.hlp)
while 'CreateWindowFromHande' cannot be found at all and neither of them
can be imported.

(I've fetched the latest win32all.exe but it was the same I've already
installed)

BR,
    Attila


From boud at rempt.xs4all.nl  Mon Jan  3 13:47:32 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 3 Jan 2000 18:47:32 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com> <84olef$5j1$1@news1.xs4all.nl> <84omaa$qln$1@nntp9.atl.mindspring.net> <84pg9v$rda$1@news1.xs4all.nl> <84qg3m$u0v$1@nntp1.atl.mindspring.net>
Message-ID: <84qqs4$dn7$1@news1.xs4all.nl>

Aahz Maruch <aahz at netcom.com> wrote:

> Oh, okay.  Let's suppose, for example, that you have a customer
> database, and the client wants to view the entire order history for a
> customer.  Would you say that it's inappropriate for the database to
> return thousands of rows for an active customer?

Well, it is if the customer is one of many who expects lightning
performance. If the customer wants a lot of data he will have to wait,
and he will simultaneously tie up the server for other customers. If
he's seriously going to look at thousands of records, well, so be it
then. But it remains a fact that what a database is best at is selectively
presenting a bit of data out of a large mass. If your relational database
is mostly employed serving up most of its contents every query, something
is wrong, and you will get the blame for its lacklustre performance.

Even so, in most cases I've seen, when thousands of rows are retrieved
there's designers/developers laziness or ignorance behind it - and in
most of the other cases the customer can be 'educated' in wanting a more
selective result. Nobody is going to eyeball more than a few hundred
records, and if they are going to post-process the result it is often
better to do the post-processing on the server and send out the result.

> Similarly, suppose you have millions of BLObs that change regularly.
> How would you organize them on an HTTP server?  Incidentally, HTTP is
> *not* an efficient transmission mechanism (less true if you're using
> HTTP 1.1).

If you have millions of large blobs you will have problems no matter
what - nothing I can say or do can worsen or lighten those ;-). I'm not
sure that I know the right solution to problems like that. I know of a
few sites where they regularly work with data like that, but they don't
store it in a relational database, they use custom solutions, and still
often store the blobs on the filesystem.

I know http isn't that efficient, but it allows you to a) offload the
demand to other servers very easily, and  b) present the textual data
already while the customer is still waiting for the picture. 

Anyway, while interesting, this is getting away from Python more
and more...

-- 

Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt


From fredrik at pythonware.com  Wed Jan 19 08:05:30 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: 19 Jan 2000 07:05:30 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jan 19)
Message-ID: <24CE8EEE9D554C14.DA42BAE82B20C70E.DDDF47DE33556CB7@lp.airnews.net>

    Last chance to register on-line for the Eighth International
    Python Conference:
    http://www.python.org/workshops/2000-01/

    The Software Carpentry announces a Python-oriented open
    source design competition with $100,000 in prize money (!):
        http://www.deja.com/getdoc.xp?AN=572973351
        http://www.software-carpentry.com

    The May 2000 issue of Linux Journal is going to be accompanied
    by a special edition Python Supplement! Sign up for your free
    copy at:
        http://www.linuxjournal.com/python

    On the newsgroup: Edward Welbourne proposes a unified
    namespace system:
        http://www.deja.com/viewthread.xp?AN=573542571
        http://www.chaos.org.uk/~eddy/dev/toy/python2.html

    Manuel Gutierrez Algaba has started a python web ring.
    Sign up your python pages here:
        http://www.ctv.es/USERS/irmina/pythonring.html

    Mark Adler announces an IDE in progress: VIPER (Visually
    Integrated Python EnviRonment)
        http://www.PinehillProducts.com

    Markus F.X.J. Oberhumer releases pysol 3.20 (now supports
    151 distinct solitaire games!):
        http://pysol.tsx.org

    Mark Lutz teams up with Perl folks to offer Python training
    for anyone. Next session in February:
        http://rmi.net/~lutz/boulder-classes.html

    Seen on the net: Andrew Kuchling looks at some of the recent
    Python books:
        http://starship.python.net/crew/amk/python/writing/books.html


========================================================================

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

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      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 <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From mal at lemburg.com  Sat Jan 22 10:42:38 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 22 Jan 2000 16:42:38 +0100
Subject: ANN: platform.py - Version 0.7.0
Message-ID: <3889CFEE.E2C0EF1E@lemburg.com>

ANNOUNCING:

                  platform.py - Version 0.7.0

        A Python module providing platform information APIs.


WHAT IT IS:

platform.py was intially developed for the mxCGIPython project
to automate the creation of file names including platform
specific data for users to be able to recognize and download
the appropriate Python interpreter binary for their (or their
ISPs) setup.

Since then the module has been extended and redesigned to
provide a generic interface for querying platform information
both in human readable form and for other application purposes
through platform dependent special APIs.

platform.py supports (at least) these setups: most Windows
setups, MacOS, various Linux distributions, various Unices,
OpenVMS, Java VM and DOS.


WHAT'S NEW ?

Version 0.7.0 is the first new version after the intial release
and update cycles last year.


WHERE CAN I GET IT ?

Documentation is provided through doc-strings and many
comments in the source code, which can be found via my
Python Pages:

        http://www.lemburg.com/~python/


WHAT DOES IT COST ?

It comes with a Python-style license, but is otherwise free for
commercial and non-commercial use.


WHERE CAN I GET SUPPORT ?

Commercial support for all my packages, tools and other
Python software will soon be available through my newly
founded software company. Keep an eye open on this list
for an announcement.


REFERENCE:

<P><A HREF="http://www.lemburg.com/~python/">
platform.py 0.7.0</A> - Generic platform information APIs.
(22-Jan-2000)

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





From hamish_lawson at yahoo.co.uk  Tue Jan 25 08:55:16 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Tue, 25 Jan 2000 13:55:16 GMT
Subject: newbie question about self and eval
References: <86jjkj$jmu$1@news.inet.co.th>
Message-ID: <86ka02$bg0$1@nnrp1.deja.com>

>             token = string.split(lines,'\011')
>             bucket = token[0]
>             if bucket =='fee':
>                 self.fee.append(token[1])
>             elif bucket == 'fi':
>                 self.fi.append(token[1])
>             elif bucket == 'fo':
>                 self.fo.append(token[1])
>             else:
>                 self.fum.append(token[1])

> I know there must be a more effective way of appending items into the
> various class lists, (which probably involves eval or something
> similar

Try the following:

    token = string.split(lines,'\011')
    eval("self.%s" % token[0]).append(token[1])


Hamish Lawson


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


From nic at csee.uq.edu.au  Sat Jan 29 20:43:20 2000
From: nic at csee.uq.edu.au (Nic Williams)
Date: Sun, 30 Jan 2000 01:43:20 GMT
Subject: Creating a local variable with a parameterized name
References: <38937075.18B75753@rochester.rr.com> <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <8704vo$soj$1@nnrp1.deja.com>

In article <vndk8ksmpox.fsf at betelgeuse.ccs.neu.edu>,
  Justin Sheehy <dworkin at ccs.neu.edu> wrote:
> Michael Ross <mross at rochester.rr.com> writes:
>
> > I want to create a local variable with a passed in name.
> > I realize I can do the following:
> > >>> def f(varname):
> >              cmd = '%s = 123' % varname
> >              exec(cmd)
> >              cmd = 'print %s' % varname
> >              exec(cmd)
> >
> > >>> f('mike')
> > 123
>
> As you have discovered, there are ways to do that, but none of them
> are pretty.  Are you sure that you really need the name to be a
> variable on its own, and that a dictionary wouldn't do?
[cut example]
> my_directory = {}
> my_directory[varname] = 123
> print my_directory[varname]
>
> Instead of creating a variable whose name is the value of 'varname',
> it makes a directory with a key equal to the value of 'varname'.
> There are very few situations of this type that I can think of where
> this would not be sufficient.
>
> Directories are handy, and particularly well suited to this sort of
> thing.  Why be any messier than you need to be?

Michael if you really need to insert the variable into the local name
space then be aware of the built-in functions global() and locals()
which return dictionaries holding the global and local symbol tables,
respectively. There is also vars([object]) which gives an object's
symbol table or (when no parameter is given) the local symbol table.

The result isn't much more attractive than Justin's but you do insert
the variable into the 'official' local space:

def f(name):
	locals()[name]=123
	print locals()[name]

Nic


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


From aerogems at netins.net  Thu Jan 13 18:03:29 2000
From: aerogems at netins.net (Scott Billings)
Date: Thu, 13 Jan 2000 17:03:29 -0600
Subject: best Template module
References: <85ljq1$mms$1@nnrp1.deja.com>
Message-ID: <387E59C1.B8901285@netins.net>

Well, you might try talking to Robin Friedrich, author of HTMLgen.
Particularly about the TemplateTools class. I've been doing a little
hacking of the copy I was so graciously sent, in the hopes of turning it
into a HTML pre-processor along the lines of Genpage (which is cool, only
it's written in Perl -- But, everyone's entitled a mistake here and there
;).

Right now, I have a pretty basic working program, which might do what you
want. Actually, I just wrote a bit of an interface to the TemplateTools
class (hopefully soon I can create a Tkinter based GUI over it). Anyway,
about all my program does, is feed the TemplateTools class the needed
parameters.

Looking over the "license" for the TemplateTools class, I don't see any
reason why I couldn't forward you a copy (with a few comments explaining
use). Email me at billings at eudoramail.com if you're interested.

-Scott-

> Hello,
>
> Could anyone recommend a good 'Template module'?
>
> Template module - a module that will let me create a plan text/HTML
> file with some special markup/tags that I can then replace by some text
> values from within my Python code
>
> good - fast execution and not too complex :)
>
> Sorry for this newbie question...
>
> Thanks,
>
> Otis
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



From pinard at iro.umontreal.ca  Sat Jan 22 20:44:21 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 22 Jan 2000 20:44:21 -0500
Subject: Small `du' reformatter :-)
In-Reply-To: Siggy Brentrup's message of "21 Jan 2000 03:16:45 +0100"
References: <oqogagfidt.fsf@titan.progiciels-bpi.ca> <8766womanm.fsf@baal.winnegan.de>
Message-ID: <oqk8l1ef4a.fsf@titan.progiciels-bpi.ca>

Siggy Brentrup <bsb at winnegan.de> ?crit:

> Fran?ois Pinard <pinard at iro.umontreal.ca> writes:

> > I knew I wrote a small Python tool to reformat `du' output [...]

> Quoting Tim Peters from another thread:

> Tim> [...] 99% of such a script ships with Python, in
> Tim> Tools/Scripts/pindent.py.  From the top of that file:

> s/pindent/dutree/

Yes, I took a look, thanks!

May I offer to replace the distributed script with mine?  You will get a
script which is shorter and more elegant Python.  Its output might also
be a bit better pruned and sorted, in my opinion.  Before:

        172 .
          |  21 src
          |  16 lib
          |  10 data
          |   |     4 .deps

After:

            172 \_ .
             10    \_ data
              4    |  \_ .deps
             16    \_ lib
             21    \_ src

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





From news at dorb.com  Fri Jan  7 16:27:07 2000
From: news at dorb.com (Darrell)
Date: Fri, 7 Jan 2000 16:27:07 -0500
Subject: does a variable exist
References: <B49BBF60.1373%Dan@Grassi.com>
Message-ID: <dQsd4.586$mj7.15972@typhoon.nyroc.rr.com>

"Dan Grassi" <Dan at Grassi.com> wrote in message
news:B49BBF60.1373%Dan at Grassi.com...
> I have forgotten and can not find the function (incantation) that can will
> determine if a variable currently exisis.

globals() and locals() are your best bet.

>>> import sys
>>> myVar=123
>>> sys.modules[__name__].__dict__.keys()
['sys', '__doc__', 'myVar', '__name__', '__builtins__']
>>>
>>> class A:
...     classVar=111
...     def __init__(self):
...             self._x=1
...             print self.__dict__.keys()
...
>>> a=A()
['_x']
>>> a.__class__.__dict__.keys()
['__init__', '__doc__', 'classVar', '__module__']
>>>
>>> def f1(aVar):
...     print locals().keys()
...
>>> f1(123)
['aVar']
>>>
>>> print globals().keys()
['__doc__', '__name__', 'A', '__builtins__', 'a', 'myVar', 'f1', 'sys']
>>>

--
--Darrell




From embed at geocities.com  Fri Jan 28 15:50:34 2000
From: embed at geocities.com (Warren Postma)
Date: Fri, 28 Jan 2000 15:50:34 -0500
Subject: Is there a Python/WindowsCE mailing list?
Message-ID: <ymnk4.7557$S64.220739@nnrp1.uunet.ca>

I'm interested in talking with others interested in the Windows CE port of
Python. I have contacted the author of the port, and I'm also curious who in
this newsgroup is using or planning to use Python on Windows CE.    Maybe we
can set up a mailing list if there isn't one already set up.

Warren Postma
ZTR Control Systems





From gmcm at hypernet.com  Mon Jan  3 08:02:11 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 3 Jan 2000 08:02:11 -0500
Subject: More grammar issues
In-Reply-To: <E12514Z-0005qV-00@ethereal>
Message-ID: <1265218848-24754805@hypernet.com>

Nolan Darilek asks:

> All right. I've begun translating a sizable subset of the
> standard Python grammar into Yacc. I'm building a parse tree with
> the output, but I'm again confused. Most rules have
> clearly-defined outputs. These near the end confuse me, though:
> 
> test: and_test ('or' and_test)* | lambdef
> and_test: not_test ('and' not_test)*
> ...

These are written this way so you'll get a chain instead of a 
branch. That is, instead of "AND (op1, AND (op2, op3))" you'll 
get "AND (op1, op2, op3)". You should also be aware that 
"and" and "or" return the value of the last operand evaluated.

- Gordon



From tilo.schwarz at daimlerchrysler.com  Mon Jan 31 13:51:48 2000
From: tilo.schwarz at daimlerchrysler.com (Tilo Schwarz)
Date: Mon, 31 Jan 2000 19:51:48 +0100
Subject: C++-Extension
Message-ID: <3895D9C4.858A291A@daimlerchrysler.com>

Hi all,

the

Extending and Embedding the Python Interpreter

reference says, that (1.11)

"If the main program (the Python interpreter) is compiled
and linked by the C compiler, global or static objects with
constructors cannot be used. This is not a problem if the main program
is linked by the C++ compiler"

Does that mean, C++-Extension are not "portable" - means, that I need
a C++-linked python interpreter to load my C++-Extension? Not
portable, because on a different platform I don't know if there is a
C++-linked python interpreter and some code might use global or static
objects with constructors. Am I right? (Hopefully not)

I'm asking, because I get the following error message:

> python
Python 1.5.2 (#1, Jul 23 1999, 06:38:16)  [GCC egcs-2.91.66
19990314/Linux (egcs- on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import _CircleDetector
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: ./_CircleDetectormodule.so: undefined symbol: cerr
>>>  

I'm not sure about cerr, I don't use it, but some STL-code...

Thanks for any help!

Bye
	Tilo

--------------------
Tilo Schwarz, DaimlerChrysler, Research Center Ulm, FT3/AB
Address:      Wilhelm-Runge-Str.11,  P.O. Box 23 60,  89013 Ulm, 
Germany
Phone:        +49 731 505 2376
Email:        tilo.schwarz at daimlerchrysler.com


From jam at quark.emich.edu  Fri Jan 21 09:40:31 2000
From: jam at quark.emich.edu (Jeff)
Date: Fri, 21 Jan 2000 09:40:31 -0500
Subject: database connection
In-Reply-To: <s8grjo5jeh697@corp.supernews.com>; from davidm@softdesigns.com on Fri, Jan 21, 2000 at 09:22:35AM -0500
References: <s8grjo5jeh697@corp.supernews.com>
Message-ID: <20000121094031.A12175@quark.emich.edu>

On Fri, Jan 21, 2000 at 09:22:35AM -0500, David Metzler wrote:
> I would like to create some py programs that connect to a database like
> mysql or postgres on either win 98 or Linux.
> I could not find the file to include. The python.org directions talk about
> compiling but I do not have MS C.
> Is there a mysqldb available to install in Win 98 without compiling.
> How about a rpm or binary for RH 6.0
> 
> If not for mysql perhaps another database?
> 

I cannot speak for win9x, but postgresql has a module for python with the
distribution. check http://www.postgresql.org/.. I would expect that mysql
has something similar as well. I use the postgresql module on a daily basis
on my website, and it works quite well.

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From wweston at worldnet.att.net  Tue Jan 11 18:08:08 2000
From: wweston at worldnet.att.net (wes weston)
Date: Tue, 11 Jan 2000 23:08:08 +0000
Subject: can't find tk.tcl
Message-ID: <387BB7D8.F1888924@worldnet.att.net>

Guys,
	At the command line prompt: "from Tkinter import *" then "w = Tk()" and
a "usable" tk.tcl can't be found in the several places looked at. But,
it's in one of them: C:\Progra~1\Tcl\lib\tk8.0.  The weirdest thing is
I've been running programs that use Tkinter recently on this
machine(WinNT Ws). It (Tkinter programs) works at work and on my
portable.

	I tried a fresh install in several different places. Help!

		wes


From gregm at iname.com  Sun Jan 23 06:45:44 2000
From: gregm at iname.com (Greg McFarlane)
Date: Sun, 23 Jan 2000 22:45:44 +1100
Subject: Tkinter/Pmw and focus events
In-Reply-To: <3887A383.5D921ADB@exceptionalminds.com>; from Timothy Grant on 20 Jan 2000 at 04:08:35PM
References: <3887A383.5D921ADB@exceptionalminds.com>
Message-ID: <20000123224544.13916@nms.otc.telstra.com.au>

On 20 Jan, Timothy Grant wrote:
> Back to my other pet subject at the moment, Tkinter. I've been using Pmw
> and having really good success with it, especially since I develop on
> Linux and my clients run on Win98! Pmw allows one to assign the
> attribute 'modifiedcommand' to many of its widgets. That function is
> called any time the widget is modified. Is there a similar attribute for
> got-focus and lost-focus events?
> 
> I'd like to implement something like is in Delphi/Paradox where when the
> cursor is outside the field the numbers are formatted all pretty
> ($1,234.56), and when the cursor is inside the field it is automatically
> reformatted to a 'normal' number (1234.56).

If you are using Pmw.EntryField, then you should use the bind method
of the entry component.  You could bind to '<FocusIn>' and
'<FocusOut>' (or maybe to '<Enter>' and '<Leave>'.  For example, this
changes the test to lowercase when the entry widget has focus and to
uppercase when it loses focus:

import string
import Pmw
 
def focus_in(event = None):
    entryfield.setentry(string.lower(entryfield.get()))

def focus_out(event = None): 
    entryfield.setentry(string.upper(entryfield.get()))

root = Pmw.initialise()
 
entryfield = Pmw.EntryField() 
entryfield.pack() 

entryfield.component('entry').bind('<FocusIn>', focus_in) 
entryfield.component('entry').bind('<FocusOut>', focus_out) 
 
root.mainloop()

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com



From kkreiner at carinthia.com  Sun Jan 16 10:16:49 2000
From: kkreiner at carinthia.com (Kreiner Karl)
Date: Sun, 16 Jan 2000 16:16:49 +0100
Subject: using threads with python and winnt
Message-ID: <85snka$3di@fstgal00.tu-graz.ac.at>

hi everone,
            does anyone have expiriences with using threads and winnt ? I
started working with the built-in module "thread" using
the function "start_new_thread" , which needs two parameters : a function
and a list of araguments to be passed to the
function. but no matter what I do , i just get the error-message:

type error: first arg must be callable

I consulted the python manual but it didn't provide help to my problem. so,
does anyone know the correct syntax of this
function and are there generally any problems with using the whole library
with windows nt ?

any help appreciated,

thx Charly






From PClaerhout at CREO.BE  Thu Jan 20 05:02:17 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 20 Jan 2000 11:02:17 +0100
Subject: How to differentiate Win98 from NT ?
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA7070@MSGEURO1>

Try Marc's platform.py module. Here is the URL:

http://starship.skyport.net/~lemburg/platform.py

Pieter

-----Original Message-----
From: Svante Kleist [mailto:Svante.Kleist at edt.ericsson.se]
Sent: Thursday, January 20, 2000 8:04 AM
To: python-list at python.org
Subject: How to differentiate Win98 from NT ?



Dear Pythonists,

"os.name" says "nt" for both Windows NT and Windows 98.

In order to differentiate between the two,
I could search for the existence of "cmd.exe",
but that feels rather clumsy. 

Better suggestions, please?

 / Svante Kleist
/NEMESIS systemDesign, Stockholm
-- 
http://www.python.org/mailman/listinfo/python-list

-------------- next part --------------
A non-text attachment was scrubbed...
Name: platform.py.url
Type: application/octet-stream
Size: 103 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000120/4d2205d4/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: platform.py.url
Type: application/octet-stream
Size: 103 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000120/4d2205d4/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: platform.py.url
Type: application/octet-stream
Size: 103 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000120/4d2205d4/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: platform.py.url
Type: application/octet-stream
Size: 103 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000120/4d2205d4/attachment-0003.obj>

From lalal at fd.com  Mon Jan 17 23:35:31 2000
From: lalal at fd.com (lalal at fd.com)
Date: Mon, 17 Jan 2000 23:35:31 -0500
Subject: make money while surfing!!!             11955
Message-ID: <170100233640@fd.com>

Everyone likes making easy money.  This place pays you 50 cents for every hour you browse the web!!!
Get even more if you refer people!  I just got a check for $36.00!!!

Sign up here:
http://www.alladvantage.com/go.asp?refid=IJE218







d2o(3,@]v*



From jr_dinero at latinmail.com  Sat Jan 15 19:32:25 2000
From: jr_dinero at latinmail.com (J.R.)
Date: Sun, 16 Jan 2000 00:32:25 GMT
Subject: EARN MONEY EASILY-READ THIS!!
Message-ID: <3881120A.E11F947E@latinmail.com>

JUST A FEW MINUTES.

NOTHING TO LOSE BY READING IT.

Visit this web page.
Just click on this link:


http://www1.gratisweb.com/jr_dinero/money.html




From mlh at vier.idi.ntnu.no  Thu Jan  6 11:29:05 2000
From: mlh at vier.idi.ntnu.no (Magnus L. Hetland)
Date: 06 Jan 2000 17:29:05 +0100
Subject: Simple cartesian product
References: <000801bf571c$0cf448a0$1a2d153f@tim> <iacembwr7qg.fsf@vier.idi.ntnu.no> <Pine.LNX.4.10.10001061110050.6929-100000@gaia.gfy.ku.dk> <20000106132958.E7015@xs4all.nl>
Message-ID: <iac9023qi3y.fsf@vier.idi.ntnu.no>

Thomas Wouters <thomas at xs4all.net> writes:

> On Thu, Jan 06, 2000 at 11:13:29AM +0100, Thomas Knudsen wrote:
[...]
> 
> So is this YYYY-MM-DD or YYYY-DD-MM ?

yyyy-mm-dd

Logical, considering other formats like hh:mm:ss, and decimal numbers
in general.

--

  Magnus
  Lie
  Hetland


From zeitlin at seth.lpthe.jussieu.fr  Wed Jan  5 13:23:14 2000
From: zeitlin at seth.lpthe.jussieu.fr (Vadim Zeitlin)
Date: 5 Jan 2000 18:23:14 GMT
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com> <386fa6b1.58568155@news.telus.net> <20000105085647.C9441@vislab.epa.gov>
Message-ID: <slrn877386.lok.zeitlin@seth.lpthe.jussieu.fr>

On Wed, 5 Jan 2000 08:56:47 -0500, Randall Hopper <aa8vb at yahoo.com> wrote:
>UNIX boxes ("configure", "make", "make install" - done).  Sun, SGI, DEC,
>FreeBSD, Linux, etcetc.

 Hello,

 please have a look at http://www.lpthe.jussieu.fr/~zeitlin/wxWindows/daily/
which contains the results of daily automatic rebuilds of wxWindows for
several platforms. Sun, FreeBSD, Linux, HP-UX and AIX seem to work fine
most of the time - as this page uses the current sources in the cvs server,
sometimes the things do break, but you will notice from "Last snapshot" column
that they get fixed quite quickly. And, at any rate, the upcoming 2.2 release
of wxWindows will surely build on all of these platforms without any tweaking.

>wxWindows/wxGTK/wxPython requires someone with experience and persistence
>to help it along on a standard SGI UNIX box, in compilation and
>configuration.

 There is one problem inherent to SGI: the native make doesn't support VPATH
(and never will according to SGI docs) while wxWindows makefiles require it,
so you must install GNU make. Except for this, however, there should be no
problems - if there are any, please contact us at
wxwin-developers at wx.dent.med.uni-muenchen.de. Unfortunately, we have less
resources than Tcl/Tk folks and I couldn't buy a dedicated SGI box for wxWin
testing yet <1.5 grin> so we can only rely on people having access to them to
send us bug reports.

 Regards,
VZ


From amused at webamused.com  Sun Jan 23 12:08:57 2000
From: amused at webamused.com (Joshua Macy)
Date: Sun, 23 Jan 2000 17:08:57 GMT
Subject: New install. Getting ImputError
References: <388AAE5A.B20A7BCE@slip.net>
Message-ID: <388B335E.3F9E0B16@webamused.com>

Thomas D'Urso wrote:
> 
> Just installed / started with Python and reading Learning Python. When I
> try to execute this code:
> 
>    import test
>    from test import vtest    # <-- error on this line
>    print vtest
> 
> file "test" contents:
> 
>    vtest = 'this is a test'
> 
> I get debugger error for 2nd line:    ImputError: cannot input name
> vtest
> The global variable says ImputError is:     Input can't fine module, or
> can't find name in module.
> 
> It give no other errors as far as opening the file. Working on a Mac G4
> with OS 9. Python 1.5.2.
> Doing the "import test.autotest" reports no errors.
> 
> TIA Thomas

The module that you're trying to import has to have the extension .py,
so if the file is actually named just "test" that'll be a problem. 
Otherwise, it works for me.

file test.py
============
vtest = 'This is a test'

Python 1.5.1 (#1, Apr  3 1999, 18:41:12)  [GCC egcs-2.91.60 19981201 (e
on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import test
>>> from test import vtest
>>> print vtest
This is a test

Joshua


From djmaecki at gmx.net  Mon Jan 17 13:56:31 2000
From: djmaecki at gmx.net (djmaecki da mp3masta)
Date: Mon, 17 Jan 2000 19:56:31 +0100
Subject: regular expressions
Message-ID: <20000117195631.A2020@gmx.net>

Hi out there!

Has anyone experience with regular expressions in python? Or is there a
a howto?

kindly regards
Thomas

-- 
/*----------------------------------------------------------------------------\
| In October of 1998 three Linux developers disappeared into the woods near   |
| Redmond, Washington, in an attempt to compile their kernels.         _      |
|                                                                     | |     |
| A year later their source code was found ...                       -----    |
|                                                                    < ? )    |
| THE BILL WITCH PROJECT                                             /   \    |
|                                                                  _(_\/__)   |
\----------------------------------------------------------------------------*/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000117/03f81e6b/attachment.sig>

From gmcm at hypernet.com  Sat Jan 15 18:19:30 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 15 Jan 2000 18:19:30 -0500
Subject: How to use Pythonwin's ability to access windows dll's???
In-Reply-To: <XX6g4.24$vz.3707@nnrp3-w.snfc21.pbi.net>
Message-ID: <1264144125-438654@hypernet.com>

Sam Schulenburg writes:

> I can access windows dll's using Sam Rushing's calldll.pyd and windll.py. My
> question is are thes features built into Pythonwin? 

Nope. Everything in the Win32 extensions is wrapped (much 
of it via SWIG).  Sam's stuff is regarded as dangerous, 
because it's too easy to crash Python.

Wimps.

real-men-juggle-chain-saws-ly y'rs



- Gordon



From tiddlerdeja at my-deja.com  Sun Jan 23 14:38:40 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Sun, 23 Jan 2000 19:38:40 GMT
Subject: COM and CreateObject and inheritence
References: <86d2dr$a9b$1@nnrp1.deja.com>
Message-ID: <86flbu$rl$1@nnrp1.deja.com>

I've now used the VB debugger and found that I get 99 returned from my
function bar:

x = c.bar

x now contains 99.

If I try to do anything with x:

MsgBox x

Then VB crashes.

My COM object seems to work fine from Python and Dispatch calls? Are
there other tests I could do to diagnose the problem? Could my
installation of VB be broken?

Any help apprecuated.

In article <86d2dr$a9b$1 at nnrp1.deja.com>,
  tiddlerdeja at my-deja.com wrote:
> I managed to get the HelloWorld COM Python example to work using
> CreateObject in VB. I could not get a Reference to Python.TestServer
so
> that I could:
>
> Dim Test and HelloWorld
>
> Instead I had to:
>
> Dim Test As Object
>
> Is it not possible to get a reference to it?
>
> Also, I tried to create my own test:
>
> foo.py
> -----
>
> class foo:
>     def __init__(self):
>         print 'init'
>
>     def bar(self):
>         return 99
>
> comfoo.py
>
> from foo import *
> class comfoo(foo):
>     _reg_clsid_ = "{3F90DE60-D0F5-11D3-96E4-9A059544BB30}"
>
>     _reg_progid_ = "My.Test"
>
>     _reg_desc_ = "COM Test"
>
>     _public_methods_ = ['bar']
>
>     def __init__(self):
>         print 'cominit'
>
> When I try to call 'bar' in VB it crashes. comfoo is registered and
both
> files are in my python path.
>
> Can someone tell me what I'm doing wrong?
>
> Any help appreciated.
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>


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


From andrew.henshaw at gtri.gatech.edu  Tue Jan 25 18:11:22 2000
From: andrew.henshaw at gtri.gatech.edu (Andrew Henshaw)
Date: Tue, 25 Jan 2000 18:11:22 -0500
Subject: Whitespace delimiters suck - Seen Occam?
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu> <86cmom$5pp$1@news-int.gatech.edu> <86h7jm$o7i@newton.cc.rl.ac.uk> <388C760A.1E66C20@gtri.gatech.edu> <86k8km$11p6@newton.cc.rl.ac.uk>
Message-ID: <388E2D9A.C9A62D60@gtri.gatech.edu>


Richard Brodie wrote:
> 
> "Andrew Henshaw" <andrew.henshaw at gtri.gatech.edu> wrote in message
> news:388C760A.1E66C20 at gtri.gatech.edu...
> 
> > That's interesting.  Everyone here, who has used it for any
> > reasonably-sized project, has really liked it.  I've seen opposition
> > from programmer's who try it for a day and then immediately
> > abandon it, however.
> 
> You are probably right, in that the projects were relatively small,
> and programmers didn't stick to it for long enough.
> 
> > Have you tried the Windows version of Origami from Elcom (it adds several
> > niceties, but the default color scheme is horrible)?
> 
> Windows? Colours? Linux? We stopped using Occam in about '92,
> so the state of the art might have improved since then ;) I can see
> a folding editor would be better if you had a GUI to drive it from.

I don't have much call for Occam anymore, but I still use Origami for
just about every programming project.  It supports a lot of other
languages by default, and it can be set up to support most others.  It
only took a couple of minutes to set it up for Python (including 
keyword highlighting).


From shang.remove_edu at st.jyu.fi.edu  Thu Jan 20 15:58:15 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Thu, 20 Jan 2000 22:58:15 +0200
Subject: How to store a function pointer in class?
References: <867j5p$qvq$1@mordred.cc.jyu.fi> <vndaem01s1r.fsf@betelgeuse.ccs.neu.edu> <867nbb$1ho$1@mordred.cc.jyu.fi> <vnd66wo1nz2.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <867ssq$a3c$1@mordred.cc.jyu.fi>

Justin Sheehy <dworkin at ccs.neu.edu> wrote in message
news:vnd66wo1nz2.fsf at betelgeuse.ccs.neu.edu...
> You could store the function inside a mutable class data member.

Yeah, I figured this out myself, but it still seems like an
unneccessary and ugly hack. So far the best I could come up with was
to make a wrapper class for a function. i.e.

class func:
    def __init__(self, f):
        self.f = f
    def __call__(self,*args,**keyw):
        return apply(self.f, args, keyw)

Now I can use:

>>> Test.class_attribute = func(test_func)
>>> Test.class_attribute(1)
test_func called with 1
>>> f = Test.class_attribute.f

Alternatively you could add __doc__, __code__ etc. members and pass
the func object to any code requiring a function.

> For instance:
>
> >>> class Test:
> ...   class_attribute = []
>
> >>> def test_func(x):
> ...     print "test_func called with",x

Thanks for your ideas!

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi




From Svante.Kleist at edt.ericsson.se  Thu Jan 27 07:49:31 2000
From: Svante.Kleist at edt.ericsson.se (Svante Kleist)
Date: Thu, 27 Jan 2000 13:49:31 +0100
Subject: How to differentiate Win98 from NT ?
References: <2B1262E83448D211AE4B00A0C9D61B03BA7070@MSGEURO1> <86du41$lv0$1@gxsn.com>
Message-ID: <MPG.12fa5d43d7d344f1989694@news.ericsson.se>

Not very solid, I'd say. In my case, NT lives in D:\nt40

/ Svante

In article <86du41$lv0$1 at gxsn.com>, matt at resonation.co.uk says...
> try :
>     if os.environ['windir']=="C:\\winnt":
>             # do whatever
>     else:
>             ....
> 
> 
> 


From andy at robanal.demon.co.uk  Sat Jan  1 16:59:51 2000
From: andy at robanal.demon.co.uk (Andy Robinson)
Date: Sat, 01 Jan 2000 21:59:51 GMT
Subject: FINANCIAL MODELLING PACKAGE IN PYTHON?
References: <841d0c$j0n$1@violet.singnet.com.sg> <387f6d37.25522789@news.demon.co.uk> <19991230065516.A778039@vislab.epa.gov> <W8TkoEATH4a4Ewpu@jessikat.demon.co.uk>
Message-ID: <387378c0.4549271@news.demon.co.uk>

Robin Becker <robin at jessikat.demon.co.uk> wrote:

>In article <19991230065516.A778039 at vislab.epa.gov>, Randall Hopper
><aa8vb at yahoo.com> writes
>>Andy Robinson:
>> |Anyone who wants to discuss this, come join the python-finance list at
>> |www.egroups.com/group/python-finance/.  It's been a bit quiet of late
>> |but that's where my stuff and Kevin's is being discussed..  
>>
>>Will do.  You ought to get the list sited on:
>>
>>     http://www.python.org/psa/MailingLists.html
>>
>>This is the first I'd heard of it.
>>
>I have some good analytics for Markowitz risk/return optimisation. Does
>this come into the area?

Absolutely!

- Andy


From mross at rochester.rr.com  Sat Jan 29 21:49:29 2000
From: mross at rochester.rr.com (Michael Ross)
Date: Sat, 29 Jan 2000 21:49:29 -0500
Subject: Creating a local variable with a parameterized name
References: <38937075.18B75753@rochester.rr.com> <vndk8ksmpox.fsf@betelgeuse.ccs.neu.edu> <8704vo$soj$1@nnrp1.deja.com>
Message-ID: <3893A6B9.E6963438@rochester.rr.com>


Nic Williams wrote:

> In article <vndk8ksmpox.fsf at betelgeuse.ccs.neu.edu>,
>   Justin Sheehy <dworkin at ccs.neu.edu> wrote:
> > Michael Ross <mross at rochester.rr.com> writes:
> >
> > > I want to create a local variable with a passed in name.
> > > I realize I can do the following:
> > > >>> def f(varname):
> > >              cmd = '%s = 123' % varname
> > >              exec(cmd)
> > >              cmd = 'print %s' % varname
> > >              exec(cmd)
> > >
> > > >>> f('mike')
> > > 123
> >
> > As you have discovered, there are ways to do that, but none of them
> > are pretty.  Are you sure that you really need the name to be a
> > variable on its own, and that a dictionary wouldn't do?
> [cut example]
> > my_directory = {}
> > my_directory[varname] = 123
> > print my_directory[varname]
> >
> > Instead of creating a variable whose name is the value of 'varname',
> > it makes a directory with a key equal to the value of 'varname'.
> > There are very few situations of this type that I can think of where
> > this would not be sufficient.
> >
> > Directories are handy, and particularly well suited to this sort of
> > thing.  Why be any messier than you need to be?
>
> Michael if you really need to insert the variable into the local name
> space then be aware of the built-in functions global() and locals()
> which return dictionaries holding the global and local symbol tables,
> respectively. There is also vars([object]) which gives an object's
> symbol table or (when no parameter is given) the local symbol table.
>
> The result isn't much more attractive than Justin's but you do insert
> the variable into the 'official' local space:
>
> def f(name):
>         locals()[name]=123
>         print locals()[name]
>
> Nic
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

Nic, my initial thought was the same as yours, until I ran across this in
Section 4.1 of the Python Ref Manual, referring to locals() :

" ... The built-in functions globals() and locals() returns a dictionary
representing the current global and local namespace, respectively. The
effect of modifications to this dictionary on the namespace are
undefined."

I found that changing the dictionary returned by locals() works in an
interactive shell, but not when the code is in a module.

This is what made me dig into the Python source... It seems that
internally Python holds the variables for the local frame in a different
data structure than a dictionary.  locals() simply converts this to a
dictionary for convenience.

Justin, thanks for your answer... but I'm trying to pre-condition an
environment that will be used by non-programmers, so I was trying to avoid
the dictionary syntax.

- Mike



From a.eyre at optichrome.com  Thu Jan 13 08:56:11 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Thu, 13 Jan 2000 13:56:11 -0000
Subject: converting tuple to list
In-Reply-To: <200001121522.PAA15694@lsls4p.lsl>
Message-ID: <000701bf5dcd$f2b6d350$3acbd9c2@peridot.optichrome.com>

> list(sequence) -> yields a list with the same entries as sequence
> tuple(sequence) -> yields a tuple, likewise

Just out of interest: what is the canonical way of converting a list
of tuples to a dictionary? i.e. the opposite of dict.items()

e.g.

[("key1","value1"),("key2","value2")] -> {"key1":"value1","key2":"value2"}




From wlfraed at ix.netcom.com  Sat Jan 22 00:25:27 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Fri, 21 Jan 2000 21:25:27 -0800
Subject: Python Rocks!
References: <200001211334.IAA28640@python.org> <86ajlq$661@news.or.intel.com>
Message-ID: <zTiJOIVtImux0kGF7t2TN5w=5HDX@4ax.com>

On Fri, 21 Jan 2000 13:38:40 -0800, "tye4" <tye4 at yahoo.com> declaimed
the following in comp.lang.python:

> Yes, I think I'll let you guys wallow in your ignorance.
>
	I have to wonder, ignorance of what? Programming language
development?

	I've programmed in three flavors of Fortran (IV, 77, 90), a 70's
era COBOL, at least 5 variants of BASIC, (A)REXX, C (pre- and post-
ANSI), an early C++, Pascal, toy programs in APL and LISP, and a few
classes on Ada (but no real programs). Python is just another language
in which certain decisions were made in the syntax... One of which just
happens to produce relatively legible programs vs the run-on sentences
of the COBOL, the spaghetti of early BASIC, the inconsistent usage of
";" in Pascal, the cryptic condensing of APL (and PERL, IMHO), the
inconsistent bracketing in Pascal and C (single statement if/else
branches do not need brackets), and the confusion from white-space
indifference of Fortran.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From s.schwarzer at ndh.net  Wed Jan 19 20:55:06 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Thu, 20 Jan 2000 02:55:06 +0100
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com>
Message-ID: <38866AFA.C72FD6C1@ndh.net>

Hello tye

tye4 schrieb:
> Don't get me wrong. Indentation tremendously improves readability. However,
> it should not be part of the language. That way, we don't have to put an
> immense amount of effort just trying get everything into the right block.

I don't have (and never had) to invest a "immense amount of effort"
to keep my Python code right. To me it looks you try to mold Python
into the shape of the (from your perspective) "normal" languages you
are more familiar with.

The solution to this is not, however, to change Python, but (for you)
to change your language (that is, not to use Python).

> There is not single programming language that forces such a bizarre
> indentation rule.

At first glance I thought like you, but after a while I discovered
that this "bizzare rule" is not only bizzare but a strong feature
of the language. If you open your eyes, some "bizzare" things may
enrich your life ;-) .

> If thousands of programmers have a problem with this, why isn't it fixed.

Maybe because hundreds of thousands programmers like it this way? :-)

Regards
 Stefan


From amused at webamused.com  Wed Jan 26 22:34:29 2000
From: amused at webamused.com (Joshua Macy)
Date: Thu, 27 Jan 2000 03:34:29 GMT
Subject: Creating Objects
References: <7g8v8s01mjtouok31d7elm09bvqc184tu7@4ax.com>
Message-ID: <388FBA9D.74FB0317@webamused.com>

Calishar wrote:
> 
> Hi folks,
> 
>   I'm working on solving a problem that my boss put to me the other
> day, and I have run into a solution that I don't know how to implement
> in Python (or any other language for that matter)
> 
> Basic description of problem (particulars not included)
> 
> Work out a system that can import information from (any number) of
> systems, and write the infomration out (meaningfully) to any system in
> the set.
> 
> I.e.
> Assume a set a set of systems [A-G] the system needs to be able to
> import from any of the seven systems, and write out it back out to any
> of the seven. Ther may or may not be multiple instances of each system
> type.
> 
> What I have in mind is defining a set of base classes and defining the
> actual implementation at run-time
> 
> 

  You haven't said what sort of information the systems provide, but I
would probably approach this by writing a class that represents a
canonical intermediate format, and then a set of translator classes,
each of which is only responsible for translating from a particular
system to the intermediate format or from the intermediate format to the
particular system: conversion can then be perfomed between any two
systems in two steps. That's only nine classes for an 8 system suite
(assuming that you give one class the responsibility for both the to and
the from translation, which seems reasonable in the abstract), and
adding a new system to the suite requires only writing one more to/from
tranlator class.  Deciding just what goes into the intermediate format
and how to represent it could be tricky, but assuming that what your
boss wants is remotely sensible (there is some kind of natural
correspondence of the information of any system to any other system),
you probably already have a pretty good idea of what it might look
like.  I would be tempted to defer deciding absolutely what that entails
by using something like XML as the intermediate format, since I know
that I can get parsing and nicely walkable trees practically for free
using Python's XML tools.

  Joshua


From bryan.olson at uptronics.com  Wed Jan  5 19:40:35 2000
From: bryan.olson at uptronics.com (bryan.olson at uptronics.com)
Date: Thu, 06 Jan 2000 00:40:35 GMT
Subject: Simple cartesian product
References: <000801bf571c$0cf448a0$1a2d153f@tim> <iacembwr7qg.fsf@vier.idi.ntnu.no>
Message-ID: <850oa1$pom$1@nnrp1.deja.com>

Magnus L. Hetland wrote:
> "Tim Peters" <tim_one at email.msn.com> writes:
>
> > Search in DejaNews for p13x25.
> >
> > c'mon-folks-new-millennium-new-topics<wink>-ly y'rs  - tim
>
> Now - you're not going to get that "new millennium"-discussion
> started, are you? We are still in the same millennium, century, and
> decade as last year. The third millennium starts with 2001.01.01.

The universe is about 12-billion years old, so
"second millennium" is off by a factor of 6,000,000,
while "third" only by 4,000,000.

Anyway, a thread back in the 19-hundreds was about
the product of any number of lists (well, the number
of lists has to be a non-negative integer).  For
clarity-over-efficiency I like,

def cartesian(*listList):
    if listList:
        result = []
        prod = apply(cartesian, listList[:-1])
        for x in prod:
            for y in listList[-1]:
                result.append(x + (y,))
        return result
    return [()]


--Bryan




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


From fredrik at pythonware.com  Fri Jan 21 03:48:28 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 09:48:28 +0100
Subject: Bug in Calendar Module
References: <22E71DAEC504D111B78100805FFE9DC72F03CAD4@pfs21.ex.nus.edu.sg>
Message-ID: <01e901bf63ec$52bc5850$f29b12c2@secret.pythonware.com>

Liew Pol Yee <polyp at cyberway.com.sg> wrote:
> Hello! Does a fix for the Calendar Module calendar.py exist? Currently, i
> get errors when I do the following:
>
> import calendar
> new = calendar.monthcalendar(2038,2)
> 
> It seems to be unable to accept dates after January, 2038.
> Thanks for any help!

that's a Unix problem (time_t is 32 bits), not a
Python problem per se.  

search the web for "2038" for more info.

(and sure, if Unix doesn't fix this before then, I
promise we'll work around that bug in Python)

if you really need it, I'm pretty sure you can combine
calendar.py with Demo/classes/Dates.py into something
that (literally) works until the end of time.

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! &smiley; -->




From gawron at obop.com.pl  Mon Jan 10 08:14:57 2000
From: gawron at obop.com.pl (=?iso-8859-2?Q?Przemys=B3aw?= G. =?iso-8859-2?Q?Gawro=F1ski?=)
Date: Mon, 10 Jan 2000 14:14:57 +0100
Subject: class destructor
Message-ID: <3879DB51.B9928BCE@obop.com.pl>

Is there 'automatic' destructor in python, like AClass::~AClass() in C++
?

Thanks Przemek

--
Przemyslaw G. Gawronski UIN:8358522
mailto:gawronskip at usa.net
mailto:gawron at obop.com.pl





From koffie_ at ddsNOSPAM.nl  Sun Jan  2 11:08:55 2000
From: koffie_ at ddsNOSPAM.nl (Pieter)
Date: Sun, 2 Jan 2000 17:08:55 +0100
Subject: a word of advice
References: <slrn86mrsc.2to.c.kotsokalis@ajax.noc.ntua.gr>
Message-ID: <84ns93$olc$1@tesla.a2000.nl>

I agree with that,  New Redhat releases SUCK BIGTIME.
So called, user-friendly is making the finetuning of the Linuxbundle
complicated. If you do a standard install apache and php, you still have to
modify the httpd.conf and crap like that, with phyton the same shit.



Constantinos A. Kotsokalis wrote in message ...
>I do not know if this has been said in any of these two groups
>before, but here's a bit of advice: If you try to compile
>ctsybase python module to use with Sybase under Linux (or
>any other OS, for that matter) and get a core dump when you
>try to import it (doesn't matter if it's statically linked
>or a dynamic module), make sure your locale (LC_ALL env.
>variable) is available in sybase's locales file
>(locales/locales.dat). It took me some three hours yesterday
>night to locate the problem (thanks, redhat, for stripping
>the python executable and cut the way to valuable debugging
>information :-)). If your locale is not available, either
>change locale or add it to the locales file as an ``alias''.
>
>  Regards,
>    Costas
>
>PS: I wonder why isql worked...
>
>--
>Constantinos A. Kotsokalis || c.kotsokalis at noc.ntua.gr
>National Technical University of Athens
>Network Management Center
>Tel. No: +30 1 7721861




From tseaver at aftershock.neosoft.com  Wed Jan 12 11:47:36 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 12 Jan 2000 10:47:36 -0600
Subject: class methods?
References: <20000112145448.29699.qmail@web804.mail.yahoo.com>
Message-ID: <324ECD3BD15B45F8.92AF8FDA2C29CCE6.9136D4F134428F5F@lp.airnews.net>

In article <20000112145448.29699.qmail at web804.mail.yahoo.com>,
Stuart Zakon  <sjz18 at yahoo.com> wrote:
>Is there no such thing as class scope methods in Python? (Analogous to static
>methods in Java, C++).
>
>I have a class scope dictionary that I am using to collect statistics on method
>invocations of the class instances and I wanted to have a class method to print
>them out. 
>
>However, Python seems to want a class instance to call a method. Does every
>method on a class need self as a parameter?
>
>Certainly I could make a separate statistics class and instantiate it once;
>however, the nature of the problem is such that it is more natural for the
>statistics collection to be private to the original class.
>
>If I am missing something, I apologize, but coming from Java and C++ it just
>seemed natural to have this feature. If there can be class attributes, then why
>not class methods? My understanding is that this is natural for Smalltalk as
>well.

This one surprised me too, coming to Python from C++/Java.  The usual response
in Python is that class methods solve a problem (a need to access "private"
class data) which doesn't exist in Python -- class members are always
accessible.  The usual cognate is to declare a module-scope function
which does the work of the class method.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From News at Titanic.co.uk  Tue Jan  4 09:14:23 2000
From: News at Titanic.co.uk (Simon Faulkner)
Date: Tue, 04 Jan 2000 14:14:23 GMT
Subject: CGI Script using Python
References: <D1bc4.40$hJ1.1161@news.rdc1.tx.home.com> <20000103192528.A19258@quark.emich.edu>
Message-ID: <l3047s43sjp65fg0n8561r7uke8f26iv3p@4ax.com>

Just for my info,

why?

>>
>>     print "Content-type: text/html\n"
>>     print "<H3>"
>>     print "TITLE"
>>     print "</H3>"

works OK for me.

is it flawed? (I know it's not perfect :-)

Simon




On Mon, 3 Jan 2000 19:25:28 -0500, Jeff <jam at quark.emich.edu> wrote:

>On Tue, Jan 04, 2000 at 12:20:51AM +0000, Paul E. Bible wrote:
>> Hi there,
>> 
>> I'm currently developing a CGI script using Python under Apache Web Server.
>> I've written a simple script, placed it in the /cgi-bin subdirectory and
>> perform a chmod 755 on it.  Here is the script:
>> 
>> -----------------------------------------------------------
>> #!/usr/local/bin/python
>> import os, string, sys
>> from cgi import *
>> 
>> def Main():
>>     print "Content-type: text/html\n\n"
>>     print "<HTML>"
>>     print "<HEAD>"
>>     print "This is a test"
>>     print "</HEAD>"
>>     print "<BODY></BODY>"
>>     print "</HTML>"
>>     sys.exit(0)
>> 
>> Main()
>
>change the above to:
>
>print "Content-Type: text/html"
>print
>print "<HTML>"
># continue with the rest of the script. notice extra 'print' above
>
>hope that helps.
>
>regards,
>J

Simon Faulkner


From composzioni at tiscalinet.it  Tue Jan 11 00:32:31 2000
From: composzioni at tiscalinet.it (Giovanni Riccardi)
Date: Tue, 11 Jan 2000 06:32:31 +0100
Subject: Newby question: how manage user input??
Message-ID: <85efp1$6pi$1@pegasus.tiscalinet.it>

I want to make a simple script that print out
some text (actually questions), get what the user
wrote at the displyed prompt and make some calculation
with that.
I tried to implement the user input with stdin or raw_input but after
returning the value the function I wrote doesn't
get back to the main loop??

Some refs (with examples) on the net for this problems?
Or could someone explain it to me??

Thanx in advance
Giovanni

-------------------------------------------
GIOVANNI RICCARDI * TERRACINA (LT) * ITALY

EMAIL: composizioni at tiscalinet.it
  ICQ: 40670340
-------------------------------------------





From P.J.W.S.VRIJLANDT at INT.azg.nl  Thu Jan 27 05:38:42 2000
From: P.J.W.S.VRIJLANDT at INT.azg.nl (P.J.W.S. Vrijlandt)
Date: Thu, 27 Jan 2000 11:38:42 +0100
Subject: Keywordargument in string.split
In-Reply-To: <1D9EA098C@nds-3.azg.nl>
Message-ID: <13B8DD29BC@nds-3.azg.nl>

I wrote:

> I have this error I don't quite understand.
> Why can't I give 'sep' as a keyword argument?
> 
> It seems that the string module I import, is not the string.py in the 
> standard distribution but a not-quite-equivalent c-implementation of 
> it.

I got replies pointing to 'strop', which is a c-implementation of 
string, and is imported by string.py

But what is the header (def) of split in strop? 
(and why is it different)



> 
> 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 string
> >>> string.split('abcd', sep = 'c') 
> Traceback  (innermost last):
>   File "<pyshell#1>", line 1, in ?
>     string.split('abcd', sep = 'c')
> TypeError: this function takes no keyword arguments
> >>> def split(s, sep=None, maxsplit=0):  # = string.split
>            print s, sep, 
> maxsplit >>> split('abcd', sep = 'c')
> abcd c 0
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 
Met vriendelijke groet,

Patrick Vrijlandt



From brunson at level3.net  Fri Jan  7 13:34:34 2000
From: brunson at level3.net (Eric Brunson)
Date: Fri, 07 Jan 2000 11:34:34 -0700
Subject: Converting an intergar to a string
References: <19991231140532.A8331@dark.net> <082a01bf533f$a61d3400$0100a8c0@rochester.rr.com> <19991231164828.A8614@dark.net> <386EE5E2.4E9F7B38@wjk.mv.com>
Message-ID: <387631BA.1180319F@level3.net>

More flexible would be to start using the % construct early and get
used to it.  It's very powerful.

print("Some Text %d Some more Text %s blah %3d" % (num1, string, num2))

"Wm. King" wrote:
> 
> Hi: As I am relatively new to Python, I had asked basically the same
> question some time ago... Here is what I found:
> 
> print "Some Text " + str(num) + "Some text"
> 
> Hope this helps.
> 

-- 
Eric Brunson                * _ o  *
brunson at brunson.com        * / //\         Faster and faster,
brunson at l3.net                \>>|   *       until the thrill of speed
                               \\,             overcomes the fear of
death


From psoares at consiste.pt  Wed Jan  5 09:48:10 2000
From: psoares at consiste.pt (Paulo Soares)
Date: Wed, 5 Jan 2000 14:48:10 -0000 
Subject: using python to press keys
Message-ID: <51998F46F8B7D011BF7C0060B03502AD2F3577@orion.consiste.pt>

If you are using Windows search the Python message archives for sendkeys. It
has other functions to ease automation (not the OLE kind).

Best Regards,
Paulo Soares

> -----Original Message-----
> From:	Black Mantis [SMTP:themantis at trojanslair.zzn.com]
> Sent:	Wednesday, January 05, 2000 4:18
> To:	python-list at python.org
> Subject:	using python to press keys
> 
> Can you use python to "press down" keys? If so, how would you do that?
> 
> thanks
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From benwittner at mediaone.net  Mon Jan 17 07:43:40 2000
From: benwittner at mediaone.net (Ben Wittner)
Date: Mon, 17 Jan 2000 12:43:40 GMT
Subject: conflict between my package and os standard module
References: <Dkcg4.19933$zU5.216301@wbnws01.ne.mediaone.net> <38823BF3.40D3AB96@easics.be> <20000117095108.C21652@xs4all.nl>
Message-ID: <08Eg4.25980$zU5.258198@wbnws01.ne.mediaone.net>

Thomas,

Thanks for replying.

> Dont forget that you can replace:
>
> import os # you want the 'enhanced' os, you get the standard one
> os.mygetuser(...)
>
> with:
>
> import myproj_os
> os = myproj_os
>
> To avoid rewriting too much code. (add a 'del myproj_os' if you want to
> avoid pollution.)

Cleaver trick.  I'll try to remember it.
Actually, though, my problem is the reverse; in my_proj/os.py, I don't have
access to the standard os services because when I say
import os
in my_proj/os.py, I get my_proj/os.py instead of the standard os.

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

That's pretty darn funny.  :)

-Ben

benwittner at yahoo.com






From hamish_lawson at yahoo.co.uk  Fri Jan  7 12:09:52 2000
From: hamish_lawson at yahoo.co.uk (Hamish Lawson)
Date: Fri, 07 Jan 2000 17:09:52 GMT
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <387400DB.3C683629@callware.com>
Message-ID: <8556kj$vn6$1@nnrp1.deja.com>

Ivan Van Laningham <ivanlan at callware.com> wrote:

> Guido is not listed as the co-author.  It says clearly "Foreword by
> Guido van Rossum, Creator of Python" right there on the cover.

I suspect that amazon.com may be the culprit for causing the confusion.
They list the book's authors as

   David M. Beazley, Guido Van Rossum

For other books I've seen Amazon listing the writer of the foreword as
a co-author.

Hamish Lawson


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


From thomas at xs4all.net  Mon Jan 31 08:44:12 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 31 Jan 2000 14:44:12 +0100
Subject: nesting class definitions?
In-Reply-To: <871vqt$a8d$1@nntp8.atl.mindspring.net>; from aahz@netcom.com on Sun, Jan 30, 2000 at 06:27:41PM +0000
References: <20000127155551.A2069@stopcontact.palga.uucp> <20000128182031.C16093@xs4all.nl> <871vqt$a8d$1@nntp8.atl.mindspring.net>
Message-ID: <20000131144411.A19725@xs4all.nl>

On Sun, Jan 30, 2000 at 06:27:41PM +0000, Aahz Maruch wrote:

> Yup.  But try this one on for size:

> def SpamGen ( foo ):
> 	class Spam:
> 		bar = foo
> class Viking:
> 	def __init__(self, foo):
> 		self.spamClass = SpamGen(foo)

Er, correct me if i'm wrong, but shouldn't this generate a NameError (asside
from SpamGen returning None, of course) ? Or did you intend to show that
this is very clear and very useful code, but impossible at the moment ?

> (Yeah, this is essentially meta-class stuff, which I know you know
> about, but I think presenting it slightly differently like this will
> make sense to more people.  It certainly makes more sense to me --
> metaclasses usually make my head hurt.  ;-)

Ssshhht, not so loudly, I've been wrestling with sendmail all last week...
my head still hurts enough, thank you very much ;) I still don't quite
understand Metaclasses, just as I dont quite understand closures - I think I
know what they are, but if they really are what I think they are, they are
really simple concepts, and I can't figure out why everyone is making such a
big fuss about them. Perhaps it is because I've never _needed_ those
concepts, I've always (in my entire 1 year experience with python) been able
to do everything I want using the simple and elegant namespace & scoping
rules used by Python.

The absolute greatest thing about Python, for me, is that everything I care
about (almost everything there is) is immediately clear, incredibly
intuitive, and comes naturally when writing Python. The only problem with
that is that I don't quite trust it, still. I'll be happily writing away,
and suddenly notice that I try to do something that might not work (force of
habbit, I still write C and Perl more often than Python) and fire up the
interpreter to test it out. I have yet to find something that didn't behave
as I expected ;)

Please, whatever else changes, whatever crufty C++ habbits get added,
whatever Java-lookalike constructs are thought up, whatever new truly
pythonic features are inserted, dont change that aspect of Python.

Loving-(Python)-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From bjorn at roguewave.com  Mon Jan 17 18:00:51 2000
From: bjorn at roguewave.com (bjorn)
Date: Mon, 17 Jan 2000 16:00:51 -0700
Subject: re-opening stdin in raw (binary) mode?
References: <388388c9.12001026@news.iol.ie> <slrn8876la.bos.kc5tja@garnet.armored.net>
Message-ID: <38839F22.28E03FA9@roguewave.com>


"Samuel A. Falvo II" wrote:

> In article <388388c9.12001026 at news.iol.ie>, Sean Mc Grath wrote:
> >The subject says it all. Can some kind soul shed light
> >on how to re-open sys.stdin in raw mode?
>
> If memory serves me correctly, I don't have the complete answer you're
> looking for, but under Linux, you can perform an ioctl() on file descriptor
> 1 (or is it 0?) to set a binary-mode flag.

Don't you just close it, and then posix.fdopen it (I can't remember if it's 0
or 1 either...)

-- bjorn




From sschen at ux8.cso.uiuc.edu  Sat Jan 15 07:57:01 2000
From: sschen at ux8.cso.uiuc.edu (steve s chen)
Date: Sat, 15 Jan 2000 12:57:01 GMT
Subject: Does python support long longs?
Message-ID: <x8_f4.26$tX3.648@vixen.cso.uiuc.edu>

Our organization is on the verge of switching to Python.  In working
with it in the last few days, we've been unable to do certain things.
I was wondering if perhaps modules or patches are out there to
do these?

1) Does Python support long long datatypes?  
2) Does Python support unsigned 32-bit ints?
3) A less significant issue -- is there a utility that analyzes 
	 a segment of python code verifying the dynamic name-bindings?
	 NameError exceptions weren't being reported until run-time;
	 does that mean we must QA each control path to ensure 'correct'
	 execution?

Thanks!
Steve
steve at schen.net


From sanderson at ttm.com  Fri Jan 21 10:45:54 2000
From: sanderson at ttm.com (Scott Anderson)
Date: Fri, 21 Jan 2000 10:45:54 -0500
Subject: A modest proposal
References: <6851DA17DBCA29508525686C00031BBB.00031C238525686C@ttm.com>
Message-ID: <38887F32.3EE0D4F9@ttm.com>

Good Lord, people.

All of these issues (readability, etc.) were solved long ago with
INTERCAL.

PLEASE-DON'T-FORGET-IT-'ly y'rs,

-scott

ivanlan at callware.com wrote:
[...]
> Martijn Faassen wrote:
[...]
> > > Therefore, I plan to redesign
[...]
> No, no, it *is* the core language.  No more def.  Instead, you will have
> to use the glyph for God K.  You will have to write your variable names
> in the Mayan syllabary.  All numbers will be modified base 20 by default
> (if you prefix a number with the "elephant" glyph, however, you will be
> able to use Roman numbers).  No more reserved words, only reserved
> glyphs.



From jody at sccsi.com  Sat Jan  1 18:26:10 2000
From: jody at sccsi.com (Jody Winston)
Date: 01 Jan 2000 17:26:10 -0600
Subject: Python and GUI options
References: <84lreq$hib$1@nnrp1.deja.com> <84lv5k$rqv$1@news1.xs4all.nl>
Message-ID: <wkr9g1gyr1.fsf@sccsi.com>

boud at rempt.xs4all.nl (Boudewijn Rempt) writes:

> PyQt has the disadvantage that you need to spend money if you want to
> run it on Windows - PyKDE and PyGnome obviously don't run on Windows at
> all. WxPython depends on a lot of other stuff - WxWindows, and GTK or
> Motif on Unix, and it can be tough to install everything; more difficult
> still to demand from users that they go the same route. I don't really
> like the widgets offered by Fox, on which FxPy is based - they look a
> bit old-fashioned. 

On the SGI wxWdindow and GTK still are not as stable as either Linux
or Windows 98
-- 
Jody Winston


From moonseeker at my-deja.com  Mon Jan  3 05:27:36 2000
From: moonseeker at my-deja.com (moonseeker at my-deja.com)
Date: Mon, 03 Jan 2000 10:27:36 GMT
Subject: pickle != cPickle
References: <Q2upQCAKRjb4EwQS@jessikat.demon.co.uk>
Message-ID: <84ptie$v5d$1@nnrp1.deja.com>

In article <Q2upQCAKRjb4EwQS at jessikat.demon.co.uk>,
  Robin Becker <robin at jessikat.demon.co.uk> wrote:
> I notice that cPickle and pickle sometimes give different results. This worries me as I thought
> these two were now strictly comparable. What sort of things can cause this? It appears they
> still give the same answer on unpickle.

Hi!

As you can find in the documentation cpickle can produce different pickled
data but the unpickled data is strictly the same you dumped. You can use
pickle to save data and cpickle to unpickle it.

Regards,
Mike


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


From garabik at melkor.dnp.fmph.uniba.sk.spam  Wed Jan  5 04:22:50 2000
From: garabik at melkor.dnp.fmph.uniba.sk.spam (Radovan Garabik)
Date: 5 Jan 2000 09:22:50 GMT
Subject: popen and spawnv: losing the battle with Win98.
References: <3872C9AC.C4414F27@es.co.nz>
Message-ID: <947064114.762328@cdwork.cvt.stuba.sk>

Matthew Miller <matthewm at es.co.nz> wrote:
 : Hi Everyone,

 : Problem#1) I've tried and tried but I cannot get
 : os.popen('command','w').write('data') under Win98 to accept data from
 : the pipe - it justs ignores it. HELP!

 : Problem#2) Likewise, I cannot get spawnv() to work under Win98. Could
 : someone give me a simple example.

I was biten by this some time ago.
The answer is to use the full path:
exefile=r"f:\python\pythonw.exe"
os.spawnv(os.P_NOWAIT, exefile, (exefile, r"j:\programs\program.pyw", frm, to, message))

-- 
 -----------------------------------------------------------
| Radovan Garabik http://melkor.dnp.fmph.uniba.sk/~garabik/ |
| __..--^^^--..__    garabik @ melkor.dnp.fmph.uniba.sk     |
 -----------------------------------------------------------
Antivirus alert: file .signature infected by signature virus.
Hi! I'm a signature virus! Copy me into your signature file to help me spread!


From loewis at informatik.hu-berlin.de  Tue Jan  4 15:14:31 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 04 Jan 2000 21:14:31 +0100
Subject: not pickling dicts?
References: <84ptmq$v65$1@nnrp1.deja.com>
Message-ID: <p6qwvppegrc.fsf@informatik.hu-berlin.de>

moonseeker at my-deja.com writes:

> I'm trying to pickle an object with a dictionary in it, but it seems that the
> data in the dict isn't pickled (or unpickled).

That must be a different problem. The dictionary contents should get
pickled, unless __getstate__ is defined and returns an empty
dictionary, or some other user-defined mechanism applies (setstate,
getinitargs, etc).

> Another question:
> Can I have shelve to use cpickle to improve performance?

shelve will automatically use cpickle if available.

Regards,
Martin


From pbible at home.com  Mon Jan 24 19:33:12 2000
From: pbible at home.com (Paul E. Bible)
Date: Tue, 25 Jan 2000 00:33:12 GMT
Subject: Python CGI and Accessing MS QL Server Database
References: <86ioek$iek$1@lure.pipex.net>
Message-ID: <cb6j4.823$0v3.5834@news.rdc1.tx.home.com>

How about trying mxODBC
(http://starship.python.net/crew/lemburg/mxODBC.html).
I found that it's quite easy to install and works great.  Besides, the Win32
ODBC module, according
to the Python Website, is not currently being maintained by anyone.

Using mxODBC, I found that the following line will create a connection to a
database within a CGI script:

 # Open connect to Database via ODBC
 db = Odbc.Windows.DriverConnect('DRIVER={Microsoft ODBC for
Oracle};SERVER=Test;UID=userid;PWD=password;')

Obviously, I'm using the ODBC driver for Oracle, so replace this driver with
SQL Server.

Good Luck,
Paul

"Navtej Riyait" <Navtej.Riyait at ukgateway.net> wrote in message
news:86ioek$iek$1 at lure.pipex.net...
> Hello,
>
> I am trying to create a little test web application in python/HTML using
> IIS. I have setup IIS so that .cgi scripts are sent to the python
> interpreter. I am trying to access a SQL Server database, 'pubs' (using
> Python DB APIs). I am importing odbc, dbi in my script. Database access
> works fine when I run a dbtest script at the command line, but the same
bit
> of code gives an error when run in the .cgi script. It is not finding the
> data source name (DSN). I have set up the DSN in the ODBC Driver Manager
> correctly.
>
> I can't see what environment parameters to set to get the .cgi python
script
> to access the 'pubs' database. Can anyone help ?
>
>
> Thanks
>
>
>




From a.eyre at optichrome.com  Thu Jan 20 05:00:47 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Thu, 20 Jan 2000 10:00:47 -0000
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <865rur$5ud$1@nnrp1.deja.com>
Message-ID: <001a01bf632d$393fbda0$3acbd9c2@peridot.optichrome.com>

> I like the following syntax for readability, and feel the this should be
> implimented in Python:
> 
> 	sub	ax,bx		# Lets keep the python comment symbol
> 	jne	Mylabel         # We also need labels
>         rtn			# Notice all commands are three chars
> Mylabel
> 	add	00h,ax
> 

I feel this may be too confusing. I'd prefer a syntax like this:

1000101001011101010010101001010100010010010001000100100101000101
1011000001100100100010010010001010011101001001010100100101010010

etc.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From s.schwarzer at ndh.net  Sun Jan  2 18:21:55 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Mon, 03 Jan 2000 00:21:55 +0100
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl> <386E7BD4.514764EA@ndh.net> <84nv0u$bu9$1@nntp9.atl.mindspring.net>
Message-ID: <386FDD93.DBF3A4C1@ndh.net>

Hello Aahz,

Aahz Maruch schrieb:
> Anyone who claims that tuples are unnecessary has never used a tuple as
> a dict key.  That requirement necessitates differentiating between a
> mutable and immutable list -- might as well stick with tuples.

I said, I never needed this immutability. I didn't say they are
never necessary :) . Nevertheless you are right; I never used
tuples as dictionary keys - yet ;-) .

Thank you for the clarification; I forgot this application.

Stefan


From wtanksle at hawking.armored.net  Wed Jan 19 23:58:00 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 20 Jan 2000 04:58:00 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net> <8631bt$1nu@news.or.intel.com>
Message-ID: <slrn88d63f.13f.wtanksle@hawking.armored.net>

On Tue, 18 Jan 2000 16:41:06 -0800, tye4 wrote:

>So you don't think this is a problem?? Hello!, read all the messages in this
>thread. Seems to me a lot of programmers have a problem with this.

ONE person aside from yourself.  Go ahead, count.

>Thanks to this quirk, I now have to worry how an editor saves tabs to files.

No, you don't.  Calm down and think for once.

>I have offered you many reasons why we need 'end' block support..

No, you haven't.  All you've done is claim that other languages don't use
indentation.  So what?  Python uses assignment.  ML doesn't.  Does that
mean ML is bad and Python is good, or is it the other way around?

>You have not offered a single reason why this is not a problem..

Yes, we have.  For example, we've offered the fact that it doesn't cause
any manner of problems in our collective, extensive experience which
wouldn't be caused by any other choice just as well.

Add to this the fact that the indentation syntax -- please read this -- is
one of the main reasons why we LIKE the language.  It's one of our
favorite features.  Take it away, and Python would be just another
scripting language (although with other good choices elsewhere).  Why, if
Python didn't use indents, there might be more than one way to do it!
That would be terrible.

>-tye4

-- 
-William "Billy" Tanksley, in hoc signo hack


From ivanlan at callware.com  Tue Jan  4 17:22:25 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 04 Jan 2000 15:22:25 -0700
Subject: Python & Apache
References: <A4uc4.411$hJ1.3104@news.rdc1.tx.home.com>
Message-ID: <387272A1.F036408C@callware.com>

Hi All, Paul--

"Paul E. Bible" wrote:
> 
> Has anyone had any success getting Python CGI scripts to run under the
> following configuration?
> 
> OS:                Windows/NT
> Web Server:   Apache 1.3.9
> Python:           v1.5.2
> 

Sure.  That's exactly what my webserver uses.  All the tools are written
in Python:

	http://www.pauahtun.org/tools.html

I think there's a few leftover dregs of perl on the system, but I
*never* use it.

The biggest gotcha is that the Python scripts *must* start with

	#!c:\Python\python.exe

(or wherever Python actually lives). You can't use 

	#!/usr/local/bin/python

because Apache won't find it, even if the path exists on Apache's
drive--it insists on the driveletter and the backslashes.  I think, but
am not certain, that it pukes on spaces in that path, too.

I made no other changes to accomodate NT.  I'm using NT4 with Service
Pack 5.  I've heard bad things about SP6.

Anything else?

<too-wrapped-up-in-base-20-to-be-funny>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From tim_one at email.msn.com  Tue Jan 18 23:28:32 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 18 Jan 2000 23:28:32 -0500
Subject: More problems....
In-Reply-To: <2E7h4.230$Pe6.2431@newsfeed.slurp.net>
Message-ID: <000201bf6235$a53d25e0$9a2d153f@tim>

[posted & mailed]
[Guido, is this a known bug?  Not in the bugs list, but it's
 become a frequent complaint.]

[TechSpood]
> ...
> I will write a simple program (in the IDLE enviroment) then
> when I run it, it will run it til it gets to the first input
> statement, at which time it will stop reading the program.
> I don't get a syntax error or any other kind of error.
> ...

Let's track this down.  For starters, I'm assuming you also *run* (as well
as write) the program within IDLE.  Scream if that's wrong.

You haven't said which version of Python you're using (I assume 1.5.2),
which OS (since your post identified itself as coming from Outlook Express,
I assume some flavor of Windows), or given a minimal failing example.  So
there's little to go on.  I don't have any problem running this code under
Win95 under IDLE:

for i in range(3):
    pw = raw_input(("Please enter your password: ",
                    "Try again: ",
                    "Last chance, pal: ")[i])
    print "mumble, mumble, ... hey!", pw, "is no good!"
print "Go away."

Do you?  Here's a sample session pasted from my IDLE:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12)
[MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
IDLE 0.5 -- press F1 for help
>>>
Please enter your password: xyz
mumble, mumble, ... hey! xyz is no good!
Try again: try what again?
mumble, mumble, ... hey! try what again? is no good!
Last chance, pal: same to you!
mumble, mumble, ... hey! same to you! is no good!
Go away.

Assuming that doesn't work for you, there's one obvious (to me <wink>)
difference in what I'm doing:  I'm running the latest version of IDLE, not
the one that shipped with 1.5.2.

If you want to try that, get it from the CVS archive at python.org, or, if
you're not set up for CVS, David Ascher kindly maintains a current snapshot
at

http://starship.python.net:9673/crew/da/

info-is-more-useful-than-frustration<wink>-ly y'rs  - tim





From g_will at cyberus.ca  Mon Jan 24 08:10:35 2000
From: g_will at cyberus.ca (Gordon Williams)
Date: Mon, 24 Jan 2000 13:10:35 GMT
Subject: Tkinter and wxPython
References: <388B060B.D13F7B64@exceptionalminds.com>
Message-ID: <01bf666c$1cbb47c0$4c42c3d1@Cyberus.cyberus.ca>


Timothy Grant <tjg at avalongroup.net> wrote in article
<388B060B.D13F7B64 at exceptionalminds.com>...
> OK, After just spending three weeks learning Tkinter on-the-fly, I have
> to ask this question, what advantages/disadvantages are there to each
> Tkinter and wxPython?
> 

My reason for moving to wxPython from Tkinter is due to the slow speed in
drawing lines on canvas. wx was faster on the order of 200 to 1800 times
for the 1000 to 10000 lines that I plotted.  These benchmark times were
posted on the list earier if you are interested.

I found that Tkinter was easier to learn and use primarily due to the
package small size.  wxPython has a *lot* of features many of which are
redundant.  There are two, three and sometimes four ways of doing the same
thing.  This does not add to the clarity and adds a lot to my memory
overload.  wxPython is much more powerful than Tk from what I have seen of
both.

Robin has been very good at answering wx questions on the wx mailing list. 
This is a *big* plus, but I dont know how he is going to keep up with the
volume.  The number of people that are now using wx is growing
exponentially.

My experience is only with MSW, but loading the package was very simple:
just download and install in your python directory.

Good luck,

Gordon Williams


From mal at lemburg.com  Sun Jan 23 08:58:05 2000
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sun, 23 Jan 2000 14:58:05 +0100
Subject: ANN: platform.py - Version 0.7.0
References: <3889CFEE.E2C0EF1E@lemburg.com> <3889F0EC.6E8BDA4B@lemburg.com>
Message-ID: <388B08ED.638F1949@lemburg.com>

[Sorry for posting this twice: the version I posted yesterday
 had a typo in the URL]

ANNOUNCING:

                  platform.py - Version 0.7.0

        A Python module providing platform information APIs.


WHAT IT IS:

platform.py was intially developed for the mxCGIPython project
to automate the creation of file names including platform
specific data for users to be able to recognize and download
the appropriate Python interpreter binary for their (or their
ISPs) setup.

Since then the module has been extended and redesigned to
provide a generic interface for querying platform information
both in human readable form and for other application purposes
through platform dependent special APIs.

platform.py supports (at least) these setups: most Windows
setups, MacOS, various Linux distributions, various Unices,
OpenVMS, Java VM and DOS.


WHAT'S NEW ?

Version 0.7.0 is the first new version after the intial release
and update cycles last year.


WHERE CAN I GET IT ?

Documentation is provided through doc-strings and many
comments in the source code, which can be found via my
Python Pages:

        http://www.lemburg.com/python/


WHAT DOES IT COST ?

It comes with a Python-style license, but is otherwise free for
commercial and non-commercial use.


WHERE CAN I GET SUPPORT ?

Commercial support for all my packages, tools and other
Python software will soon be available through my newly
founded software company. Keep an eye open on this list
for an announcement.


REFERENCE:

<P><A HREF="http://www.lemburg.com/python/">
platform.py 0.7.0</A> - Generic platform information APIs.
(22-Jan-2000)

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





From bob at no.com  Sun Jan 16 15:05:04 2000
From: bob at no.com (bob)
Date: Sun, 16 Jan 2000 20:05:04 GMT
Subject: SERIAL I/O ON WIN32
Message-ID: <Qvpg4.581$253.12322@news>

Hi there...newbie to Python here.

I need to pass information across a COM port. I have the serial-i/o module
but no clue on how to use it.

Thanks for any and all help!

Bob




From jcollins at pacificnet.net  Thu Jan 27 14:42:33 2000
From: jcollins at pacificnet.net (Jeffery D. Collins)
Date: Thu, 27 Jan 2000 11:42:33 -0800
Subject: Installing quirk...
References: <389007A5.6B22105C@emw.ericsson.se>
Message-ID: <38909FA9.374A3DE@pacificnet.net>

You need to have the gnu readline library installed on your machine.
See the readline entry Modules/Setup for further details.

Jeff


Per Soderstam wrote:

> Hi!
>
> I'm trying to build the Python1.5.2 release on a SPARC5 workstation
> running Solaris5 and i want another functionality for my pyhon promt
> (tsch type, with the up/down/left/right arrow positioning and command
> history as well as backspace instead of delete). Setting the SHELL
> (../tcsh) variable does not seem to work, but I'm no guru :). Do anyone
> understand my question? If so, do anyone have an answer?
>
> /Per
> --
> Per S?derstam, Ericsson Microwave Systems AB, SE-431 84 Molndal, SWEDEN,
> tele: (+46) (0)31 7470195, e-mail: per.soderstam at emw.ericsson.se
> ********************************************************************
> The views expressed above is my own and does not nessecarily
> reflect the policy of my employer. http://www.ericsson.se/microwave/



From hei at adtranzsig.de  Tue Jan 25 07:43:01 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 25 Jan 2000 13:43:01 +0100
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be>
Message-ID: <948807929.472741@news.adtranzsig.de>

Thomas Hamelryck schrieb in Nachricht <86joac$l2s$1 at mach.vub.ac.be>...
>Good analogy, actually. Sex is great, but should never be forced
>on you.

You misunderstood the "whitespace sucks" topic.

>
>BTW, people here seem to have adopted the idea that if you don't like
>a particular feature of a language, you should abandon it (e.g., "You
>don't like indentation? Use Perl!"). But many languages have evolved
>exactly this way: users criticize, complain, praize and whine, thereby
>influencing the shape of the languge in question. You can find a very
>nice description of this process with respect to C++ in "The Evolution
>of C++ : Language Design in the Marketplace of Ideas", by Jim Waldo
>(Editor), James Waldo (Editor).

Not all languages evolve like C++. Whether this is good or bad,
is in the eye of the beholder. Again, have a try to convince the C++
people to use "begin" and "end". If C++ is so evolvable, it should be
easy going, shouldn't it? Much easier than arguing against those
boneheaded Python people. C++ needs fresh ideas all the time.
(At least, that's what it looks like nowadays to me)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From ivanlan at callware.com  Wed Jan 19 17:41:55 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 19 Jan 2000 15:41:55 -0700
Subject: olemsg32 module used in win32com exchange server example?
References: <slrn8897gs.3cm.grant@grante.comtrol.com> <864ult$fot$1@nnrp1.deja.com> <slrn88ceoc.5ga.grant@grante.comtrol.com>
Message-ID: <38863DB3.7FAEC820@callware.com>

Hi All--

Grant Edwards wrote:
> 
> >>  2) Does the win32com stuff talk directly to the Exchange
> >>     server, or does it go through a local copy of Outlook?
> >
> >It goes through the MAPI DLLs - the same DLLs that the local copy of
> >Outlook itself uses.
> 
> Cool.  Does that mean that I don't even need to have Outlook
> installed in order for a Python/COM/MAPI program to work?
> 

I suspect that is true, but there's a gotcha.  Unless I'm wrong, both
NetScape Mail and Outlook use MAPI dll's, but they expect different
functions to be found there.  So, when I want to use Outlook to retrieve
mail, I can configure everything else correctly but I have to change the
MAPI.DLL to the one the Outlook expects or it barfs.  Then put it back
when I want to use NetScape. ...

<life-ain't-easy-for-the-dll-impaired>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From mjackson at wc.eso.mc.xerox.com  Wed Jan 19 12:42:22 2000
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 19 Jan 2000 17:42:22 GMT
Subject: def args idiom and signatures
References: <02b001bf629c$d7ce3960$01ffffc0@worldnet.att.net>
Message-ID: <864t1u$4mr$3@news.wrc.xerox.com>

"Emile van Sebille" <emile at fenx.com> writes:
> In /F's Standard Python Library I spotted:
> 
>     def _d(y, m, d, days=(0, 31, 59, ..., 334, 365)):
> 
> where the balance of the function relies on days having this value.
> 
> Trying to understand why it may be an argument vs defined within
> the function, I suspect that as the def is only processed to bytecode
> once, whereas an in-line assignment would be re-processed with
> each execution, that the decision is performance based.
> 
> Now, in light of the various discussions pertaining to function 
> signatures, it seems that this idiom presents a particular problem.
> 
> Various munging of the name or introduction of a static type keyword
> may help address this, but I'm curious if this common usage and
> what people's thoughts are.
> 
> Or-even-if-my-understading-of-it-is-correct?-ly y'rs

Um. . .flexible leap year handling?

Sweden-once-had-a-February-30th-ly y'rs,

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
    Faced with the choice between changing one's mind and proving
    there is no need to do so, almost everybody gets busy on the proof.
				- John Kenneth Galbraith




From arcege at shore.net  Tue Jan 11 19:50:34 2000
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 12 Jan 2000 00:50:34 GMT
Subject: Please, someone tell me what's going wrong here. ;>
References: <GzLe4.54805$Ke.42098@news21.bellglobal.com>
Message-ID: <udQe4.3161$Ef6.805744@news.shore.net>

Jason Maskell <backov at nospam.csolve.net> wrote:
: I posted a while back on embedding, and after scouring deja and reading the
: demos, I can't even get this simple example to work. It's essentially a C++
: Builder'ized version of the simple embedding demo.. PyRun_SimpleString
: returns -1 and foo is never executed.

: TForm1 *Form1;

: __fastcall TForm1::TForm1(TComponent* Owner)
:     : TForm(Owner)
: {
: }

: void initxyzzy(); /* Forward */

: /* A static module */

: static PyObject * xyzzy_foo(PyObject *self, PyObject *args)
: {
:     Form1->Memo1->Lines->Add("Blah");

:  return PyInt_FromLong(42L);
: }

: static PyMethodDef xyzzy_methods[] = {
:  {"foo",  xyzzy_foo, METH_VARARGS},
:  {NULL,  NULL}  /* sentinel */
: };

: void initxyzzy()
: {
:  PyImport_AddModule("xyzzy");
:  Py_InitModule("xyzzy", xyzzy_methods);
: }

: void __fastcall TForm1::Button1Click(TObject *Sender)
: {
:     int a;
:     Py_SetProgramName(Application->ExeName.c_str());
:     Py_Initialize();
:     initxyzzy();
:     a=PyRun_SimpleString("xyzzy.foo()");
:     Py_Finalize();

: }

: void __fastcall TForm1::FormDestroy(TObject *Sender)
: {
:     Py_Finalize();
: }


I would try to stay away from calling the module initializers, they
are designed with a purpose in mind.  Go for the documented route:

void initxyzzy()
  {
    (void)Py_InitModule("xyzzy", xyzzy_methods);
  }

void TForm1::Button1Click(TObject *Sender)
{
    int a;
    Py_SetProgramName(Application->ExeName.c_str());
    Py_Initialize();
    a=PyRun_SimpleString("import xyzzy"); /* import into the namespace */
    a=PyRun_SimpleString("xyzzy.foo()");
    Py_Finalize();
}

The module never got imported into the __main__ module, only
initialized.

  -Arcege



From fredrik at pythonware.com  Tue Jan 11 10:28:26 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: 11 Jan 2000 09:28:26 -0600
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Jan 11)
Message-ID: <AE394FE7A7985025.628D930F611635E3.89C1ECAB49EC72F3@lp.airnews.net>

Just a few quickies, this time:

    Going to the Software Development 2000 conference in March? 
    Don't miss the Python path:
        http://www.sdexpo.com/2000/west/conference.html
        http://www.sdexpo.com/cgi-bin/main.pl?x-a=q&x-tb=q&x-id=35631

    And yes, you can still register for the Python conference:
        http://www.python.org/workshops/2000-01/


    Stuart Zakon discusses Python and UML in a very interesting
    article:
        http://www.deja.com/getdoc.xp?AN=571129415
        http://www.objectsbydesign.com/projects/python_uml.html


    What happens if Python is modified to let you return arbitrary
    objects from the __init__ method?
        http://www.deja.com/viewthread.xp?AN=569166793


    Neel Krishnaswami has implemented multimethods in Python (that
    is, methods that change their behaviour based on all arguments,
    rather than just 'self', I think):
        http://www.deja.com/getdoc.xp?AN=569792454
        http://www.sff.net/people/neelk/open-source/

    Gordon McMillan's Installer tool now available for Linux/Unix:
        http://www.deja.com/getdoc.xp?AN=570043368
        http://starship.python.net/crew/gmcm/distribute.html


    Greenpeace International is looking for a Python-capable system
    administrator for their Amsterdam office:
        http://www.deja.com/getdoc.xp?AN=569058379


========================================================================

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

      The Vaults of Parnassus ambitiously collects Python resources
          http://www.vex.net/~x/parnassus/   

      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 <claird at neosoft.com> to subscribe.  Mention "Python-URL!".


-- The Python-URL! Team--

Dr. Dobb's Journal (http://www.ddj.com) is pleased to participate in and
sponsor the "Python-URL!" project.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From skip at mojam.com  Tue Jan  4 23:33:35 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 4 Jan 2000 22:33:35 -0600 (CST)
Subject: Importing a dicrionary from a file
In-Reply-To: <3872C4EC.20ADE046@trojanslair.zzn.com>
References: <3872C4EC.20ADE046@trojanslair.zzn.com>
Message-ID: <14450.51615.249959.909830@beluga.mojam.com>

    Keavan> I am a newbie, and i am working on a program that uses a really
    Keavan> big dictionary. i was wondering if i could store the dictionary
    Keavan> in a seperate file, let's say dict.dat.

Sure.  Use a dbm-like object.  Something like

    import anydbm
    db = anydbm.open("dict.dat", "c")	# "c" says "make writable & create 
					# if necessary"
    db["foo"] = "bar"
    ...
    db.close()

should work.  If your definition of big corresponds roughly to mine, you
almost certainly don't want to read the entire thing into memory at once.

Your only restriction is that your keys and values are limited to strings
unless you jump up a level and use shelve.  The lib doc for the shelve
module has a simple example:

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

Cheers,

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 dworkin at ccs.neu.edu  Tue Jan  4 11:05:54 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 04 Jan 2000 11:05:54 -0500
Subject: Quitting my program
In-Reply-To: volumeregeling@hotmail.com's message of "Tue, 04 Jan 2000 15:45:19 GMT"
References: <3872148f.18589850@122.0.0.250>
Message-ID: <vndogb1u8il.fsf@betelgeuse.ccs.neu.edu>

volumeregeling at hotmail.com (volumeregeling) writes:

> I have a problem ,I don?t know how to stop my program from a function.

Either:

import sys
sys.exit()

or:

raise SystemExit

will cause Python to silently exit.

-Justin

 




From JoeSmith at bogusaddress.com  Wed Jan 19 05:02:46 2000
From: JoeSmith at bogusaddress.com (Joe Smith)
Date: Wed, 19 Jan 2000 10:02:46 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <87ya9muahl.fsf@fun77.kivikko.hoas.net>
Message-ID: <38858C94.4135FAFD@bogusaddress.com>

It looks like "Nick>" is indented 4 spaces.  What does this mean?  Is somebody
or somebody's editor not sane?  (I am joking.)

Erno Kuusela wrote:

> >>>>> "Nick" == Nick Belshaw <nickb at earth.ox.ac.uk> writes:
>
>     Nick> ! - whats wrong with tabs?
>
> Nothing at all, since no sane editor would ever use a tab
> width other than 8, an no sane person would ever want anything
> else than 8-character indents.
>
>   -- dr. 8



From gregm at iname.com  Tue Jan 25 06:31:25 2000
From: gregm at iname.com (Greg McFarlane)
Date: Tue, 25 Jan 2000 22:31:25 +1100
Subject: Tab notebook control available for Tkinter?
In-Reply-To: <388D24E8.EB1318B8@austin.rr.com>; from Daniel Weber on 25 Jan 2000 at 03:19:59AM
References: <388D24E8.EB1318B8@austin.rr.com>
Message-ID: <20000125223125.15917@nms.otc.telstra.com.au>

Python megawidgets (Pmw) is based on Tkinter and has a tabbed
notebook.  The home page is:

    http://www.dscpl.com.au/pmw/

On 25 Jan, Daniel Weber wrote:
> I'm trying to write a new front-end to a product (written in OS/2) to
> make it cross-platform.  However, I need a notebook tab control, which
> I haven't been able to find in the archives or any reference in
> deja.com.
> 
> Other than moving to wxWindows (which isn't available on OS/2 that I can
> see), are there any Tkinter based solutions?
> 
> TIA -- daniel
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com



From bsb at winnegan.de  Mon Jan 10 15:37:27 2000
From: bsb at winnegan.de (Siggy Brentrup)
Date: 10 Jan 2000 21:37:27 +0100
Subject: More Tkinter trouble
In-Reply-To: Timothy Grant's message of "Mon, 10 Jan 2000 03:20:15 -0800"
References: <3879C06F.FCCD7978@exceptionalminds.com>
Message-ID: <873ds5mznc.fsf@baal.winnegan.de>

Timothy Grant <tjg at avalongroup.net> writes:

[...]

> # Everything works great if I terminate the function here
>         
>         self.SupplierE = Entry(SupplierFrame)

change this to 
         self.SupplierE = Entry(SupplierFrame.interior())

and have a look at Pmw examples.

HIH
  Siggy

-- 
Siggy Brentrup - bsb at winnegan.de - http://www.winnegan.de/
                  bsb at north.de - http://www.north.de/~bsb/
****** ceterum censeo javascriptum esse restrictam *******



From news at dorb.com  Mon Jan 17 21:10:42 2000
From: news at dorb.com (Darrell)
Date: Mon, 17 Jan 2000 21:10:42 -0500
Subject: Trouble with Python and ASP
References: <860gpo$882$1@nnrp1.deja.com>
Message-ID: <XVPg4.6141$U5.236616@typhoon.nyroc.rr.com>

I'm no expert here.

Since you know it happens when two users hit the page at once, can you
simplify that page ? Get it down to the minimum required to cause the
problem.

--
--Darrell
<larsde at telepost.com> wrote:
> I'm experiencing disturbing problems with using Python as an ASP
> scripting engine. I've been working on a project for quite some time,
> and so far everything has been peachy.
>





From jbauer at rubic.com  Sun Jan  9 10:08:23 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Sun, 09 Jan 2000 09:08:23 -0600
Subject: urllib.urlopen returns 'connection refused' on Solaris
References: <3875CBCF.72E26CE7@ina.fr> <38789474.6887@cs.bham.ac.uk>
Message-ID: <3878A467.1B795100@rubic.com>

Stuart Reynolds wrote:
> Richard Gruet wrote:
> > On Solaris 2.6 (Sun OS 5.6), the fct urllib.urlopen() returns
> > an error 'Connection refused' for all URLs:
> 
> I had the same problem. Locally, we have some network security
> features installed that prevent direct access to http and ftp 
> address by any program. We can run Netscape as it uses a 
> proxy server.

If a proxy server is causing your connection problem, you
should be aware that urllib checks the environment for names 
that end in _proxy  (i.e. 'http_proxy').  You can assign
a value to the environment or as part of URLopener
instantiation.  A quick read of the urllib source will
explain all.

Proxies for the ftp protocol aren't as consistent
as http, IIRC.  The technique varies with the kind
of proxy used.  Since there are a limited number
of techniques, it might prove useful to write an
ftpprobe script to pursue known ftp proxy techniques
and report the successful strategy back to the user.

Or is this becoming increasing irrelevant as
security administrators shut down unencrypted 
ftp ports?

Jeff Bauer
Rubicon Research



From dnichols at sourcesignal.com  Fri Jan 14 15:02:45 2000
From: dnichols at sourcesignal.com (David Nichols)
Date: Fri, 14 Jan 2000 12:02:45 -0800
Subject: Tcl/Tk Path problems in Python 1.5.2 installation
Message-ID: <387F80E5.A0AF82A5@sourcesignal.com>

I just installed python 1.5.2 (running py152.exe and answering 'yes - I
want to install Tcl/Tk') and have been unable to get the path set
up correctly for graphical applications.  For instance, if I go to the
idle directory (//E/Programs/Py152/Tools/idle) and try to launch it I
get the following result.

============================================================
BASH.EXE-2.02$ python idle.py
Traceback (innermost last):
  File "idle.py", line 3, in ?
    PyShell.main()
  File "E:\Programs\Py152\Tools\idle\PyShell.py", line 611, in main
    root = Tk()
  File "E:\Programs\Py152\Lib\lib-tk\Tkinter.py", line 886, in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
    {} ./lib/tcl8.0 . E:/Programs/Py152/tcl8.0/library
E:/Programs/Py152/Tools/library

This probably means that Tcl wasn't installed properly.
============================================================

The directories in sys.path are as follows ...
E:\PROGRAMS\PY152
E:\PROGRAMS\PY152\DLLs
E:\PROGRAMS\PY152\PIL
E:\PROGRAMS\PY152\lib
E:\PROGRAMS\PY152\lib\lib-tk
E:\PROGRAMS\PY152\lib\plat-win
E:\Programs\Py152\DLLs
E:\Programs\Py152\Lib
E:\Programs\Py152\Lib\lib-tk
E:\Programs\Py152\Lib\plat-win

Tcl in installed in E:\PROGRAMS\Py152\\Tcl, and init.tcl exists in
E:\PROGRAMS\Py152\Tcl\lib\tcl8.0 ... but this directory does not
seem to be searched.

Do I need to set additional environmental variables or make changes
to the registry beyond those made by the installer?

Thanks in advance for any help

David Nichols
Source Signal Imaging
dnichols at sourcesignal.com



From pbible at home.com  Sun Jan 16 19:11:05 2000
From: pbible at home.com (Paul E. Bible)
Date: Mon, 17 Jan 2000 00:11:05 GMT
Subject: If you can only buy *one* Python book...
References: <N%rg4.4360$3b6.20735@ozemail.com.au> <20000116182502.A1106@better.net>
Message-ID: <t6tg4.305$iL4.4241@news.rdc1.tx.home.com>

I found 'Learning Python' to be very easy to read and a great book to get
started programming in Python...
The 'Python Essential Reference' is a good addition to the online
documentation as well.  I found 'Programming Python'
somewhat difficult to follow myself.

Paul

"William Park" <parkw at better.net> wrote in message
news:20000116182502.A1106 at better.net...
> On Sun, Jan 16, 2000 at 10:55:41PM +0000, Jason Stokes wrote:
> > Which one should you get?  I quite like the look of the New Riders
> > Python Reference, but is it comprehensive enough?  I already know Perl,
> > and I actually prefer learning from reference works.  Suggestions?
> >
> > --
> > Jason Stokes: jstok at bluedog.apana.org.au
> > --
> > http://www.python.org/mailman/listinfo/python-list
>
> Read the online 'tutorial'.  After that you have many choices:
>     - Learning Python
>     - Quick Python Book
>     - etc.
> I have printout of all online docs, except 'library'.  But, I still
> prefer and recommend a bound book.  I refer to 'language' and 'library'
> references only when I have syntax or library problems.
>
> William Park
>




From kc5tja at garnet.armored.net  Fri Jan 14 16:56:31 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 14 Jan 2000 21:56:31 GMT
Subject: XML-RPC Server
References: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de><slrn87ne3i.qla.kc5tja@garnet.armored.net> <xqxr9fkk94u.fsf@skewer.ca.boeing.com> <008701bf5ea8$d51d7f20$f29b12c2@secret.pythonware.com>
Message-ID: <slrn87v78u.gt3.kc5tja@garnet.armored.net>

In article <008701bf5ea8$d51d7f20$f29b12c2 at secret.pythonware.com>, Fredrik
Lundh wrote:
>- the goal is to have full implementations of both XML-RPC
>and SOAP available in a not too distant future. (and with a
>little luck, we might even be able to persuade GvR to ship
>them with Python 2.0 or so...)

I cast my vote for this:  include xmlrpclib and soaplib!!  :)

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From sei at sunse.jinr.ru  Wed Jan 19 16:20:12 2000
From: sei at sunse.jinr.ru (Evguenii Smogailov)
Date: Thu, 20 Jan 2000 00:20:12 +0300 (MSK)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <388670D2.2FA842F6@starmedia.net>
Message-ID: <Pine.SOL.3.91.1000120001532.29235B-100000@sunse>

On Wed, 19 Jan 2000, Andrew Csillag wrote:
> Just because I know lisp doesn't mean I can't learn python.  I don't
> find it a defect that python doesn't support prefix notation like lisp
> does (wouldn't it be awful if it did?  Yeesh...).  Python doesn't have
> to be Perl for Perl programmers to be able to learn it.  I bet there are
> a bunch of Perl programmers that even know Java.

I have made no such statements (like someone can't learn something), it 
is your extrapolations:)

> You're saying that Perl is readable? HAHAHAHA... Oh, you were serious...
> Ok, 1 eq 2 is easier to read that 1==2?  Hmmmm... I think not.
> 

Yes, i'm saying that 'eq', 'ne' are more readable then '==' '!=', '<>',
'next' shorter then 'continue' and '..' operator can be a good thing to 
consider.

evguenii




From shevitz at my-deja.com  Wed Jan 19 13:52:47 2000
From: shevitz at my-deja.com (shevitz at my-deja.com)
Date: Wed, 19 Jan 2000 18:52:47 GMT
Subject: python mode generating lots of files when C-c C-c...
Message-ID: <86515p$hme$1@nnrp1.deja.com>

I'm a newbie to XEmacs, python mode, and wxPython. I'm running on NT.
I am trying to learn wxPython so I have a very simple script which puts
up a frame window. I run it a lot to see what single new lines of code
do.

If I execute the script from the buffer directly (via C-c C-c) it seems
to work fine. And when I close the window (via it's close window box),
everything still seems to work fine. The problem is that it would
appear every time I do this, a temp file is created in the same
directory as the script, and it appears to have the same contents as
the script. The files have names like "python-lDA65O". And as I just
discovered the problem I have about 50 of them in the directory.

I don't know what's going on here, but I would sure appreciate if
someone can tell me how to stop it. Am I leaving processing dangling?
Is there a flag I can set in XEmacs, or some line of code I can put in
the script to prevent this.

thanks,
Danny


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


From P.J.W.S.VRIJLANDT at INT.azg.nl  Thu Jan 27 05:29:45 2000
From: P.J.W.S.VRIJLANDT at INT.azg.nl (P.J.W.S. Vrijlandt)
Date: Thu, 27 Jan 2000 11:29:45 +0100
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
In-Reply-To: <slrn8903r7.al2.thantos@brimstone.mecha>
Message-ID: <1392272911@nds-3.azg.nl>

> From:          thantos at chancel.org (Alexander Williams)
> Subject:       Re: Which idiom is better? try: ... except Attr..Err or hasattr?
> Date:          Thu, 27 Jan 2000 09:27:03 GMT
> Organization:  Wyvern Enterprises/Palantir Heavy Industries
> Reply-to:      thantos at gw.total-web.net
>>  Which coding idiom is better?  The second form is one line 
>>  shorter
> 
> Personally, I prefer the former idiom with try/except; some object
> because its slightly slower than the other, but I find it clearer.
> 

According to the manual,  hasattr is implemented by calling getattr 
and seeing if an exception occurs, so hasattr will not be faster than 
try...except...

Patrick



From just at letterror.com  Sun Jan 23 15:23:16 2000
From: just at letterror.com (Just van Rossum)
Date: Sun, 23 Jan 2000 21:23:16 +0100
Subject: Win32Installer + Numeric, anyone succeeded?
In-Reply-To: <CC59868A0A76D311B1F50090277B321C1F3097@VRTMAIL>
Message-ID: <l03102803b4b1135d1811@[193.78.237.179]>

At 8:01 AM -0500 23-01-2000, Mike  Fletcher wrote:
>Yes, that was it (loads and runs now).  No, there doesn't appear to be an
>import _numpy in my numeric directory anywhere.  Could be it's only in the
>CVS repository?

I just checked, it's indeed in the CVS repository, but I don't know since
when. It's in my version since release 11 (which I still use...), but it's
possible I patched it myself back then.

Just





From jmarshall at alum.mit.edu  Tue Jan 25 09:34:19 2000
From: jmarshall at alum.mit.edu (Joe Marshall)
Date: Tue, 25 Jan 2000 09:34:19 -0500
Subject: What about REBOL (www.rebol.com)
References: <000e01bf66cb$670a8480$a42d153f@tim>
Message-ID: <Mvij4.49146$Rj5.31961@dfw-read.news.verio.net>

Tim Peters <tim_one at email.msn.com> wrote in message
news:000e01bf66cb$670a8480$a42d153f at tim...
> [posted & mailed]
>
> [Tim]
> > ... (despite that it *looks* nothing like Lisp, a REBOL
> > "block" is a lexical closure, and the semantics are
> > virtually identical to Scheme's)
>
> [Joe Marshall]
> > REBOL abandoned lexical closures in version 2.0, so the
> > semantics have become quite different from Scheme's.
>
> I eagerly yield to Joe on all matters REBOL -- he was the primary
> implementer of the version I used most.  So, Joe, should the original
> questioner learn Python or REBOL?

Python, of course!

(Who in their right mind would advocate language `X' in comp.lang.`Y'?)

But seriously, it would depend a lot on what the original poster wanted
to accomplish.  He said he was interested in learning a `multi-platform
script language'.  Most likely he wants to improve his marketable skills.
I'd say Perl, TCL, and Python provide the greatest opportunity for that.
REBOL is too immature a language, is missing necessary connectivity
(the core product does not support 3rd party dynamic libraries, foreign
functions, COM or CORBA, or even the ability to spawn subprocesses
and capture their output), and has a miniscule user base compared to
the others.

So  I think Python is his best bet.  Unless, of course, he finds that
Perl appeals to his sense of aesthetics.....

~jrm




From just at letterror.com  Mon Jan 31 13:21:30 2000
From: just at letterror.com (Just van Rossum)
Date: Mon, 31 Jan 2000 19:21:30 +0100
Subject: where is bbpy?
In-Reply-To: <3895d03e@dnews.reed.edu>
Message-ID: <l03102808b4bb82e1d179@[193.78.237.139]>

At 10:11 AM -0800 31-01-2000, Noah Iliinsky wrote:
>Howdy.
>
>The Python FAQ section 7.10 ( http://www.python.org/doc/FAQ.html#7.10 )
>claims that there is a Python plug-in for BBedit in the Mac distribution
>called BBPy. However, there is no sign of it in the distribution or
>anywhere outside of the FAQ. Where has it gone? Has anyone seen it?

It has deceased. It is no more. It has long since been replaced by a
"proper" Mac IDE.

(In case you'd like to revive it, I could dig out the source.)

Just





From quinn at pfennig.ugcs.caltech.edu  Fri Jan 21 03:48:06 2000
From: quinn at pfennig.ugcs.caltech.edu (Quinn Dunkan)
Date: 21 Jan 2000 08:48:06 GMT
Subject: Whitespace delimiters suck
References: <000001bf63a1$832e5740$6ea0143f@tim> <868bbv$lug@news.or.intel.com> <roy-11C630.23024320012000@netnews.nyu.edu>
Message-ID: <slrn88g7ct.n42.quinn@pfennig.ugcs.caltech.edu>

On Thu, 20 Jan 2000 23:02:43 -0500, Roy Smith <roy at popmail.med.nyu.edu> wrote:
>When I was first learning programming (a quarter of a century ago!), 
>there was the idea of the "telephone test", i.e. your code should be 
>simple enough that if you read it to somebody on the phone they would 
>understand it.  Python not only fails the phone test, it fails the 
>printed page test; I can print two programs on a piece of paper, which 
>differ only in tabs and spaces, and they will appear identical yet which 
>do different things.  Barf.
>
>I'm in love with python, but this idea of white space as significant 
>syntax is the biggest wart on the language by far.

Hey, you're right!  All those times I had to write programs by telling someone
what to write over the phone would have been so much easier if I could have
said "if x equals ten colon y equals three plus x endif" instead of "if x
equals ten colon y equals three plus x dedent".  No, wait, maybe it wouldn't
have.

Let's try again.

All this time I've been writing python by hand on paper, and scanning it in
with an ocr program.. no, uh...  I usually mix tabs and spaces randomly, and
then check programs for syntax errors by printing them out and staring at
them.  It's really hard to find some errors because python is
whitespace-significant.  Why can't some clever person design a syntax-checking
program to do this automatically?  In the process, it could even reduce it to
some unambiguous, machine-friendly level.

Fortunately, python passes the telegraph test:
IF X EQUALS TEN COLON Y EQUALS THREE DEDENT STOP

But it fails the upside-down test and the purple-trashcan test...

As far as simplicity goes that's programmer-dependent.


From niels at endea.demon.nl  Wed Jan 12 14:32:58 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Wed, 12 Jan 2000 20:32:58 +0100
Subject: Compile snmpy problem
References: <85ht8l$o4j1@imsp212.netvigator.com>
Message-ID: <387CD6EA.3DD73A24@endea.demon.nl>


"DORAEMON.HK" schreef:
> I try to compile SNMPy in a RedHat 6.1 server.  But I got,
> 
> checking whether byte ordering is bigendian... configure: error: can not run
> test program while cross compiling

Unless you're really trying to cross-compile, this means that configure
can't get your C compiler to work.

-- 
Niels Diepeveen
Endea automatisering



From chrisw at nipltd.com  Tue Jan 25 12:40:56 2000
From: chrisw at nipltd.com (Chris Withers)
Date: Tue, 25 Jan 2000 17:40:56 +0000
Subject: Multidimensional Dictionaries: Possible?
References: <slrn88ppsg.49t.w6uv@voyager.home.com> <oqpuuqx6s1.fsf@titan.progiciels-bpi.ca>
Message-ID: <388DE028.ED24A3B1@nipltd.com>

> > Here's how I do it in Perl:
> > %dict = ();  # create empty dictionary
> > $dict{$key1}{$key2} = $string;       # make an entry
> > $newstring = ${dict{$key1}{$key2}};  # access an entry
> 
> dict = {}
> dict[key1, key2] = string
> newstring = dict[key1, key2]

What's the quickest way to do the equivalent of the following Perl in Python,
again, using the above example:

$dict{$key1}{$key2}++

I can only think of the following, I hope there is a better way! :

a =  dict[key1, key2]
a = a + 1
dict[key1, key2] = a

cheers,

Chris


From mikael at isy.liu.se  Wed Jan 19 07:04:44 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Wed, 19 Jan 2000 13:04:44 +0100 (MET)
Subject: Python Rocks! - get rid of colons
In-Reply-To: <Pine.WNT.4.21.0001191125160.243-100000@wks_3.e-shopwork.com>
Message-ID: <XFMail.000119130444.mikael@isy.liu.se>

On 19-Jan-00 Sposhua wrote:
 >  (oh, and wouln't it make life easier to change == to = and let 
 >  python work out when it is assingment and when it is comparison?)

We had this discussion a few months ago. Today

>>>a=b=c

makes both a and b equal to c (or the same as). But

>>>a=b==c

tests if b==c and throws the result to a. With your proposal, what would

>>>a=b=c

be?

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    19-Jan-00
Time:    12:59:42

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From skip at mojam.com  Sat Jan  8 20:13:18 2000
From: skip at mojam.com (Skip Montanaro)
Date: Sat, 8 Jan 2000 19:13:18 -0600 (CST)
Subject: Finite State Machines (was "Re: control structures (was "Re: Sins")")
In-Reply-To: <018e01bf58b6$74ff16f0$0100a8c0@rochester.rr.com>
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
	<14452.45436.641890.197509@beluga.mojam.com>
	<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
	<14452.60226.853060.116123@beluga.mojam.com>
	<011501bf58a4$7b0db540$0100a8c0@rochester.rr.com>
	<14453.15229.833652.832323@beluga.mojam.com>
	<018e01bf58b6$74ff16f0$0100a8c0@rochester.rr.com>
Message-ID: <14455.57518.358206.265887@beluga.mojam.com>

    Darrell> From: "Skip Montanaro"
    >> Yup...  Check out
    >> http://www.musi-cal.com/~skip/python/

    Darrell> Would have book marked your page, but I already had.  Have you
    Darrell> made good use of your FSM class ?

Yup, it's a core class used in the interpretation of commands in Musi-Cal's
email interface.

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 tim_one at email.msn.com  Tue Jan 25 03:19:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jan 2000 03:19:08 -0500
Subject: using threads with python and winnt
In-Reply-To: <388D556E.1DC8C084@zebrasoft.co.nz>
Message-ID: <000801bf670c$dab7a480$632d153f@tim>

[Matthew Miller]
> Maybe in the heat of the moment I was overly glib about Win9x
> and python threading ...

There's an exceedingly rare timing hole in the released 1.5.2 Python (fixed
in the CVS version) that had been there for years, showing up in rare
reports under extremely high rates of thread death and only on top-end
multiprocessor systems.

Now I had never seen a threading problem under Python on my uniprocessor
Win95, but constructed a program to provoke this error after finding some
*suspicious* code in Python "by eyeball".  In the course of tracking it down
and verifying Guido's fix, I chewed through hundreds of thousands of
threads, with several hundred active simultaneously, over a period of
nonstop days and in a program designed to be as stressful as possible.  No
problems at all on Win95.

Threads are one thing Windows does very well (indeed, I never worked on a
Unix(tm) system I didn't write <wink> where threads were a tenth as
reliable).  So I'd look for the cause of your problems elsewhere -- luckily
<wink>,  there are *plenty* of other places to look under Windows.

e.g.-try-unplugging-the-keyboard<wink>-ly y'rs  - tim





From warlock at eskimo.com  Sun Jan 23 19:58:28 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Sun, 23 Jan 2000 16:58:28 -0800
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com>  <86fl0m$gmj$1@inputplus.demon.co.uk>
Message-ID: <slrn88n8tk.4ff.warlock@raven.myth>

On 23 Jan 2000 19:32:38 -0000, 
 Ralph Corderoy, in the persona of <ralph at inputplus.demon.co.uk>,
 brought forth the following words...:

>> > _Python Essential Reference_ is just what it says; a reference.
>> > If, like me, you are an experienced (which says nothing about the
>> > quality of my code ;^)) coder, you can use it to learn Python.
>> 
>> Seconded.  If you already know a few languages, and already now OOP,
>> then PER is fine to learn from.
>
>OK, I learnt C from the excellent K&R (Old Testament), all the Unix
>little languages, and many more besides.  I learnt Perl 4 years ago
>purely from the man pages, and Perl 5 at the same time from
>eavesdropping on the perl5-porters mailing list.  I like well written,
>terse, books.  Given all of that it sounds to me as if I should remove
>_Programming Python_ from my shopping basket and add _Python Essential
>Reference_.  Your views are sought.
>
>
>Ralph.
>
I am new to python, I picked up learning python, and between that
and the tut and online docs, am enjoying it immensly. But I also
picked up PER, and found it a nice reference, I wouldn't want to
have to learn from just that book, but for an essential
reference, I think it is well named (can't comment on Programming
python tho', haven't bought it.)

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From gmcm at hypernet.com  Tue Jan 18 18:16:09 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 18 Jan 2000 18:16:09 -0500
Subject: More DBI question
In-Reply-To: <3884BC08.CC920737@exceptionalminds.com>
Message-ID: <1263885125-1047272@hypernet.com>

Timothy Grant writes:

> I guess I have a misunderstanding of how cursors are supposed work, or
> more likely how I am supposed to work with cursors.

Different databases have different concepts of cursors. 
Different drivers may warp those concepts in unexpected 
ways. The least common denominator is forward only.
 
Most DBs use the word cursor to mean "result set iterator", 
but the DBI spec uses the term a bit more loosely.

> Current understanding:
> 
> Open Database Connection
> Create a Cursor
> Execute SQL queries using that cursor
> Process  results
> 
> Now, I need to make all sorts of different queries using that cursor,
> some are selects, some are updates, some are deletes, some deal with
> functions
> 
> Now my misunderstanding comes when I do something like this
> 
> c.execute("""SELECT NEXTVAL ('id')""")
> 
> c.execute("""INSERT INTO mytable ( name, rank, serialno)
> 		VALUES( ?, ?, ?""", (namevar, rankvar, serialnovar))
> 
> This works as advertised
> 
> Howevever, the second time through this sequence, the first execute
> returns an error along the lines of "Expecting 3 arguments received 1"
> 
> Now I understand where the three arguments part comes from, the cursor
> just did something that required three arguments, but how do I "reset"
> what the cursor expects, it seems clunky to call a db.commit() after
> each execute.

As you might expect by now, much depends on what DB, 
what driver and what DBI implementation you're using. I 
suggest, however, closing a cursor once you're done with an 
execute and using a new cursor.

If your DBI implementation is smart enough, it can be much 
more efficient to associate a cursor with an SQL string.

cust_cursor = conn.cursor()
while something:
  cust_cursor.execute(
          'select nm from customer where id = ?', id)
  # other stuff

If you're lucky, the statement will only get parsed once.



- Gordon



From reic0024 at ub.d.umn.edu  Wed Jan 19 17:02:24 2000
From: reic0024 at ub.d.umn.edu (Aaron J Reichow)
Date: Wed, 19 Jan 2000 16:02:24 -0600
Subject: Newbie Q: Zope with static content?
In-Reply-To: <38864030.F1A08EA6@digicool.com>
References: <Pine.SO4.4.02.10001180010450.27409-100000@ub.d.umn.edu> <38864030.F1A08EA6@digicool.com>
Message-ID: <Pine.SO4.4.02.10001191600290.29076-100000@ub.d.umn.edu>


On Wed, 19 Jan 2000, Michel Pelletier wrote:

> >         Is there anyway Zope can read a file straight off the disk,
> > presumably within it's own directory structure? (I've Zope installed at
> > /var/zope)  
> 
> Use an External Method: open file, return it to the browser, close
> file.  Remember, it's all python!

I considered that, but didn't know what Zope would do with some returned
image... I'll play around! In the mean time, I managed to get a FTP-put
script written and it's working, although it seems to consume a little
more resources than xtp would've, but alas!

Thanks!
Aaron



From brian at brian.cbs.umn.edu  Thu Jan 13 15:05:38 2000
From: brian at brian.cbs.umn.edu (Brian Langenberger)
Date: 13 Jan 2000 20:05:38 GMT
Subject: generating .pyc without running .py?
Message-ID: <85lb6i$js2$1@news1.tc.umn.edu>

Is there a way to generate Python bytecode files (.pyc)
without executing the original (.py) script?  I'd like
to automate the generation process, but can't find a 
command line parameter that seems appropriate.


From gmcm at hypernet.com  Mon Jan 31 14:30:39 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 31 Jan 2000 14:30:39 -0500
Subject: C++-Extension
In-Reply-To: <3895D9C4.858A291A@daimlerchrysler.com>
Message-ID: <1262775501-10797579@hypernet.com>

Tilo Schwarz wrote:
> 
> the
> 
> Extending and Embedding the Python Interpreter
> 
> reference says, that (1.11)
> 
> "If the main program (the Python interpreter) is compiled
> and linked by the C compiler, global or static objects with
> constructors cannot be used. This is not a problem if the main program
> is linked by the C++ compiler"
> 
> Does that mean, C++-Extension are not "portable" - means, that I need
> a C++-linked python interpreter to load my C++-Extension? Not
> portable, because on a different platform I don't know if there is a
> C++-linked python interpreter and some code might use global or static
> objects with constructors. Am I right? (Hopefully not)

It is _definitely_ not portable to write a C++ extension that 
uses static objects. It is only _quasi_ portable to write a C++ 
extension without them.

> I'm asking, because I get the following error message:
> 
> > python
> Python 1.5.2 (#1, Jul 23 1999, 06:38:16)  [GCC egcs-2.91.66
> 19990314/Linux (egcs- on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import _CircleDetector
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: ./_CircleDetectormodule.so: undefined symbol: cerr
> >>>  

That's a library problem. You probably want to get the C++ std 
lib statically linked to your .so.

- Gordon



From munif at dnet.net.id  Sat Jan  1 10:48:00 2000
From: munif at dnet.net.id (Ahmad Munif)
Date: Fri, 31 Dec 1999 16:48:00 -2300
Subject: text-based builder 
Message-ID: <386E21AF.BEFE9E1D@dnet.net.id>

i'm using wxPython to build GUI Applications.
but i don't know what tool that i can use to build text based
application.

any idea ?


From hei at adtranzsig.de  Mon Jan 24 04:09:21 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Mon, 24 Jan 2000 10:09:21 +0100
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de> <869uel$suj$1@nntp1.atl.mindspring.net> <slrn88hnpb.brp.quinn@auto.ugcs.caltech.edu> <slrn88n40a.2qh6.cjc26@bridget.mindriot.net>
Message-ID: <948708715.22839@news.adtranzsig.de>

Cliff Crawford schrieb in Nachricht ...
>Pada 21 Jan 2000 22:33:56 GMT, Quinn Dunkan bilang:
>|
>| A curried function always takes one argument.  If it needs more values to
>| 'finish', it returns a closure which will accept the next arg.
>
>My semantics prof always made us call currying "Schoenfinkelization",
>because "Schoenfinkel came up with it first".
>Do you know how hard it is to spell Schoenfinkelization? ;)

That's very easy, but then, i'm german.
I don't really know what a closure is, not being
functional-programming-minded,
but could one say that a closure
is a sort of specialization of a function, i mean, a variant of a function
where a certain value replaces a "variable"s value?
And does the word "closure" play an important role outside
of the functional programming world? And what does it have
to do with scopes?

(When you have curried a lot, this would probably mean in
german: "Heute hab ich viel geschoenfinkelt." Try that!)
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From thomas at xs4all.net  Mon Jan 31 09:33:29 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 31 Jan 2000 15:33:29 +0100
Subject: tabs do WHAT?
In-Reply-To: <874sby5m6l.fsf@iname.com>; from david.maslen@iname.com on Fri, Jan 28, 2000 at 03:00:03PM +0000
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com> <874sby5m6l.fsf@iname.com>
Message-ID: <20000131153329.B19725@xs4all.nl>

On Fri, Jan 28, 2000 at 03:00:03PM +0000, David Maslen wrote:

> However, not everyone uses emacs. As an emacs user I wouldn't wish it
> on you for all it's virtues. I think the python form of indenting
> might be rather annoying if it weren't for the programing tool I
> use.

Actually, most text-editors I know under UNIX, all but Pico (which
braindead, linewrapping, rotting carcass is a 'nono' in any case) do their
indenting fine. My personal poison is Joe, which is a versatile, fast,
intuitive and non-bloated editor with wordstar-based commands (but, of
course, configurable.) It uses filenameglobs to set initial modes on
auto-indent, wordwrap and the like, but it's all easily configurable using
simple keystrokes.

And though actual tab-size is configurable, a tab stays a tab. Just using a
real tab (worth 8 spaces) saves a lot of trouble, since the interpreter also
considers it 8 spaces. To quote Linus Torvalds:

"""
Tabs are 8 characters, and thus indentations are also 8 characters. 
There are heretic movements that try to make indentations 4 (or even 2!)
characters deep, and that is akin to trying to define the value of PI to
be 3. 

Rationale: The whole idea behind indentation is to clearly define where
a block of control starts and ends.  Especially when you've been looking
at your screen for 20 straight hours, you'll find it a lot easier to see
how the indentation works if you have large indentations. 

Now, some people will claim that having 8-character indentations makes
the code move too far to the right, and makes it hard to read on a
80-character terminal screen.  The answer to that is that if you need
more than 3 levels of indentation, you're screwed anyway, and should fix
your program.
""" (CodingStyle, linux kernel source, linux/Documentation/CodingStyle)

With Python you can go a bit farther than 3 levels of indentation and still
keep it readable, thanks to the lack of visible block delimiters ;) but it's
generally good advice anyway. 

Indentation is the first item Linus addresses in his CodingStyle document...
Perversely, the second is block delimiter placement, and he spends a lot
more time explaining his guidelines in that ;)

Use-python--saves-pain-ly y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

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



From info at ibuyer.net  Mon Jan  3 17:40:38 2000
From: info at ibuyer.net (iBuyer Support Team)
Date: Mon, 03 Jan 2000 22:40:38 GMT
Subject: Nikon Coolpix950 help
Message-ID: <3871246d.5767993@news.planetdial.com>

Nikon Coolpix950 help

Help needed! :) 
I need to know what you guys think of the Nikon Cool Pix 950. Let me
know if it's worth the bucks being spent on it. Let me know all the
dirty little secrets about it. And let me know how well it worked for
you. Please send your comments to me via email, or you can just post
it here:
http://www.ibuyer.net/prod_comments.html?id=153799, I will be checking
it daily and will email you back individually with other responses
etc. 

Remember what we did with Abit? (They actually read your comments! and
liked that you as users are responding to their thirst for feedback.
We hope to be able to speak with Nikon soon. We are already meeting
with Nokia, and will try to get feedback on their product as soon as
we have selected one :)

So email me!!! I'm excited to see what you have to say about this
digital camera.

Thanks in advance for responding!

iBuyer Support
Aque




From fredrik at pythonware.com  Sun Jan 16 12:43:28 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 16 Jan 2000 18:43:28 +0100
Subject: SUGGEST: os.getuser()
References: <20000116174004.A5995@stopcontact.palga.uucp>
Message-ID: <000b01bf6049$36532140$f29b12c2@secret.pythonware.com>

Gerrit Holl wrote:
> may I suggest this function for in the posix module:
> 
> def getuser():
>     import pwd
>     return pwd.getpwuid(getuid())[0]

you'll find a more complete implementation of that
function in the getpass module:

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

</F>




From ken at be.com  Wed Jan  5 11:44:59 2000
From: ken at be.com (Kenneth McDonald)
Date: Wed, 5 Jan 2000 08:44:59 -0800
Subject: "Quick Python" (from one of the co-authors)
Message-ID: <001801bf579c$36795820$bcd8710a@HomeW98>

I haven't had the chances I would've liked to, to participate in
discussions about "Quick Python", but have been getting
comments forwarded to me by my great co-author, Daryl
Harms. There have been a number of favorable comments,
and we thank you for all of those.

As it turns out, a significant number of errors were also
introduced into the code examples during the typesetting
process, so please be on the lookout for those. This is a
pity, as Daryl (who was the main author during the second
half of the book's life) went to a fair bit of effort to run all of
the examples through the Python shell, to verify their
correctness. From the feedback so far, it appears that the
errors are not too serious, but still, we do apologize.

I hope to be getting a bit more time to participate in this
mailing list in the near future. Feel free to email questions.

Cheers,
Ken McDonald
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000105/cb5ba890/attachment.html>

From wware at world.std.com  Mon Jan 10 22:20:45 2000
From: wware at world.std.com (Will Ware)
Date: Tue, 11 Jan 2000 03:20:45 GMT
Subject: Embedding python in custom hardware
References: <t4td58.2j.ln@netcom1.netcom.com>
Message-ID: <Fo5IML.K4q@world.std.com>

mvh at ix.netcom.com wrote:
: Can I embed Python in some custom hardware where I have no OS, just
: some C libs?  How small of a memory footprint can I strip Python down
: to?  Can I then run imported code in a 'safe' environment?

You might want to check out Deeply Embedded Python:
http://www.abo.fi/~iporres/python/small.html
-- 
-----------------------------------+---------------------
 22nd century: Esperanto, geodesic | Will Ware
 domes, hovercrafts, metric system | wware at world.std.com


From ysamama at cybercable.fr  Tue Jan 18 18:19:22 2000
From: ysamama at cybercable.fr (Yann SAMAMA)
Date: Wed, 19 Jan 2000 00:19:22 +0100
Subject: WIN32 : how to list the current processes by their executable name?
References: <326536345498D311B3BC00105A39802A07457E@newsintern.dspace.de>
Message-ID: <862sff$2n4c$1@news4.isdnet.net>

Stefan Migowsky wrote in message
<326536345498D311B3BC00105A39802A07457E at newsintern.dspace.de>...
>Hi,
>
>see the example given by Mark Hammond in the win32 package
>
>...\Python\win32\scripts\killProcName.py
>


Sorry to bother you again, but I have downloaded the latest version of
Hammond's win32 package (build 128) and the script you mentioned above
doesn't work.

It complains that it cannot import win32pdh module. And I haven't found
anything in Python's subdirectories corresponding to this module?


So where do I miss the point?

Is this an NT-related issue?

I'm running these tests on my old WIN95 PC...


completely lost-ly yours

Yann







From salvador at my-deja.com  Sat Jan  8 05:39:03 2000
From: salvador at my-deja.com (Salva)
Date: Sat, 08 Jan 2000 10:39:03 GMT
Subject: Zope and jpython
Message-ID: <857447$bc5$1@nnrp1.deja.com>

Is anybody porting Zope to jpython ?


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


From sameerc1 at yahoo.com  Sun Jan 23 11:56:40 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Sun, 23 Jan 2000 08:56:40 -0800
Subject: python and www
Message-ID: <003501bf65c2$d2fe3880$918ffea9@desktop>

I want to invoke the python interpreter, pass it my script, and direct the script's output as a response to a request made by a client browser.  Without using Python with ASP, does anyone know how I would go about doing that (say using IIS or Apache).  Thanks in advance.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20000123/e64c6e00/attachment.html>

From lemiss at my-deja.com  Thu Jan 27 03:32:18 2000
From: lemiss at my-deja.com (G. Lewis)
Date: Thu, 27 Jan 2000 08:32:18 GMT
Subject: CGI: keeping mySQL awake between hits?
References: <86nn6m$tkg$1@nnrp1.deja.com> <87oga848z5.fsf@fun77.kivikko.hoas.net>
Message-ID: <86ovqi$rgp$1@nnrp1.deja.com>

In article <87oga848z5.fsf at fun77.kivikko.hoas.net>,
  Erno Kuusela <erno at iki.fi> wrote:
> zope is one other option, or fastcgi / pcgi apache modules.

Don't forget about PSI (Python Script Interface)
http://www.vex.net/parnassus/apyllo.py?i=42505591

It is interesting too.



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


From alex at kawo2.rwth-aachen.de  Tue Jan 11 22:12:54 2000
From: alex at kawo2.rwth-aachen.de (Alex Farber)
Date: Wed, 12 Jan 2000 04:12:54 +0100
Subject: Embedding from C++: ImportError: /path/to/pydmanager.so: undefined 
 symbol: FMtype
Message-ID: <387BF136.B16D7AB8@kawo2.rwth-aachen.de>

Dear colleagues,

I have to take over some C++ program which embeds Python.
Unfortunately I am new to Python and I also don't understand
the program source yet. So I am reading documentation. However 
if someone has an idea, why do I get this error message
(nm --dynamic pyfemmesh.so |c++filt |grep -i fmtype
 prints: "00007d40 D FMtype" and LD_LIBRARY_PATH is set):

  Traceback (innermost last):
    File "/home/alex/.viewer2/pystartup.py", line 6, in ?
      from pydmanager import *
  ImportError: /path/to/pydmanager.so: undefined symbol: FMtype

I would be very glad to hear it. Otherwise/anyway I will 
keep reading (any online-pointers for embedding Python?). 

Thank you
/Alex



Here is an excerpt from pyfemmesh.h
-----------------------------------
extern PyTypeObject FMtype;



Here is an excerpt from pyfemmesh.cc
------------------------------------
static char FMtype__doc__[] =
"MeshType - is equivalent to FemMesh, uses FemMEsh::SmPtr"
;

PyTypeObject FMtype = {
        PyObject_HEAD_INIT(&PyType_Type)
        0,                              /*ob_size*/
        "Mesh",                 /*tp_name*/
        sizeof(FMobject),               /*tp_basicsize*/
        0,                              /*tp_itemsize*/
        /* methods */
        (destructor)FM_dealloc, /*tp_dealloc*/
        (printfunc)FM_print,            /*tp_print*/
        (getattrfunc)FM_getattr,        /*tp_getattr*/
        (setattrfunc)0, /*tp_setattr*/
        (cmpfunc)0,             /*tp_compare*/
        (reprfunc)0,            /*tp_repr*/
        0,                      /*tp_as_number*/
        0,              /*tp_as_sequence*/
        0,              /*tp_as_mapping*/
        (hashfunc)0,            /*tp_hash*/
        (ternaryfunc)0,         /*tp_call*/
        (reprfunc)0,            /*tp_str*/

        /* Space for future expansion */
        0L,0L,0L,0L,
        FMtype__doc__ /* Documentation string */
};



Here is the content of pystart.py
----------------------------------
# Python default startup file
import sys
import pyfempool
import pyfemmesh
import pyfilehandlers
from pydmanager import *
import vlib
# import simba
print "Default modules loaded\n"


From wlfraed at ix.netcom.com  Wed Jan 19 23:07:49 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Wed, 19 Jan 2000 20:07:49 -0800
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com> <slrn889o9j.3oa.grant@grante.comtrol.com> <slrn889ptb.3rd.grant@grante.comtrol.com> <3885D3E0.4F10F28D@netscape.com> <slrn88bsde.4pp.grant@grante.comtrol.com>
Message-ID: <kISGOPd+LYTKFb4Ca3+qITH5oHFx@4ax.com>

On Wed, 19 Jan 2000 17:17:35 GMT, grant at nowhere. (Grant Edwards)
declaimed the following in comp.lang.python:


> 
> I'm still pretty fuzzy on MAPI.  At first I thougt it was the
> protocl used to talk to Exchange servers.  Then I thought it

	Add to the confusion, it has all the letters of IMAP <G>

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From moonseeker at my-deja.com  Mon Jan  3 10:41:49 2000
From: moonseeker at my-deja.com (moonseeker at my-deja.com)
Date: Mon, 03 Jan 2000 15:41:49 GMT
Subject: setting time?
References: <Pine.LNX.4.21.0001031055360.10225-100000@fep132.fep.ru>
Message-ID: <84qfvq$ba4$1@nnrp1.deja.com>

In article <Pine.LNX.4.21.0001031055360.10225-100000 at fep132.fep.ru>,
  phd at phd.russ.ru wrote:
> On Mon, 3 Jan 2000 moonseeker at my-deja.com wrote:
> > Noone an advice to set the system time?
>
>    It is TOOOO system-dependent...

Tell me for your favorite system:
Win, AIX, Solaris, IRIX, OS/2

But I thought there should be a chance to set the system time with python.

Regards,
Mike


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


From djc at itga.com.au  Tue Jan 11 16:55:28 2000
From: djc at itga.com.au (Dave Cole)
Date: 12 Jan 2000 08:55:28 +1100
Subject: Sybase module - release 0.4
Message-ID: <87k8lgclyn.fsf@heresy.itga.com.au>

This is the "first" release of my Sybase module for Python

Sybase is a module which provides a Python interface to the Sybase
relational database system (surprise). The Sybase package supports
almost all of the Python Database API, version 2.0 with extensions.

The Sybase package is designed to work with Python versions 1.5.2 and
later and Sybase versions 11.0.3 and later. It is based on the Sybase
Client Library (ct_* API), and the Bulk-Library Client (blk_* API)
interfaces.

The module contains the following items: 

      Module interface 
      Connection objects 
      Cursor objects 
      Bulkcopy objects 

The homepage http://www.itga.com.au/~djc/sybase.html contains a
description of the deviations from the Database API specification.

The module can also be downloaded from
        ftp://ftp.itga.com.au/pub/python/Sybase-0.4.tar.gz

- Dave


From wtanksle at hawking.armored.net  Tue Jan  4 16:58:02 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 4 Jan 2000 21:58:02 GMT
Subject: "sins" (aka, acknowledged language problems)
References: <6D8A17398E28D3119F860090274DD7DB4B3D51@pces.cadlab.it>  <38656610.E4DA287B@maxtal.com.au> <slrn86ar28.4b4.neelk@brick.cswv.com> <006b01bf4f7f$b3dfe920$bf2b2bc1@martelli> <3870B7C1.35D545F7@maxtal.com.au>
Message-ID: <slrn873vc5.3bk.wtanksle@hawking.armored.net>

On Tue, 04 Jan 2000 01:52:49 +1100, skaller wrote:

>	One problem with this approach is that it based around
>a generic calling protocol -- which is as slow as a wet week in Python.
>For example, I have extended the for statement in Viper, to accept
>dictionaries and sets. It is very inconvenient to get the i'th, or
>even the next, element from a set: I have a natural way to apply
>a function to each element.

>	As a result, Viper loops are one of the things that Viper
>currently does faster than CPython -- the iteration time
>is effectively zero, and that is what pystone reported on my machine.

You know, I didn't realize that.  You asked earlier about whether you
should implement 'ifor' to augment 'for'.  I suggested that you use a
seperate function, such as "for i,x in list.items():".  I see now that you
want entirely different behavior.

Okay, two suggestions.  First and best, don't add any additional keywords
or syntax; instead, build your optimizer to recognize and optimize that
idiom.  Second, add some non-keyword syntax: for example, the above could
be written as "for x[i] in list:".

-- 
-William "Billy" Tanksley, in hoc signo hack


From JoeElliott at ProfessorJoe.com  Sun Jan  9 01:33:07 2000
From: JoeElliott at ProfessorJoe.com (Joe Elliott)
Date: Sat, 08 Jan 2000 22:33:07 -0800
Subject: Amateur Adult REAL 18-24yr College Girls Videos
Message-ID: <38782BA2.6068C7CB@ProfessorJoe.com>

Hello I am Professor Joe Elliott. I produce and direct Amateur college girls
who have never before been in front of a VIDEO CAMERA and They MASTURBATE and
SUCK AND F**K me too. Check out my FREE WEB SITE. HAPPY NEW YEAR, Professor Joe



From grant at nowhere.  Mon Jan 10 12:04:53 2000
From: grant at nowhere. (Grant Edwards)
Date: Mon, 10 Jan 2000 17:04:53 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> 	<14452.45436.641890.197509@beluga.mojam.com> 	<005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <38774FBC.EB8281CF@maxtal.com.au>
Message-ID: <slrn87k49l.pup.grant@grante.comtrol.com>

In article <38774FBC.EB8281CF at maxtal.com.au>, skaller wrote:
>Skip Montanaro wrote:
>> 
>>     Darrell> Don't know if this if fair. But exceptions are slow when raised.
>> 
>> Agreed in the case of your two-branch case.  If you have a state machine
>> with a bunch of states, try/except/except/... appears to perform about as
>> well as if/elif/elif/.../else:
>
>	Yeah, but this is an artefact, since Python does not
>have a proper switch statement. If/elif ... is clearly
>linear time .. a switch is constant time. 

At least with the compilers I've used, that depends on the case
values.  If the switch case values are sparse, the code
generated is the same as it would be for If/elif.

For example:

  switch (c)
    {
    case 0: [...] break;
    case 1: [...] break;
    case 2: [...] break;
    case 3: [...] break;
    case 4: [...] break;
    case 5: [...] break;
    case 6: [...] break;
    case 7: [...] break;
    }

May generate table-lookup code that is O(1), but

  switch (c)
    {
    case      0: [...] break;
    case   3450: [...] break;
    case -49393: [...] break;
    case 233450: [...] break;
    case -14000: [...] break;
    case  -9834: [...] break;
    }
    
Would hopefully generate O(n) linear-search code.

For FSM code, one can probably chose non-sparse switch case
values for everything...

-- 
Grant Edwards                   grante             Yow!  .. he dominates the
                                  at               DECADENT SUBWAY SCENE.
                               visi.com            


From fdrake at acm.org  Fri Jan 21 12:42:05 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 21 Jan 2000 12:42:05 -0500 (EST)
Subject: ConfigParser and .ini files
In-Reply-To: <86a1s2$7ha$1@nnrp1.deja.com>
References: <86a1s2$7ha$1@nnrp1.deja.com>
Message-ID: <14472.39533.223900.313932@weyr.cnri.reston.va.us>

super_banane at yahoo.com writes:
 > The ConfigParser class in the default Python libraries
 > does not support comments given on the value lines, separated
 > by a semicolon, such as:
 > 
 > [Section]
 > keyword = value ; put your comments here

  You're right, it doesn't.
  Does anyone know of a *real* specification for the .ini format?  I
didn't know this was allowed.

 > Adding one line of code in the ConfigParser class
 > adds support for this feature. Any chance of submitting
 > this for the next Python release?
 > 
 > file ConfigParser.py line 311, add:
 > optval = string.split(optval)[0]
 > (before optname=string.strip(optval))

  This doesn't do it either.  If I have:

[foo]
bar=baz bat

your patch would give me 'baz', not 'baz bat', so this seems broken.
  In general, I'm somewhat concerned that the syntax for ConfigParser
is overloaded; it tries to handle two different formats: one contains
RFC822 headers inside sections, and the other is the .ini format.
Both are subject to some pretty funky value interpolation.
These should really be separated, because what most people want is a
conventional .ini file parser.
  Anyway, I've attached a patch that allows ConfigParser to strip out
';' comments from .ini-style entries (using '=' instead of ':' as the
value separator); RFC822-style entries should *not* be affected.  I
think this is closer to what you're looking for.
  I can check this in if this is the right approach.  I'm still not
happy with the overloaded syntax, but this might be closer to getting
the syntaxes properly implemented.  People should tell me if this is
really the right thing to do, and if it breaks anything.


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives


Index: ConfigParser.py
===================================================================
RCS file: /projects/cvsroot/python/dist/src/Lib/ConfigParser.py,v
retrieving revision 1.14
diff -u -c -r1.14 ConfigParser.py
*** ConfigParser.py	1999/10/12 16:12:48	1.14
--- ConfigParser.py	2000/01/21 17:29:54
***************
*** 297,303 ****
          )
      OPTCRE = re.compile(
          r'(?P<option>[-.\w]+)'                # - . _ alphanum
!         r'[ \t]*[:=][ \t]*'                   # any number of space/tab,
                                                # followed by separator
                                                # (either : or =), followed
                                                # by any # space/tab
--- 297,303 ----
          )
      OPTCRE = re.compile(
          r'(?P<option>[-.\w]+)'                # - . _ alphanum
!         r'[ \t]*(?P<vi>[:=])[ \t]*'           # any number of space/tab,
                                                # followed by separator
                                                # (either : or =), followed
                                                # by any # space/tab
***************
*** 327,333 ****
              if string.strip(line) == '' or line[0] in '#;':
                  continue
              if string.lower(string.split(line)[0]) == 'rem' \
!                and line[0] == "r":      # no leading whitespace
                  continue
              # continuation line?
              if line[0] in ' \t' and cursect is not None and optname:
--- 327,333 ----
              if string.strip(line) == '' or line[0] in '#;':
                  continue
              if string.lower(string.split(line)[0]) == 'rem' \
!                and line[0] in "rR":      # no leading whitespace
                  continue
              # continuation line?
              if line[0] in ' \t' and cursect is not None and optname:
***************
*** 356,363 ****
                  else:
                      mo = self.OPTCRE.match(line)
                      if mo:
!                         optname, optval = mo.group('option', 'value')
                          optname = string.lower(optname)
                          optval = string.strip(optval)
                          # allow empty values
                          if optval == '""':
--- 356,365 ----
                  else:
                      mo = self.OPTCRE.match(line)
                      if mo:
!                         optname, vi, optval = mo.group('option', 'vi', 'value')
                          optname = string.lower(optname)
+                         if vi == '=' and ';' in optval:
+                             optval = optval[:string.find(optval, ';')]
                          optval = string.strip(optval)
                          # allow empty values
                          if optval == '""':



From garry at sage.att.com  Fri Jan  7 16:08:36 2000
From: garry at sage.att.com (Garrett G. Hodgson)
Date: Fri, 7 Jan 2000 21:08:36 GMT
Subject: UDPserver bug / documentation error?
References: <8557h8$1536@newton.cc.rl.ac.uk> <38764319.2291C75B@rubic.com>
Message-ID: <387655D4.2BF0EC83@sage.att.com>

Jeff Bauer wrote:

> Here's a simple UDP server that I use for debugging
> CGI scripts:
> 
> http://starship.python.net/crew/jbauer/creosote/

very nice!  simple, elegant.

-- 
Garry Hodgson			Every night 
garry at sage.att.com		a child is born
Software Innovation Services	is a Holy Night.
AT&T Labs			 - Sophia Lyon Fahs


From anders.eriksson at morateknikutveckling.se  Fri Jan 28 10:50:27 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Fri, 28 Jan 2000 16:50:27 +0100
Subject: pytags ?????
References: <Pine.LNX.4.20.9912110027080.1090-100000@leviathan.kdart.com>
Message-ID: <sLmROD6o5Y6LSSRhCxBnVwmArO=O@4ax.com>

On Sat, 11 Dec 1999 00:31:40 -0800, Keith Dart <kdart at pacbell.net>
wrote:

>Is anyone aware of a program that generates ctags-like tags file, except
>for Python code? I guess it would be named something like pytags (ptags is
>already taken by perl-tags and prolog-tags). I can't seem to find one.

I found two in the Python 1.5.2 distribution.

/Tools/Scripts

eptags.py
>Create a TAGS file for Python programs, usable with GNU Emacs (version 18).

ptags.py
>Create a tags file for Python programs, usable with vi.

As a newbee I realize more and more that 'most' things are already
done in Python :-)


// Anders

---
English isn't my first or second language, 
so errors or anything that you find offending is not there on 
purpose it's just due to the translation.


From fredrik at pythonware.com  Wed Jan 19 15:07:06 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 21:07:06 +0100
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com> <8653jg$esi@news.or.intel.com>
Message-ID: <002401bf62b8$c30da8a0$f29b12c2@secret.pythonware.com>

AC "tye4" <tye4 at yahoo.com> wrote:
> A newer version of Python would compile both versions since it is backward
> compatible (think of the end indicators as optional.)

why bother with Python when you can use
corbascript instead?

http://corbaweb.lifl.fr/CorbaScript/

> As time goes on, the code using 'end' will increase substantially while the
> legacy code without 'end' will reduce.

so you mean that if you make 'end' optional, everyone
will add extra end statements to their code?

yeah, right...




From ivanlan at callware.com  Wed Jan 19 15:20:54 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 19 Jan 2000 13:20:54 -0700
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com>
Message-ID: <38861CA6.F99F7C65@callware.com>

Hi All--

tye4 wrote:
> 

[snip]

> 
> Perfect example.
>    if (a > 3)
>      if (b == 4)
>         c = 2;
>     else
>      c = 3;
> 
> If you add a space, the 'else' will belong to the inner 'if'. If you delete
> one, the 'else' will belong to the outer one. Highly error prone.
> 

Don't code that way to start with.  It's dumb.  Treat yourself to a good
editor.

[snip]

> Don't get me wrong. Indentation tremendously improves readability. However,
> it should not be part of the language. That way, we don't have to put an
> immense amount of effort just trying get everything into the right block.
> There is not single programming language that forces such a bizarre
> indentation rule.
> If thousands of programmers have a problem with this, why isn't it fixed.
> The solution is backward compatible:
> if x = 5:
>     print 'x is 5'
> end if
> 

You got about as much chance of getting this into Python as you do of
getting hit by a meteor.  Less, actually, because the odds for the
latter are measurable.  What you're asking for is for Python to not be
Python.  You don't like it, design your own language, build the
interpreter or compiler, and publish it.  Maybe you can get some people
to use it.  One of them won't be me.

You're saying to people who know and love Python, "Python is a great
language except that it's Python.  Let's fix it."  We don't think it's
broken.  Your insistence on swimming upstream on this issue is not only
wrong-headed but boring.

<if-we-had-some-spam-we-could-have-spam-and-eggs-if-we-had-some-eggs>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From wang at rjka.com  Fri Jan 14 17:24:22 2000
From: wang at rjka.com (Huayin Wang)
Date: Fri, 14 Jan 2000 16:24:22 -0600 (CST)
Subject: class.__getattr__ behavior:  feature or bug?
Message-ID: <200001142224.QAA27912@ws89.krumm0.com>

[Tim Peters]
> [Huayin Wang]
> > >>> class S:
> > ...    def __getattr__(self, name):
> > ...       if name == 'c':
> > ...          self.a = self.a + 1
> > ...          return self.a
> > ...       else: return None
> > ...
> > >>> s=S()
> > >>> s
> > Traceback (innermost last):
> >   File "<stdin>", line 1, in ?
> > TypeError: call of non-function (type None)
> > >>>

> Feature.  Try sticking "print name" in __getattr__ to see what's happening.
> Putting "s" on a line by itself implicitly asks Python to invoke repr(s) (to
> build an output string), so Python tries to look up the special method
> s.__repr__.  Your __getattr__ returns None, and Python complains that None
> isn't a sensible __repr__ method.

I used to think that I know what's a bug and what's a feature, now I am 
confused.  I do not understand why I would want to use __getattr__ special
method to control/define other special methods like '__repr__' when I can
just define it if I want to.

I know I got a mouthful with the last sentence :)

By the way, does any of you tested you code/idea ;) and
is infinite __getattr__ loop a bug?

> your-next-post-will-be-wondering-whether-an-infinite-
>    __getattr__-loop-is-a-bug<0.5-wink>-ly y'rs  - tim

never mind :)



From aahz at netcom.com  Mon Jan 17 15:01:35 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jan 2000 20:01:35 GMT
Subject: Pardon my ignorance, data conversion
References: <38829357.ED633932@home.com> <85u6cs$85d$1@nntp9.atl.mindspring.net> <oqpuv04ixm.fsf@titan.progiciels-bpi.ca>
Message-ID: <85vsev$et5$1@nntp9.atl.mindspring.net>

In article <oqpuv04ixm.fsf at titan.progiciels-bpi.ca>,
Fran?ois Pinard  <pinard at iro.umontreal.ca> wrote:
>aahz at netcom.com (Aahz Maruch) writes:
>> Robert W. McGwier <rwmcgwier at home.com> wrote:
>>>
>>>I need to do the simple task of taking a nonnegative floating point
>>>variable and taking its integer part and storing it in an integer.
>>>Damned if I can find how to do it.
>>
>> import math
>> i = int(math.floor(real))
>
>Would not a mere:
>   integer_value = int(real_value)
>be sufficient?

Quite possibly; not being an expert on the intricacies of floating-point
conversions, I elected to provide a straight translation -- besides, if
Robert is asking about floor(), zie probably wants other math functions,
too.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From shang.remove_edu at st.jyu.fi.edu  Tue Jan 25 06:03:01 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Tue, 25 Jan 2000 13:03:01 +0200
Subject: newbie question about self and eval
References: <86jjkj$jmu$1@news.inet.co.th>
Message-ID: <86jvu7$oit$1@mordred.cc.jyu.fi>

Prateep Siamwalla <teep at danaicorp.com> wrote in message
news:86jjkj$jmu$1 at news.inet.co.th...

--8<--

I don't know if this is the cleanest way to do it in Python, but I'd
use self.__dict__ which is a dictionary of the object's data (see
below).

> class myclass:
>
>     def __init__(self,filename='c:\\test.txt'):
>         self.fee = []
>         self.fi = []
>         self.fo = []
>         self.fum = []
>         f = open(filename,'r')
>         rl = f.readlines()
>         f.close()
>         for lines in rl:
>             token = string.split(lines,'\011')
              self.__dict__[token[0]].append(token[1])

Note that the dictionary is (of course) case sensitive, so you might
want to use self.__dict__[string.lower(token[0])].

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi

Get paid to surf the Web!
http://www.alladvantage.com/home.asp?refid=BOT021




From lcd at cotabank.com.tw  Tue Jan 11 21:19:06 2000
From: lcd at cotabank.com.tw (Anders Lin)
Date: 12 Jan 2000 02:19:06 GMT
Subject: python module for form library of ncurses?
Message-ID: <387be4c4.6293789@news.seed.net.tw>

Hi you all,

  Does anybody know if there is python module for form library of
ncurses available?

Best regards,


From marut at my-deja.com  Wed Jan 12 16:59:57 2000
From: marut at my-deja.com (marut at my-deja.com)
Date: Wed, 12 Jan 2000 21:59:57 GMT
Subject: MySQLdb, blobs, and inserting
Message-ID: <85itgj$mlc$1@nnrp1.deja.com>

I've searched python.org, deja news, and the MySQLdb docs without
finding any good examples of using INSERT.

Could someone post a few?

# file = mydoc.html
# Does file need quoting? I thought the DB 2.0
# removed the need for that...?


>>> c.execute(r"INSERT INTO items (story) VALUES (%s)" % file )
Traceback (innermost last):
File "<stdin>", line 1, in ?
File "MySQLdb.py", line 153, in execute
File "MySQLdb.py", line 189, in _query
_mysql.OperationalError: (1064, 'You have an error in your SQL syntax
near \'<!DOCTYPE HTML PUBLIC "-//W3C
//DTD HTML 4.0 Transitional//EN" "http://www.w3.org\' at line 1')


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


From jkraai at murl.com  Thu Jan  6 17:09:17 2000
From: jkraai at murl.com (jim kraai)
Date: Thu, 06 Jan 2000 16:09:17 -0600
Subject: XML to dictionary?
References: <8518kq$59n$1@nnrp1.deja.com>
Message-ID: <3875128D.C5D23D1D@murl.com>

There's a little problem:
  dictionary keys must be unique.
  XML can have more than one child node of the same name.

Depending on the DTD:
	<Newsgroup>
	  <comp.lang.python>
	    <Message>
	      <Subject>Hi there</Subject>
	      <Body>This is a test</Body>
	    </Message>
	    <Message>
	      <Subject>Re: Hi there</Subject>
	      <Body>This is a reply to:\n> This is a test</Body>
	    </Message>
	  </comp.lang.python>
	</Newsgroup>
is a valid hunk of xml.

If you're controlling this aspect of the DTD, that's fine, but I 
don't know how you'd write a DTD to encompass:
<Newsgroup>
	<Message1>...</Message1>
	<Message2>...</Message2>
	<Message...>...</Message...>
	<MessageN>...</MessageN>
</Newsgroup>

This is a 'bad idea' in XML, but then again, I haven't written 
that many DTDs.

--jim

johnjensen at my-deja.com wrote:
> 
> I have a relatively simple XML question, but I can't seem to figure out
> how to do it simply and easily with the xmllib module.
> 
> Assume I have some XML (stylized below):
> 
> <Newsgroup>
>   <comp.lang.python>
>     <Message1>
>       <Subject>Hi there</Subject>
>       <Body>This is a test</Body>
>     </Message1>
>   </comp.lang.python>
> </Newsgroup>
> 
> How can I quickly and easily stuff this into a nested dictionary, so
> that:
> 
> print dict['Newsgroup']['comp.lang.python']['Message1']['Subject']
> spits out 'Hi there'
> 
> The XML I am working with is perfect for this type of solution.
> 
> Any help would be appreciated!
> 
> John



From michel at digicool.com  Wed Jan 19 17:52:32 2000
From: michel at digicool.com (Michel Pelletier)
Date: Wed, 19 Jan 2000 17:52:32 -0500
Subject: Newbie Q: Zope with static content?
References: <Pine.SO4.4.02.10001180010450.27409-100000@ub.d.umn.edu>
Message-ID: <38864030.F1A08EA6@digicool.com>

Aaron J Reichow wrote:
> 
> Good day,
>         I hope this is an appropriate newsgroup to discuss this -- I read
> this newsgroup offen, but do not subscribe to the Zope mailing list(s).
>         I am currently using Zope to develop some web-based applications,
> and am really liking it.  Also, I am needing to have a webcam for this
> site.  I had it working fine under Apache simply updating the jpeg on the
> disk every 30 seconds.  But Zope, as far as I can see, cannot access
> "static" content straight off the disk -- like I was doing with the webcam
> jpegs with Apache.
>         Is there anyway Zope can read a file straight off the disk,
> presumably within it's own directory structure? (I've Zope installed at
> /var/zope)  

Use an External Method: open file, return it to the browser, close
file.  Remember, it's all python!

> could run an Apache process just to server the webcam, but I
> would like to avoid doing that.  The other option I see is to have the
> jpeg uploaded to the Zope FTP server.  I'm having problems getting xtp to
> work.
>         Also, I think it'd be handy, for other reasons, to know how one
> can get Zope running zserver to access static content saved as flat files
> on the filesystem.

Look at medusa's http_server, it can server static content (Zope's
zhttp_server subclasses http_server).  You will need to hack up z2.py a
bit to get it to load medusa's static server as well as Zope's dynamic
one (or instead of).

-Michel


From sei at sunse.jinr.ru  Wed Jan 19 15:37:45 2000
From: sei at sunse.jinr.ru (Evguenii Smogailov)
Date: Wed, 19 Jan 2000 23:37:45 +0300 (MSK)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
Message-ID: <Pine.SOL.3.91.1000119233305.25615A-100000@sunse>

Hi all,

Maybe someone (like PSA member) would like to submit this (below) as an 
entry to the To-Do list (in minor language design section)?
--------------------------------

New operators:
1) 'eq' (like '==');
2) 'ne' (like '!=','<>');
3) 'last' (like 'continue');
4) '..' for generating sequences (like '1..20' or 'a..z'),
like in Perl.

Generally for improving readability,
(and Perl compatibility).

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

evguenii



From gmcm at hypernet.com  Sun Jan  9 21:11:25 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 9 Jan 2000 21:11:25 -0500
Subject: no return values for __init__ ??
In-Reply-To: <387929AF.AC816307@mdlink.de>
Message-ID: <1264652208-23225161@hypernet.com>

Helge Hess wrote:

> Gordon McMillan wrote:
> > Is inheritance a matter of taste?
> 
> Strange questions ?!

Questioning a strange assertion.

> > You claim to have patched your interpreter. Try this:
> > 
> > class A:
> >   def __init__(self):
> >     pass
> > 
> > class B:
> >   def __init__(self):
> >     return A()
> > 
> > class C(B):
> >   def __init__(self):
> >     self = B.__init__(self)
> >   def method(self):
> >     print "C.method called"
> > 
> > c = C()
> > c.method()
> > 
> > Produces an AttributeError, doesn't it?
> 
> (First, *no*, it doesn't, since C.__init__ doesn't return a new
> object, but only reassigns 'self').

And how does "method" become an available to 'self'? 

> Why should I do the above ? 

To prove that returning a value from __init__ is a sane thing to 
do.

> It's a completly constructed example.

Of course. I don't have an interpreter which allows __init__ to 
return a value, so what else could I produce?

> It's similiar to this:
> 
> class A:     def __init__(self):  pass
> class B(A):  def __init__(self):  return self.method()
>
> Produces an AttributeError as well ...

Of course it does. You didn't define "method". So yours is a 
programmer error. I did define it.
 
> The problem in your example is that you are misusing return
> values, eg
>
>   class C(B):
>     def __init__(self):
>       s = B.__init__(self)
>       if (s is not None) & (s is not self): return s
> 
> will work perfectly.

I didn't realize that return values were something one could 
abuse. Fine. Add those lines to the example.

And C().method() will produce what result with your interpreter?

Or by insisting that __init__ is "just another method", have 
you succeeded in producing a special kind of class that looks, 
smells and tastes like "just another" class, except it may 
blow up on you at run time?

- Gordon



From quinn at ngwee.ugcs.caltech.edu  Fri Jan 14 03:47:22 2000
From: quinn at ngwee.ugcs.caltech.edu (Quinn Dunkan)
Date: 14 Jan 2000 08:47:22 GMT
Subject: converting tuple to list
References: <000701bf5dcd$f2b6d350$3acbd9c2@peridot.optichrome.com> <004a01bf5dd0$a77d9830$f29b12c2@secret.pythonware.com>
Message-ID: <slrn87tonc.234.quinn@ngwee.ugcs.caltech.edu>

On Thu, 13 Jan 2000 15:14:59 +0100, Fredrik Lundh <fredrik at pythonware.com>
wrote:
>Adrian Eyre <a.eyre at optichrome.com> wrote:
>> Just out of interest: what is the canonical way of converting a list
>> of tuples to a dictionary? i.e. the opposite of dict.items()
>
>I doubt you can do much better than:
>
>mydict = {}
>for k, v in mylist:
>    mydict[k] = v

You can if you don't mind an extension: mal's mxNewBuiltins has a function
that does that.


From anders.eriksson at morateknikutveckling.se  Fri Jan 14 05:59:00 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Fri, 14 Jan 2000 11:59:00 +0100
Subject: Cookies
References: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com> <14461.65478.595789.251070@ghost.bbn.com>
Message-ID: <E=9+ODir28sPnaT7jaR6PHrOEoMd@4ax.com>

On Thu, 13 Jan 2000 11:57:19 -0500 (EST), "Tim O'Malley"
<timo at bbn.com> wrote:

<snip>

>You've got lots of problems here:
>  - you switch the case of nPos in the for loop

Just a typo ;-)
>  - you never print out the cookie, so it never gets sent to the client
>

Aha! Here is my error!

>kaka = Cookie.SmartCookie( os.environ.get("HTTP_COOKIE", "") )
>

SmartCookie? You must have a newer version of Cookie than I have,
which is the one downloaded from Parnassus with version:
Cookie.py,v 2.15 1999/01/07 21:48:33

Where can I find the new one?



Thanks for your answer and for the (.now that I understand how to use
it.) excelent module!

// Anders



---
Enlish isn't my first or second language, 
so anything that you find offending is not there on 
purpose it's just due to the translation ;-)


From python-list at teleo.net  Thu Jan 13 11:32:44 2000
From: python-list at teleo.net (Patrick Phalen)
Date: Thu, 13 Jan 2000 08:32:44 -0800
Subject: Z39.50
In-Reply-To: <85kgkj$4l0u$1@www.univie.ac.at>
References: <85kgkj$4l0u$1@www.univie.ac.at>
Message-ID: <00011308340902.04499@quadra.teleo.net>

[Matthias Huening, on Thu, 13 Jan 2000]
:: Hi,
:: 
:: Im am looking for a possibility to access a Z39.50 server through CGI or
:: ASP. Would that be possible with Python? Is there a Z39.50-module or some
:: other Python interface to the Z39.50 standard?


Take a look at http://www.bsn.com



From aahz at netcom.com  Mon Jan 17 16:37:29 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Jan 2000 21:37:29 GMT
Subject: [Q] HowTo extract the system time within python
References: <85vv88$m18$2@bw107zhb.bluewin.ch>
Message-ID: <86022p$5qu$1@nntp4.atl.mindspring.net>

In article <85vv88$m18$2 at bw107zhb.bluewin.ch>,
Markus Meng <meng.engineering at bluewin.ch> wrote:
>
>I need to know the System Time on a DOS-Only system using the python
>interpreter for a script.  What is needed to get the time/date from DOS
>into python?

import time
print time.ctime(time.time())
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From aahz at netcom.com  Fri Jan 21 15:51:31 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Jan 2000 20:51:31 GMT
Subject: Bug in Calendar Module
References: <22E71DAEC504D111B78100805FFE9DC72F03CAD4@pfs21.ex.nus.edu.sg>
Message-ID: <86agsj$gst$1@nntp3.atl.mindspring.net>

In article <22E71DAEC504D111B78100805FFE9DC72F03CAD4 at pfs21.ex.nus.edu.sg>,
Liew Pol Yee <polyp at cyberway.com.sg> wrote:
>
>It seems to be unable to accept dates after January, 2038. Thanks for any
>help!

I think mxDateTime might work better for you.  Don't remember whether it
has the equivalent of calendar in it.
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From anders.eriksson at morateknikutveckling.se  Wed Jan 12 11:15:59 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Wed, 12 Jan 2000 17:15:59 +0100
Subject: Ping in Python
Message-ID: <SKd8OPFYifsAUCG=19+7XNNTjqDx@4ax.com>

Hello!

Everyday about 1.30 pm my Intenet just dies. Everything takes forever
and IE nearly almost will TimeOut on all pages. So I need to test and
document this so I can tell my ISP that there is something wrong.

So I thought: Why don't I create an Python program that will PING a
list of hosts and log the response! 

But how does a PING work?

tips, articles, source code will be very wellcome!!

// Anders

---
Enlish isn't my first or second language, 
so anything that you find offending is not there on 
purpose it's just due to the translation ;-)


From thomas at xs4all.net  Mon Jan  3 03:51:20 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Mon, 3 Jan 2000 09:51:20 +0100
Subject: Python language spec
In-Reply-To: <20000103094852.B5669@xs4all.nl>; from thomas@xs4all.net on Mon, Jan 03, 2000 at 09:48:52AM +0100
References: <E124x1i-00053N-00@ethereal> <20000103094852.B5669@xs4all.nl>
Message-ID: <20000103095120.C5669@xs4all.nl>

On Mon, Jan 03, 2000 at 09:48:52AM +0100, Thomas Wouters wrote:

> Markus Stenberg wrote a small patch to embed Python into TinyMUSH 3.0 and

And I forgot to include a URL. My apologies:

http://www.helsinki.fi/~mstenber/tm30b5pv001.diff

Early-mornings-*sigh*-ly 'yrs.

-- 
Thomas Wouters <thomas at xs4all.net>

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



From pinard at iro.umontreal.ca  Sun Jan 23 10:55:26 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 23 Jan 2000 10:55:26 -0500
Subject: Small `du' reformatter :-)
In-Reply-To: Derek Moeller's message of "Sun, 23 Jan 2000 00:20:19 -0600"
References: <oqogagfidt.fsf@titan.progiciels-bpi.ca> <8766womanm.fsf@baal.winnegan.de> <oqk8l1ef4a.fsf@titan.progiciels-bpi.ca> <20000123002019.A27888@networksplus.net>
Message-ID: <oqhfg4lr4h.fsf@titan.progiciels-bpi.ca>

Derek Moeller <moeller at networksplus.net> ?crit:

> I must say, [the `dutree' script] is quite a bit nicer. Requires most
> recent version of python, I believe, but considering that it would ship
> with it it's not a problem.

Thanks, Derek.  By the way, I needed and added options to prune the tree
either from nodes being too deep (past a certain level) or too small
(below a threshold size).  It makes the script a bit bigger, but not to
the point of ugliness, I think.  Here is the updated version.

-------------- next part --------------
#!/usr/bin/python
# Produce directory hierarchy with sizes, well sorted and indented.
# Copyright ? 1996, 1999, 2000 Progiciels Bourbeau-Pinard inc.
# Fran?ois Pinard <pinard at iro.umontreal.ca>, 1996.

# Idea from Pierre Rioux <riouxp at bch.umontreal.ca>, 1996-07-12.

"""\
Usage: tree [-l LEVEL] [-s SIZE] [PATH]...

  -l LEVEL    disregard directories more than LEVEL levels deep
  -s SIZE     disregard directories smaller than SIZE Kb
"""

import getopt, os, string, sys

def main(*arguments):
    # Decode options.
    level_option = size_option = None
    options, arguments = getopt.getopt(arguments, 'l:s:')
    for option, value in options:
        if option == '-l':
            level_option = int(value)
        elif option == '-s':
            size_option = int(value)
    # Get data properly sorted.
    items = map(presort,
                os.popen('du %s' % string.join(arguments)).readlines())
    items.sort()
    items = map(postsort, items)
    # Filter the results.
    if level_option or size_option:
        new_items = []
        for item in items:
            split, size = item
            if not level_option or len(split) <= level_option:
                if not size_option or size >= size_option:
                    new_items.append(item)
        items = new_items
    # Erase unneeded fragments of vertical lines.
    clean(items)
    items.reverse()
    clean(items)
    items.reverse()
    # Erase leftmost white columns.
    skip = 0
    split, size = items[0]
    while not split[skip]:
        skip = skip + 1
    # Produce display.
    for split, size in items:
        sys.stdout.write('%7d ' % size)
        for item in split[skip:-1]:
            if item:
                sys.stdout.write('|  ')
            else:
                sys.stdout.write('   ')
        sys.stdout.write('\\_ %s' % split[-1])

def presort(line):
    size, path = string.split(line, '\t')
    return path, int(size)

def postsort((path, size)):
    return string.split(path, '/'), size

def clean(items):
    # The basename is always written.  Intermediate directories
    # normally print as `|', yet None is used to inhibit the line.
    draw = []
    for split, size in items:
        while len(draw) > len(split) - 1:
            draw.pop()
        while len(draw) < len(split) - 1:
            draw.append(0)
        draw.append(1)
        for counter in range(len(split) - 1):
            if not draw[counter]:
                split[counter] = None

if __name__ == '__main__':
    apply(main, sys.argv[1:])
-------------- next part --------------

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

From I.Clarke at strs.co.uk  Tue Jan  4 12:01:13 2000
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Tue, 04 Jan 2000 17:01:13 +0000
Subject: Quick Python v.s. Learning Python
References: <387152EB.10DD0E17@Lextek.com>
Message-ID: <38722759.55DE54F8@strs.co.uk>

Strangely, the best introduction to Python I ever read was the supplied
tutorial.  It, combined with the library reference, are the only two
pieces of Python documentation I have ever needed or wanted.

Ian.


--
Ian Clarke                        Email: I.Clarke at strs.co.uk
Homepage: http://www.gnu.demon.co.uk/
Also see: http://freenet.on.openprojects.net/
"Let me assure you that at First National, you're not just a number.
You're two numbers, a dash, three more numbers, another dash, and
another
number." - James Estes


From steriana at gvsu.edu  Thu Jan  6 13:27:53 2000
From: steriana at gvsu.edu (Andrew Sterian)
Date: Thu, 6 Jan 2000 13:27:53 -0500
Subject: Simple cartesian product
References: <000801bf571c$0cf448a0$1a2d153f@tim>
Message-ID: <3874deef@news.gvsu.edu>

"Tim Peters" <tim_one at email.msn.com> wrote in message
news:000801bf571c$0cf448a0$1a2d153f at tim...
> Search in DejaNews for p13x25.

...and Tim's p13x25 Cartesian product along with a small test suite and my own
version of the Cartesian product is available from my Python page:

  http://claymore.engineer.gvsu.edu/~steriana/Python

Andrew.



From wtanksle at hawking.armored.net  Mon Jan 24 16:00:28 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 24 Jan 2000 21:00:28 GMT
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com>
Message-ID: <slrn88pg1h.scl.wtanksle@hawking.armored.net>

On Mon, 24 Jan 2000 00:43:42 -0500, Christian Tismer wrote:

>SLP has been successfully compiled on Solaris (Thanks to Jeff
>Collins). It will be used on PalmIII, to solve the 4k stack problem.

Congrats on the huge success!

I have a question, though.  What's that about the Palm?  When will the
Palm version be available?  What will it require?  What will it provide?
Will it only work on the Palm III, or will it be okay on anything above as
well (Palm V, Visor, TRG, etc)?

-- 
-William "Billy" Tanksley, in hoc signo hack


From steve at acer.spvi.com  Sat Jan  8 15:39:04 2000
From: steve at acer.spvi.com (Steve Spicklemire)
Date: Sat, 8 Jan 2000 15:39:04 -0500 (EST)
Subject: pipe problem on FreeBSD....
Message-ID: <200001082039.PAA12903@acer.spvi.com>

Hi Folks,

   Hmmm.. what I thought started out as a "Zope Bug" isn't. Let me 
forgo all the history here and cut to the current analysis. There
seems to be a problem on FreeBSD with python sending large amounts of 
data through a pipe that blocks. I saw it first in Zope when the
DC folks changed their File object to break large files into chunks
and started sending them one chunk at a time rather than as one
big chunk. Here is the reference to that problem report:

http://lists.zope.org/pipermail/zope/1999-December/015874.html

I was oddly dissapointed when I found that in fact this was only 
a problem on FreeBSD (2.2.8, and 3.2 both exhibit the problem.)
I think I've now simplified it to it's most basic form:

This python script:

tryit.py:
----------------------------------------------------------------------
#!/usr/local/bin/python

import sys

f = open('blah')

while(1):
    s = f.read(1024)
    if not s:
        break

    sys.stdout.write(s)

f.close()
----------------------------------------------------------------------

fails on FreeBSD when invoked thusly:

pluto.spvi.com> python tryit.py | head -3
(83d0"@eKBdP5!!!!-m4F"e&*2!GP8I3!!!!!!!!!!!GD,D#@R`!!!!!!!!!!!!!
!!!!!!!!!!!!+Mj!!!!!!!2rrrrp"8&"-8(PdB5%!Y(r$aE4r`mF!#5KQ!"-ic`!
#f3d!"lAp(RrP2!!!!!!!!,iF"X"9[CqGcE0IZ3fdcPNc0pZq!aYXRD4GeXSrbYF
Traceback (innermost last):
  File "tryit.py", line 12, in ?
    sys.stdout.write(s)
IOError: [Errno 32] Broken pipe

where this c program does not:

tryit.c
----------------------------------------------------------------------

#include <stdio.h>

FILE *f;
size_t charCount;
int moreStuff;
char buffer[1024];

int main() {
  f = fopen("blah","r");
  if (f) {

    moreStuff = 1;

    while(moreStuff) {
      charCount = fread(buffer, 1, 1024, f);
      write(fileno(stdout), buffer, charCount);
      if (charCount < 1024) {
        moreStuff = 0;
      }
    }
    fclose(f);
  }
  else
    {
      printf("Can't open file... blah\n");
    }
}

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

pluto.spvi.com> ./tryit | head -3
(83d0"@eKBdP5!!!!-m4F"e&*2!GP8I3!!!!!!!!!!!GD,D#@R`!!!!!!!!!!!!!
!!!!!!!!!!!!+Mj!!!!!!!2rrrrp"8&"-8(PdB5%!Y(r$aE4r`mF!#5KQ!"-ic`!
#f3d!"lAp(RrP2!!!!!!!!,iF"X"9[CqGcE0IZ3fdcPNc0pZq!aYXRD4GeXSrbYF
pluto.spvi.com>


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

My thinking is that they should behave in the same way. (This also
happens with 'more' and other such blocking programs.) Where in python
should I look for the difference between tryit.c and tryit.py? Of
course on linux both programs *do* behave the same way... I just
created the 'c' version to see if it was deep in FreeBSD or somewhere
in the python port. The python and c version work correctly in linux.
<dang!>

thanks,
-steve




From robin at jessikat.demon.co.uk  Fri Jan  7 18:57:18 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Fri, 7 Jan 2000 23:57:18 +0000
Subject: bsddb item size?
References: <mrE03AA5nOd4Ew2R@jessikat.demon.co.uk>
 <wzd7rd4kcd.fsf@sunshine.cs.uu.nl>
Message-ID: <1y9ghGAe1nd4Ewk9@jessikat.demon.co.uk>

In article <wzd7rd4kcd.fsf at sunshine.cs.uu.nl>, Piet van Oostrum
<piet at cs.uu.nl> writes
>>>>>> Robin Becker <robin at jessikat.demon.co.uk> (RB) writes:
>
>RB> Anybody know if there are item size limitations with bsddb?
>
>I don't think so. They specifically chose an algorithm with no limitations
>(other than total memory size).
thanks that's a great help. How about collisions are they possible?
-- 
Robin Becker


From aahz at netcom.com  Sat Jan 22 19:32:19 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Jan 2000 00:32:19 GMT
Subject: Python dumps core with this.. how come?
References: <000001bf651d$f05ac240$132d153f@tim> <m31z79wzlm.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <86di6j$uch$1@nntp6.atl.mindspring.net>

In article <m31z79wzlm.fsf at atrus.jesus.cam.ac.uk>,
Michael Hudson  <mwh21 at cam.ac.uk> wrote:
>
>I'm gradually forming the opinion that writing *really* good code in
>Python is no easier than in any other language, it's just that writing
>merely very good code is much easier.

Bingo.  And writing clean, efficient code gets changed from extremely
difficult to simply "difficult".
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From milton at isomedia.com  Thu Jan 20 19:57:18 2000
From: milton at isomedia.com (Stephen Milton)
Date: Thu, 20 Jan 2000 16:57:18 -0800
Subject: NetUserAdd problem
Message-ID: <s8fbpq3heh6108@corp.supernews.com>

I have a script that attempts to use the NetUserAdd function from the
win32net library.  On most of my machines, it works great, but on one NT4
server system, it fails with the following error:

Traceback (innermost last):
  File "C:\Scripts\QUICKT~1.PY", line 26, in ?
    win32net.NetUserAdd(server, 1, d)
pywintypes.api_error: (2351, 'NetUserAdd', 'This computer name is invalid.')

I have tested using the win32netdemo.py file that comes with win32all-125,
and I get the same error.  I can change the computer name to all variations
(IP address, FQDN, machine name, UNC-style, etc.) and it just won't work on
this machine.

Any ideas of what to try to do to this machine to get it to work?  Has
anyone else seen similar problems?

E-mail replies appreciated, I can post a fix summary once I figure it out.

Steve Milton
ISOMEDIA.COM





From hartmut at oberon.noris.de  Wed Jan  5 16:38:51 2000
From: hartmut at oberon.noris.de (hartmut Goebel)
Date: Wed, 05 Jan 2000 22:38:51 +0100
Subject: RPM-interface/module
References: <386B3CDB.FF843280@bibsyst.no> <16600370@oberon.noris.de> <84r14b$l1rms@fido.engr.sgi.com>
Message-ID: <16600389@oberon.noris.de>

Paul Jackson (pj at sgi.com) schrieb:

 >I see it on Red Hat 6.1 CD #1, under /misc/src/anaconda

Ah, thanks! I missed this, since my RH-based distribution has this directory
packed to gain some space on CD.

+++hartmut


From phd at phd.russ.ru  Mon Jan  3 05:44:41 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 3 Jan 2000 10:44:41 +0000 (GMT)
Subject: How to handle and create cookie with python??
In-Reply-To: <4oXb4.1127$UO3.31620@wagner.videotron.net>
Message-ID: <Pine.LNX.4.21.0001031039370.10225-100000@fep132.fep.ru>

On Mon, 3 Jan 2000, Akhar wrote:
> I would like to know how can I create cookies and use them for various cgi

   Using module Cookie.py. It is not in the standard distribution - just search
it through python.org. It is very nice python-style module!

> with python? Will python be fast enough for  a busy web site? Will it scale
> properly?

   There are many way to speed things up. You can rewrite critical parts in
C; or better yet, reorganize your site to serve static files and recreate
these files from cron every few minutes. For really dynamic sites to speed
things up you need to try PyApache or httpdapy - builtin apache modules.
You can try also PersistentCGI.

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From kelly at poverty.bloomington.in.us  Fri Jan 14 16:57:53 2000
From: kelly at poverty.bloomington.in.us (Kelly Lynn Martin)
Date: Fri, 14 Jan 2000 16:57:53 -0500
Subject: cgi use root uid 
In-Reply-To: Your message of "Fri, 14 Jan 2000 16:10:57 +0800."
             <85mlpc$2uac$1@news.cz.js.cn> 
Message-ID: <200001142157.QAA05130@poverty.bloomington.in.us>

On Fri, 14 Jan 2000 16:10:57 +0800, "????" <jfzhou at 188.net> said:

>Can someone tell me how to run a python cgi that can have root right.
>the main perpose is to modify some file like passwd,hosts

Since Python is a script interpreter, you'll have to use a cgi wrapper
to avoid the setuid script race security hole (which is why most UNIX
kernels disallow setuid scripts).  The Perl perlsec manpage explains
how to do this for Perl; essentially the same approach should work
with Python.

Kelly



From georgethomas at pspl.co.in  Wed Jan 12 23:53:38 2000
From: georgethomas at pspl.co.in (George Thom@s)
Date: Thu, 13 Jan 2000 10:23:38 +0530
Subject: Python and MAPI
Message-ID: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in>

Hi everyone,
I'd like to know if anyone has worked with MAPI using Python with the aid of
either Mark Hammond's win32 extensions or Sam Rushing's calldll/dynwin
package. Some sample code would also be useful, since there is none in the
distribution.
Regards,
George
--------------------------------------------------------------------------
George Thomas
Resident,Planet Earth,Third Rock from the Sun
georgethomas at pspl.co.in


Flat No. FI-5, Fifth Floor,
Himagirinath Co-op. Hsg. Society,
486A Elphinstone Road,
Kirkee, Pune - 411003,
Maharashtra State,
India.
--------------------------------------------------------------------------




From felixt at dicksonstreet.com  Wed Jan  5 22:56:35 2000
From: felixt at dicksonstreet.com (Felix Thibault)
Date: Wed, 05 Jan 2000 21:56:35 -0600
Subject: Question about dictionaries (Question about is ?)
Message-ID: <3.0.5.32.20000105215635.0091d420@mail.dicksonstreet.com>

Hi Everybody-

I've been working with nested dictionaries whose keys are slices of text
from a document, and I've come up with two ways to make them:

>>> text = 'spam-a-lam-a-spama'
>>> names = {'spam': text[:4]}
>>> dict1 = {text[:4]: {text[-5:-1]: 'eggs'}}
>>> dict2 = {names[text[:4]]: {names[text[-5:-1]]: 'eggs'}}

with the difference being:

>>> dict1.keys()[0] is dict1['spam'].keys()[0]
0
>>> dict2.keys()[0] is dict2['spam'].keys()[0]
1

...but I'm new to programming as well as to Python, so I'm not sure what
this means- does dict2
take up less memory, or something, or is this just a Stupid Programming
Trick? Is this like 
intern(..)'ing ? 


Thanks in advance,

Felix






From flight at mathi.uni-heidelberg.de  Mon Jan 31 11:37:28 2000
From: flight at mathi.uni-heidelberg.de (Gregor Hoffleit)
Date: Mon, 31 Jan 2000 17:37:28 +0100
Subject: Reading pipermail archives with mutt
Message-ID: <20000131173728.D5068@mathi.uni-heidelberg.de>

Somehow mutt is not able to read the mail archives downloadable from
Python.org's SIG pages. The simple /bin/mail will grok them correctly
and list the contents, but mutt doesn't find any mails.

IIRC, that's true not only for Python.org's mailing list archives, but
also for other archives. I'm not sure, though, if these archives perhaps
are all written by pipermail.

Has anybody else noticed this before ? Is there any simple resolution ?

    Gregor



From spidaman at well.com  Mon Jan 10 19:36:16 2000
From: spidaman at well.com (spidaman at well.com)
Date: Tue, 11 Jan 2000 00:36:16 GMT
Subject: [job] infoseek python patching
Message-ID: <85dttu$28g$1@nnrp1.deja.com>

Looking for an experienced python hand to come work with us periodically
to extend and customize the behavior of our ultraseek search engine.
This is on a short term contracting and consulting in downtown San
Francisco (though if there are other programming skills that fit with
other areas of our needs, we are hiring for a full time s/w engineer :).

Please reply to spidaman at well.com with rates, resume and availibility.


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


From m.faassen at vet.uu.nl  Sat Jan 15 11:31:18 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Jan 2000 16:31:18 GMT
Subject: ANN - PyRecode
References: <20000113144712.A666@starship.python.net> <FoAJ3M.FDA@world.std.com>
Message-ID: <85q7cm$kio$1@newshost.accu.uu.nl>

Will Ware <wware at world.std.com> wrote:
[snip]
> Alas, there is no English<->Dutch translation
> in Babelfish.

That shouldn't be necessary, right? I mean, everybody knows everybody who
uses Python speaks Dutch. :)

Regards,

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


From wpostma at earthlink.net  Sun Jan 23 22:07:40 2000
From: wpostma at earthlink.net (Warren Postma)
Date: Mon, 24 Jan 2000 03:07:40 GMT
Subject: tabs do WHAT?
Message-ID: <0mPi4.3349$bp2.155298@newsread2.prod.itd.earthlink.net>

Let's see how many languages I have run into that make whitespace and/or
tabs important:

1. Makefiles: I once had a problem where a rule had a space where one was
not allowed, at the end of a line, causing me to spend three days finding
the problem.

2. Multi-line C Macros. Double Blech!  The '\' macro-continuation-character
must be the last character on the line.  Ever been hit with this one? Yuck.
At least you only do it once and then you learn.

3. Anything programmed on punch cards.  I seem to remember that the first 8
columns were used for line numbers (Fortran?)  and the middle bunch were for
code, and the last column was special somehow, and whitespace could
sometimes make or break your code.  Extra Yucky.

4. Python.

The two polar opposites (otherwise beautiful syntax, and the tabbed scoping)
are so fascinatingly horrible that you have to do a double- It's like the
commercial of the beautiful girl who's otherwise perfect smile is marred by
yellow nicotine stains. Sorry for the visual image, but hey, I gotta say, it
just doesn't fit with the rest of the language, and I don't buy it one bit.

A programmer/professor I once had for a software engineering course told us
about the "principle of least amazement": Anything weird or hard to explain
that you write into your software will later become the thing you regret.

As an embedded language, Python needs to be usable beyond the ivory towers
where functional programming or object oriented programming or metaclasses
are fodder for discussion.   Python looked eminently useable by
non-programmers until I found out that whitespace was meaningful.  Try to
explain scope to someone who doesn't write programs for a living.

Even BASIC, ugly as it is from a functionality point of view, is miles ahead
from the point of view of useability by people who are skilled technical
people, yet who are not "programmers".    Ergo, Visual Basic's popularity.
The language is a dog's breakfast, but at least the basic feature set you
use every day is clear and clean enough to keep in your head, and still have
room for your other domain-specific knowledge.

Python seems one of the few languages that is elegant enough for a computer
science type to appreciate, and yet not cluttered up with syntactic garbage
as to become an impenetrable mess (TCL, Perl, and the Granddaddy of
Unreadable Languages, )...   So close, and yet so far away. Has anyone built
a version of Python that totally ignores tabs and whitespace, or a filter
that you can use on top of Python? Perhaps  it could support both 'begin'
and 'end', or  '{' and '}' for scoping.  Heck, at least you could write it
in Python.  An interpreter within an interpreter. That would probably be
very slow, wouldn't it?

If Guido reads this, please note my personal feature request for Python 2.0:
Make the "whitespace" stuff optional, for people who like weird syntactical
features only. I'll love you forever. Thanks.

Warren Postma




From erno at iki.fi  Wed Jan 26 20:08:30 2000
From: erno at iki.fi (Erno Kuusela)
Date: 27 Jan 2000 03:08:30 +0200
Subject: CGI: keeping mySQL awake between hits?
References: <86nn6m$tkg$1@nnrp1.deja.com>
Message-ID: <87oga848z5.fsf@fun77.kivikko.hoas.net>

>>>>> "jhefferon" == jhefferon  <jhefferon at my-deja.com> writes:

    jhefferon> I understand that I can put Python into Apache as a
    jhefferon> module, so that Python will always be awake.  I want to
    jhefferon> also keep mySQL up (maybe a half-dozen clients).  I
    jhefferon> believe that I can do this with Perl's Apache::DBI, but
    jhefferon> I'd much to prefer to go with Python.

there are 2 python modules for apache, PyApache and httpdapi.
PyApache wouldn't help you here as it creates a new python interpreter
for each request, it only saves the fork/exec overhead. i'm currently
using httpdapi.

zope is one other option, or fastcgi / pcgi apache modules.

  -- erno


From tseaver at starbase.neosoft.com  Fri Jan 21 21:22:21 2000
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 21 Jan 2000 20:22:21 -0600
Subject: In defence of the two-namespace rule
References: <000601bf6452$c123ada0$caa0143f@tim>
Message-ID: <9E9B938C3350EEE6.5AB8C6ED94CC65E7.EFF3CDC64D031C30@lp.airnews.net>

In article <000601bf6452$c123ada0$caa0143f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>every-bird-in-the-forest-is-flawed-ly y'rs  - tim

If-a-bird-flaws-in-the-forest-but-noone-hears?

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From ivanlan at callware.com  Sat Jan  1 23:02:23 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Sat, 01 Jan 2000 21:02:23 -0700
Subject: Need help with canvas widget in Tkinter please
References: <386EC3E6.FAC5BC0@telebot.com>
Message-ID: <386ECDCF.A6B91CD5@callware.com>

Hi Flounder

flounder wrote:
> 
> I cannot seem to figure out how to add images to a canvas with python/Tkinter
> 
> I can do it with Tcl/Tk and Perl/Tk but not with python i have tried this but
> 
> it did not work at all. Here is the code i used
> 
> -----------CODE--------------
> 
> #!/usr/bin/python
> 
> from Tkinter import *
> 
> mw = Tk()
> mw.title("Map Editor")
> photo = PhotoImage(file="tile0.gif")
> 
> display = Canvas(mw, width=640, height=480).pack()
> 
> item = display.create_image(0, 0, anchor=NW, image=photo)
> 
> image = display.addImage
> mainloop()
> 
> --------------END OF CODE---------------
> 
> when i run this it gives me this error
> 
> -----------ERROR MESSAGE--------------
> 
> Traceback (innermost last):
>   File ".//test.py", line 11, in ?
>     item = display.create_image(0, 0, anchor=NW, image=photo)
> AttributeError: 'None' object has no attribute 'create_image'
> 
> -----------END OF ERROR---------------
> 
> can anyone show me how this should be done I have looked at the
> 
> documentation and can't figure it out
> 
> Thanks in advance
> 
> --
> Flounder
> 
> --
> http://www.python.org/mailman/listinfo/python-list

#!/usr/bin/python

from Tkinter import *

mw = Tk()
mw.title("Map Editor")
photo = PhotoImage(file="tile0.gif")

display = Canvas(mw, width=640, height=480)

item = display.create_image(0, 0, anchor=NW, image=photo)

display.pack()
mainloop()

----------------------------------------------
Your main problem was calling Canvas(...).pack(), since the pack()
method returns None.

And no need to add_image(), either.  create_image() does everything you
need.

<can't-add-things-to-None>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From fredrik at pythonware.com  Mon Jan 17 05:41:01 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 17 Jan 2000 11:41:01 +0100
Subject: Very Easy question concerning input
References: <D2zg4.1375$Pb4.11132@newsfeed.slurp.net>
Message-ID: <005001bf60d7$59eb4810$f29b12c2@secret.pythonware.com>

> can someone tell me how to use a variable from a previous line and get a
> variable from the currentline using raw_input, for example
> 
> I am trying to make a simple program that ask the user his name then says
> "welcome (name) would you like to do something"
> 
> I tried this...but it doesn't work right...
> 
> name=raw_input ("What is your name?")
> choice1=raw_input "welcome, %s, would you like to do something?" % name
>
> doing this displays a %s instead of the persons name

really?  I get a syntax error:

  File "sample", line 2
    choice1=raw_input "welcome, %s, would you like to do something?" % name
                                                                   ^
SyntaxError: invalid syntax

> can you tell me how to fix this?

just add parentheses:

choice1 = raw_input("welcome, %s, would you like to do something?" % name)

</F>




From larsde at telepost.com  Mon Jan 17 20:48:43 2000
From: larsde at telepost.com (larsde at telepost.com)
Date: Tue, 18 Jan 2000 01:48:43 GMT
Subject: Trouble with Python and ASP
Message-ID: <860gpo$882$1@nnrp1.deja.com>

I'm experiencing disturbing problems with using Python as an ASP
scripting engine. I've been working on a project for quite some time,
and so far everything has been peachy.

Trouble is when multiple users try to connect to the web-application at
the same time. IIS either just hangs for a long while, usually until
one of the users try to press stop and resubmits the form.

Other times, IIS just dies with the following error message logged to
the event-viewer:

An object call caused an exception.   (Package: IIS-{Default Web
Site//Root/Signup3.0})    (IID: {51372AEF-CAE7-11CF-BE81-00AA00A2FA25})
(Method: 3)  (Microsoft Transaction Server Internals Information: File:
d:\viper\src\runtime\mtxex\activity.cpp, Line: 889) (Exception:
C0000005) (Address: 0x1e124188)
python15!PyErr_GivenExceptionMatches + 0x58

This happens when two users try to access the same asp-page
simultaneously.

I'm using some COM/ADO over ODBC and things like this.

When searching the web for information about this, I found another
posting describing the same problems, even for simple ASP-files.

I would love any help on this. It would be sad if it turns out python
with asp can only handle one connection at a time. It would certainly
force me to reimplement all my code in some other language.

On my system I'm running Win32all-128.exe install with the 152-version
of the python interpreter.

Thanks,

Lars Thomas Denstad.


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


From claird at starbase.neosoft.com  Sat Jan 22 00:19:41 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 21 Jan 2000 23:19:41 -0600
Subject: CGI and Python
References: <86apvo$qie$1@nnrp1.deja.com>
Message-ID: <C34E08ABEDCAE412.8FC89D9CBBD88811.BA1C182C6E72CA84@lp.airnews.net>

In article <86apvo$qie$1 at nnrp1.deja.com>,  <yduchesne at my-deja.com> wrote:
			.
			.
			.
>a) I know there are a couple of modules out there that make embedding
>Python into Apache possible; does this allow for a programmer to have
>Python objects residing in the server's process (i.e. as for Java
>servlets) - and be accessible to all Python scripts running on the
>server? Can a python "servlet" access other python "servlets"?
Yes and no--some of each.
>
>c) How is embedded (in Apache) Python performance? Does it compare to
>mod_perl or Java servlets?
Yesssss ...  In fact, my own experience with
server-side Python showed it measurably slower
than corresponding Perl and Java work, but I
know there are domains where it's faster.
General conclusion:  Python is fast enough.
>
>d) I would like someone(s) to come back to me and tell me about what
>module they recommend - PCGI, PyApchache...(please, some -even
>telegraphic style- explanations would be appreciated)
Whenever I can make time, I push <URL:http://
starbase.neosoft.com/~claird/comp.lang.python/web_python.html>
in the direction of answering these questions.
			.
			.
			.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From grant at nowhere.  Tue Jan 18 16:55:00 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 18 Jan 2000 21:55:00 GMT
Subject: Python and MAPI
References: <002301bf5d82$285cdf50$cf02a8c0@intranet.pspl.co.in> <387DF8C4.4B3214CE@netscape.com>
Message-ID: <slrn889o9j.3oa.grant@grante.comtrol.com>

In article <387DF8C4.4B3214CE at netscape.com>, Michel Orengo wrote:

>I've just started to play around with MAPI. I'm far from being a expert but here
>is a sample of what you can do to send a message:
>
>import win32com.client
>session = win32com.client.Dispatch("MAPI.Session")
>session.Logon("Michel Orengo")
>outbox=session.Outbox
>collmsg=outbox.Messages
>msgo=collmsg.Add("my subject","my
>text",win32com.mapi.mapitags.PR_MESSAGE_CLASS,1)
>msgo.Recipients.Add("Michel","morengo at morganlabs.com")
>msgo.Sender.Address="morganea at bellatlantic.net"
>msgo.Send(1,0)

Here's the code I ended up trying to run:

------------------------------------------------------------
import win32com.client
import win32com.mapi.mapitags

session = win32com.client.Dispatch("MAPI.Session")
session.Logon("MS Exchange Settings")
outbox=session.Outbox
collmsg=outbox.Messages
msgo=collmsg.Add("subj text","message text",win32com.mapi.mapitags.PR_MESSAGE_CLASS,1)
msgo.Recipients.Add("grante at visi.com")
msgo.Sender.Address="grant.edwards at comtrol.com"
msgo.Send(1,0)
session.Logoff
------------------------------------------------------------

It bombs at line msgo.Sender.Address="" line with the following
error:

com_error: (-2147221233, 'OLE error 0x8004010f', (1271, 'Collaboration Data Objects', ' [Collaboration Data Objects - [MAPI_E_NOT_FOUND(8004010F)]]', None, 0, -2147221233), None)

I'm a complete newbie when it comes to win32 programming, so if
somebody could lend me a pointer, I'd appreciate it.

PS: Once I get the above example to work, what I _really_ want
    to do is to retreive messages from my inbox, but I can't
    figure out what operations are avaiable for mapi objects.
    
    Any pointers in that respect?
    

-- 
Grant Edwards                   grante             Yow!  Do you need
                                  at               any MOUTH-TO-MOUTH
                               visi.com            resuscitation?


From scoobysnax5336 at my-deja.com  Tue Jan 11 23:17:16 2000
From: scoobysnax5336 at my-deja.com (scoobysnax5336 at my-deja.com)
Date: Wed, 12 Jan 2000 04:17:16 GMT
Subject: strange CGI problem
References: <85ga10$p8u$1@nnrp1.deja.com>
Message-ID: <85gv81$8o1$1@nnrp1.deja.com>

In article <85ga10$p8u$1 at nnrp1.deja.com>,
  sp00fD <sp00fD at yahoo.com> wrote:
> I'm creating a little web-site; for some reason when I view the page
in
> Internet Explorer, it's fine, but Netscape shows me the raw html as if
> it's a text file.  What would cause this?
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

Sounds like you need to assign a mime type on the web server.  Some
browsers assume that they are smarter than the user and decide for you.


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


From gmcm at hypernet.com  Wed Jan 19 16:05:23 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 19 Jan 2000 16:05:23 -0500
Subject: Python Rocks! 
In-Reply-To: <00Jan19.125324pst."3619"@watson.parc.xerox.com>
References: Your message of "Wed, 19 Jan 2000 11:58:47 PST."             <002401bf62b8$c30da8a0$f29b12c2@secret.pythonware.com> 
Message-ID: <1263806570-5772209@hypernet.com>

> > why bother with Python when you can use corbascript instead?
> 
> Because Python is a widely used and mature languages with oodles of
> libraries and experience programmers?

No way. Those people don't count. They're so weird, they read 
& write code without block delimiters!

if-God-had-intended-we-use-Guidos-indentation-he-wouldn't-
have-put-a-}-key-on-all-those-keyboards-ly y'rs

- Gordon



From sholden at bellatlantic.net  Thu Jan  6 16:38:09 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Thu, 06 Jan 2000 21:38:09 GMT
Subject: Origin of double underscore convention?
References: <Y96d4.1764$oF2.5206@ozemail.com.au>
Message-ID: <38750B59.75FB528B@bellatlantic.net>

Probably just to avoid naming conflicts: since Python binds names
at run-time it looks them up in dictionaries.  If common methods
had names like "init" or "repr" the chances of conflict would be
much higher.

Of course with "mix-in" classes and/or multiple inheritance those
conflicts could more easily occur.  This was why "name mangling"
was introduced: again, to lower to odds of naming conflicts.

regards
 Steve

Jason Stokes wrote:
> 
> I'm relatively new to Python.  This is possibly a dumb question, but
> I'm wondering about the double underscore convention to mark an inbuilt
> or internal method.  I find it pretty ugly and I'm wondering if it was
> chosen for any historical reason, eg analogy to another language.  I
> know C programmers sometimes use the underscore convention.
> 
> --
> Jason Stokes: jstok at bluedog.apana.org.au


From just at letterror.com  Thu Jan 20 18:23:05 2000
From: just at letterror.com (Just van Rossum)
Date: Fri, 21 Jan 2000 00:23:05 +0100
Subject: Socket module on Mac
In-Reply-To: <388797dc.2416809@news.btx.dtag.de>
Message-ID: <l03102801b4ad4921f169@[193.78.237.120]>

At 11:21 PM +0000 20-01-2000, Stefan Franke wrote:
>Are the socket module - and based on that, asynccore and asyncchat -
>fully available by Python on the Mac platform?
>
>Sorry, I don't have a box here to check it myself.

Yes, sockets and asyncore & friends work fine on the Mac.

(Currently no threading, but it looks like that's going to change real soon.)

Just





From matthewm at es.co.nz  Wed Jan 26 00:23:14 2000
From: matthewm at es.co.nz (Matthew Miller)
Date: Wed, 26 Jan 2000 18:23:14 +1300
Subject: Module to access file attributes for Windows?
Message-ID: <388E84C1.703E3DEE@es.co.nz>

Hi everyone,

Is their a module to access file attributes like:- date
created/modified,etc for Windows? I've had a reaonable look around but
with no success. Thanks. matthew.



From duncan at rcp.co.uk  Fri Jan 28 09:08:29 2000
From: duncan at rcp.co.uk (Duncan Booth)
Date: 28 Jan 2000 14:08:29 GMT
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com> <MPG.12fa1256cc30caf0989682@or.news.verio.net>
Message-ID: <8EC991745duncanrcpcouk@194.238.50.13>

jasont at wellmed.com (Jason Taylor) wrote in 
<MPG.12fa1256cc30caf0989682 at or.news.verio.net>:

>I just installed this new DLL, and it's a night-and-day difference.  
>There's almost no wait the first time, and it's as fast as Perl the 
>second and subsequent times.  By switching back to the old DLL, the wait 
>reappears.
>
>Thanks guys again for your help!!!

I find it strange that you find this long delay, but others (myself 
included) find the standard DLL loads quickly every time (possibly a bit 
slower the first time). I can think of a few possible reasons for this.

Perhaps the python DLL loads other DLLs which take time to load, and which 
are preloaded into memory on my system, but not on yours. The python 
dependencies are KERNEL32, USER32, ADVAPI32, WSOCK32, WINMM, MSVCRT, most 
of which would be loaded on any booted windows system. WSOCK32 might be 
unloaded if you do not normally have a network connection, in which case 
running up a network application before running python should reduce the 
wait.

The other possibility that occurs to me might be a conflicting DLL already 
loaded in the system at the same address. This would mean that the python 
dll would have to be relocated every time it loads, however the delayed 
loading DLL shouldn't be faster in this case unless it has a different 
relocation base.


From phd at phd.russ.ru  Mon Jan  3 05:55:58 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Mon, 3 Jan 2000 10:55:58 +0000 (GMT)
Subject: setting time?
In-Reply-To: <84ptud$vfh$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.21.0001031055360.10225-100000@fep132.fep.ru>

On Mon, 3 Jan 2000 moonseeker at my-deja.com wrote:
> Noone an advice to set the system time?

   It is TOOOO system-dependent...

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From PClaerhout at CREO.BE  Thu Jan  6 12:07:52 2000
From: PClaerhout at CREO.BE (Pieter Claerhout)
Date: Thu, 6 Jan 2000 18:07:52 +0100 
Subject: py2bat (was: How do I make a Python .bat executable file?)
Message-ID: <2B1262E83448D211AE4B00A0C9D61B03BA7027@MSGEURO1>

Try making a .cmd or a .bat file (you can choose), and put the following
line at the top of this cmd file:

@python -x %~f0 %* & goto :EOF

Paste your python script after that, and this should work. However, there
are three things you have to keep in mind:

- You need to have Python installed
- Don't put such a script in a path where there are spaces in the path
- This doesn't work with Windows 95/98

I also attached a script that converts your .py file to the according .cmd file.

Kind regards,


Pieter

Pieter Claerhout - PClaerhout at creo.be
Response Center - Applications Support

Creo Europe - Excelsiorlaan 21 - 1930 Zaventem - Belgium
Tel: +32 (2) 711 14 00 - Fax: +32 (2) 720 96 71

 <<py2cmd.cmd>> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: py2cmd.cmd
Type: application/octet-stream
Size: 683 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000106/e50d35e6/attachment.obj>

From swepson at hks.com  Thu Jan 20 15:06:11 2000
From: swepson at hks.com (Chris Swepson)
Date: Thu, 20 Jan 2000 15:06:11 -0500
Subject: HTML Widget
Message-ID: <38876AB2.58FF3EA5@hks.com>

    I've been searching far and wide for an HTML widget which can be
used in Python/Motif.  I can't seem to find one anywhere.  It's needed
for a very important job I'm doing.  Please respond ASAP.

--
"Fight the collective, Microsoft can't win.  Bill Gates has claimed to many lives already.  Don't conform to the Microsoft side..."

                -General Aeon Wilm of the US Computer rebellion (year 2047)





From zippy at cs.brandeis.edu  Thu Jan 20 23:33:49 2000
From: zippy at cs.brandeis.edu (Patrick Tufts)
Date: Thu, 20 Jan 2000 20:33:49 -0800
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <3dzouke2v7.fsf@amarok.cnri.reston.va.us>
Message-ID: <200120002033497688%zippy@cs.brandeis.edu>

In article <3dzouke2v7.fsf at amarok.cnri.reston.va.us>, Andrew M.
Kuchling <akuchlin at mems-exchange.org> wrote:

> Bala <engbpa at liquidinformation.com> writes:
> >         Python Essential Reference (OTHER NEW RIDERS)
> >         by David M. Beazley, Guido Van Rossum
> > or
> >         Learning Python (Help for Programmers)
> >         by Mark Lutz, David Ascher, Frank Willison (Editor)
> 
> The Essential Reference is just what its title says: a reference
> guide.  You'd have to be *really* motivated to learn Python from it.


The Essential Reference is an excellent reference.  Get it anyway.  But
you won't learn the language from it -- it's too terse for that.

Oh, and don't bother with "Programming Python" (also by Mark Lutz). 
It's not a good for learning or reference.  I've read reviews that say
"Learning Python" is much better.

--Pat


From news at dorb.com  Mon Jan 17 19:15:48 2000
From: news at dorb.com (Darrell)
Date: Mon, 17 Jan 2000 19:15:48 -0500
Subject: PYTHONPATH newbie question
References: <00091c0e.fbb4d1d4@usw-ex0102-012.remarq.com>
Message-ID: <beOg4.6093$U5.232421@typhoon.nyroc.rr.com>

3.29 site -- Site-specific configuration hook

Check out the file site.pth each line is a path that's added on startup.
--
--Darrell
"johnM"  wrote:
> How does one make changes to PYTHONPATH permanent? I can import module
> sys and call sys.path.append("dir") no problem. I can call sys.path and
> see that dir has been added.





From tismer at tismer.com  Mon Jan 24 11:26:50 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 24 Jan 2000 11:26:50 -0500
Subject: Stackless 1.01 Homepage
References: <388BE68E.B26985CA@tismer.com> <$WXiIAAO+Dj4Ew2W@jessikat.demon.co.uk>
Message-ID: <388C7D4A.D873AA@tismer.com>


Robin Becker wrote:
> 
> In article <388BE68E.B26985CA at tismer.com>, Christian Tismer <tismer at tismer.com> writes
> 
> I'm very impressed. here are the pybench timing differences (absolute and expressed as % of
> original python) for my 333Mhz win32 machine. here negative is bad for stackless, positive is
> good. My only disappointment is that Zope wouldn't start up.

Whow. I guess there is a bug in exec. In any case, it cannot be a
conceptual but but just a failure of mine. Idle, PyWin, pystone,
Metakit, they all work fine.

Will look into it after the conference.
Thanks for giving the measures.

These optimizations were done last January partially, before
I started to think of SLP. They didn't make it into the core,
like SLP didn't as well. As a consequence I bundled them and pushed
optimization even further, to outperform Standard Python.
If these patches would be applied to the original, I expect about
15 % speed gain, because SLP actually looses about 8%.

But please don't apply them :-))

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From mitch.chapman at mciworld.com  Sun Jan  2 11:51:27 2000
From: mitch.chapman at mciworld.com (Mitch Chapman)
Date: Sun, 02 Jan 2000 16:51:27 GMT
Subject: Python books review ?
References: <slrn86kqi1.h99.hirend@c1020616-a.frmt1.sfba.home.com>
Message-ID: <386F82E8.CD1F283E@mciworld.com>

Hirendra Hindocha wrote:
> 
> Hi,
> 
> It seems that suddenly there are a lot of Python books on the market.
> Has anyone bought them ? If so, can u post a review of them ?
> 
> I'm looking for reviews on -
> 
> Programming With Python
> Tim Altom; Paperback

I did the tech editing for this book, so can give you an unbiased
review :)

"Programming with Python" is a cookbook, rather than a language
reference or tutorial.  Other titles would be better for full-time
software developers just getting started with Python.  Tim targeted
this book at sysadmins looking for simple, concrete code examples.

The book includes 30+ sample scripts from members of the Python
community, each w. accompanying commentary.  Most of the scripts are
short, but they show how well Python can exploit the platform on which
it runs, with a particular emphasis on Linux.

The scripts run the gamut from games to multi-threaded Tkinter apps to
system utilities.  My personal favorite is a script by Sam Rushing
which identifies the process owning a socket.

The good news is that Tim has a clear, entertaining writing style.
The bad news it that he had me for his tech editor.  When my
complementary copy arrived I dug in to see how many errors made it all
the way to press.  Unfortunately, I was able to find several.

Three types of errors were evident: simple typos, indentation errors
resulting from hard tabs in the source code, and bone-headed technical
mistakes.  Luckily I've found only two of the last -- so far.

I've asked the publisher for permission to set up an errata web site
for the book, but have not yet received a response.

The indentation errors arose in part from Tim's insistence on
respecting the work of the original script authors.  He was concerned
about making "cosmetic" changes which might inadvertently alter the
behavior of a script, then attributing the result to an unsuspecting
script author.  But if I'd had the mental presence to suggest
replacing hard tabs w. spaces, Tim would probably have agreed.

One of the technical errors noted above involved a script provided by
Joseph Strout.  It appeared in the book as 'logic.py'.  Joe's original
script was fine, but the tech-review copy which arrived in email was
severely mangled, with some source lines run together, some
incorrectly indented, etc.  Thanks to my inability to explain this
simple fact, both the mangled script and a copy of Joe's original
version appear in the book, the latter "courtesy of Mitch Chapman."
I'm still looking for a rock to crawl under.

Well, so much for the sales pitch :)  If you're looking for Python code
samples for part-time programmers, have a look at "Programming with
Python" (and probably also "Python Annotated Archives" -- dunno,
haven't seen it yet).  If you're looking for an intro text for
full-time developers, something like "Learning Python" or "Python
Essential Reference" would be more appropriate.

-- 
Mitch Chapman
mitch.chapman at mciworld.com


From Alex.Martelli at think3.com  Fri Jan 14 05:30:39 2000
From: Alex.Martelli at think3.com (Alex Martelli)
Date: Fri, 14 Jan 2000 11:30:39 +0100
Subject: best Template module
Message-ID: <6D8A17398E28D3119F860090274DD7DB4B3DD1@pces.cadlab.it>

Otis Gospodnetic writes:

> Could anyone recommend a good 'Template module'?
> 
> Template module - a module that will let me create a plan text/HTML
> file with some special markup/tags that I can then replace by some text
> values from within my Python code
> 
> good - fast execution and not too complex :)
> 
> 
Shorn of the 'statement-level' functionality (conditionals and loops),
this is what my smartcopy.py boils down to:

import sys

class copier:
    "Smart-copier class"
    def __init__(self, regex, dict, ouf=sys.stdout):
        "Initialize self's data fields"
        self.regex = regex
        self.globals = dict
        self.ouf = ouf
    def copy(self, inf=sys.stdin, block=None):
        "Copy a file, or any block of lines, expanding expressions"
        if not block: block = inf.readlines()
        def repl(match,self=self):
            "return the eval of a found expression, for replacement"
            return '%s' % eval(match.group(1),self.globals)
        for line in block:
	      self.ouf.write(self.regex.sub(repl,line))


I don't think it gets much simpler than this:-).  Typical use:

import smartcopy
import re

# we choose to bracket embedded expressions between '@...@'
rexp=re.compile('@([^@]+)@')

# define the variables on which the bracketed expressions work
dict={ 'x': 2.3, 'y': 4.5, 'z': 6.7 }

# let output stream default to standard output
copy=smartcopy.copier(rexp, dict)

# stdin to stdout with embedded-expression expansion
copy.copy()


E.g. if standard input is

	@ x @ + @ y @ = @ x+y @

standard output becomes

	2.3 + 4.5 = 6.8


Of course, this could be factored very differently depending
on needs -- I needed arbitrary markers for embedded
expressions to make sure of no conflict with the syntax
for the files being transformed, and for my specific needs
I wanted to fix such markers, and the dictionary for the
expressions' evaluation, but then have the possibility of
processing several input files (or just blocks of lines) onto
a single output stream, in sequence.


I'm sure much more professional and general template
facilities exist, but since the task is so simple in Python
it may still be wortwhile having a small, simple module to
factor and tune to specific needs.

(My full 'smartcopy' module adds embedded Python
_statements_, basically to allow conditionals and loops,
but that does make it a tad more complex:-).


Alex




From robin at jessikat.demon.co.uk  Sun Jan  2 07:03:55 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 2 Jan 2000 12:03:55 +0000
Subject: fun with stock quotes
References: <Fnn296.A3w@world.std.com> <ncREqFA3nZb4Ew1G@jessikat.demon.co.uk>
 <Fnowpn.Iun@world.std.com>
Message-ID: <r3MbaHAr6zb4Ew2G@jessikat.demon.co.uk>

In article <Fnowpn.Iun at world.std.com>, Will Ware <wware at world.std.com>
writes
>Robin Becker (robin at jessikat.demon.co.uk) wrote:
>> here's my take on that which just returns a class with all the values
>
>I had originally done that, and that's why I have something
>erroneous in the comment about it being a class rather than a C
>module. Once I started playing with the data, I found I was doing
>things that ran too slow (the bottleneck seeming to be the linear
>regression operation) so I started coding stuff in C.
I do regression using NumPy; it works fine for me. I guess if it became
really essential to go to C one should build a proper regression module.

My main problem with these public data sources is that they're not
always available. Here in the UK yahoo seems to have made quotes.csv
produce xls files; an insanity of course for us, but sensible for yahoo.


-- 
Robin Becker


From quinn at drachma.ugcs.caltech.edu  Mon Jan 17 20:04:40 2000
From: quinn at drachma.ugcs.caltech.edu (Quinn Dunkan)
Date: 18 Jan 2000 01:04:40 GMT
Subject: Best way to work in debug code?
References: <85umq4$tlr$1@nnrp1.deja.com>
Message-ID: <slrn887f3r.viv.quinn@drachma.ugcs.caltech.edu>

On Mon, 17 Jan 2000 09:19:08 GMT, rdudfield at my-deja.com
<rdudfield at my-deja.com> wrote:
>Hello,
>
>
>Just wondering how you people write debug code in your functions
>/methods?
>
>I personally write a "if (DEBUG_LEVEL >= 5):" around the code.  However
>this reduces readability.

I have a dprint(msg) function that writes msg to stderr if debugging is set,
or to a file if debug_log is set, and occaisionally use assertions.  I've
never found it useful to get more complicated than that.  The only time I've
found it useful is for long-running processes that try to do complicated
stuff.

If you put code in if __DEBUG__: (or assert expr) then python -O will
"optimize" it away.  If you don't want to see debugging stuff in code, perhaps
a folding editor is the way to go (vim will do it eventually).

And then, if your code needs that much debugging, perhaps simplifying,
componentizing, and writing test harnesses is a better solution.


From dcinege at psychosis.com  Mon Jan 17 13:12:31 2000
From: dcinege at psychosis.com (Dave Cinege)
Date: Mon, 17 Jan 2000 13:12:31 -0500
Subject: Python and MS SQL (Microsoft SQL server)
Message-ID: <38835B8F.CA022D69@psychosis.com>

Is there a python module available to access MS SQL (Microsoft SQL server)
My client really wants to use this over mySQL. (Which I know there is a module
for)

Dave



From aahz at netcom.com  Thu Jan 13 16:10:33 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 13 Jan 2000 21:10:33 GMT
Subject: class methods?
References: <20000112145448.29699.qmail@web804.mail.yahoo.com> <85l5ul$bea$1@nnrp1.deja.com>
Message-ID: <85lf09$brl$1@nntp5.atl.mindspring.net>

In article <85l5ul$bea$1 at nnrp1.deja.com>,  <zakons at objectsbydesign.com> wrote:
>
>After looking at the alternatives, it is clear that it is much cleaner
>to just create an instance to print out the statistics, as in the
>following:
>
>   [....]
>
># to print out the statistics
>Company().print_stats()

Huh.  That's interesting.  I wonder if there's any significant
performance problem with this?
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From moonseeker at my-deja.com  Mon Jan  3 05:33:49 2000
From: moonseeker at my-deja.com (moonseeker at my-deja.com)
Date: Mon, 03 Jan 2000 10:33:49 GMT
Subject: setting time?
References: <84fuk4$n83$1@nnrp1.deja.com>
Message-ID: <84ptud$vfh$1@nnrp1.deja.com>

Hi again!

Noone an advice to set the system time?

Mike


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


From bitbucket at isomedia.com  Thu Jan  6 02:31:05 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Thu, 06 Jan 2000 07:31:05 GMT
Subject: python with ASP
References: <003c01bf5826$23f7b0e0$db75fea9@desktop>
Message-ID: <38743e91.376317586@news.isomedia.com>

On Thu, 6 Jan 2000 01:12:20 -0800, "Sameer Chowdhury"
<sameerc1 at yahoo.com> wrote:
>when I run it using PWS under win98, I get the following error:
>
>Python ActiveX Scripting Engine error '80020009'=20
>
>invalid syntax=20
>
>/pytest.asp, line 3=20
>
>> <%@ LANGUAGE =3D Python %>
>> <%
>> if REQUEST.Form("title"):
>> %>
>>  <html>
>>  <b> Got Title</b>
>>  </html>
>> <% else: %>
>>  <html>

I've seen this.  HTML between code sections appears to break up the
formatting of the code as far as the interpretter is concerned (and
thus, the syntax error for the lack of statement after the 'if',
followed by another for the 'else' without a matching 'if').  Also,
after you overcome the formatting problem, you'll have to capitalize
the Request object normally (at least on IIS 4 under NT 4, you have
to).  Finally, 'Request.Form ("title")' on my machine _always_ returns
something - an instance that acts somewhat like a list.  To test if
Request.Form ("title") really got any data, I recommend you do len ()
on it, and then subscript it to get at the values if the length
indicates there are any. :)

So this code does what you want.

<%@ LANGUAGE = Python %>
<%
if len (Request.Form("title")):
	sStr = '<b> Got Title</b>'
else:
	sStr = '''<FORM ACTION="./pytest.asp" method="POST">

<b>Title:&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp&nbsp</b><BR>
		<input type="text" value="" size=60 name="title"><P>
		<input type=submit value="Submit">
		</form>'''
%>
<html>
<%= sStr %>
</html>

    -Eugene


import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From gerrit.holl at pobox.com  Fri Jan  7 08:40:58 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 7 Jan 2000 14:40:58 +0100
Subject: how to include a file -- not import??
In-Reply-To: <3874F5DA.B73A6039@starmedia.net>; from andrew@starmedia.net on Thu, Jan 06, 2000 at 03:06:50PM -0500
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com> <3874F5DA.B73A6039@starmedia.net>
Message-ID: <20000107144058.A3759@stopcontact.palga.uucp>

Andrew Csillag wrote on 947167610:
> exec open('filename').read() 

execfile('filename')

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----
If you think Linux should get more hardware drivers, _please_ sign this:
http://www.libranet.com/petition.html



From dworkin at ccs.neu.edu  Wed Jan 19 23:10:42 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 19 Jan 2000 23:10:42 -0500
Subject: Pardon my ignorance, data conversion
In-Reply-To: "Michael P. Reilly"'s message of "Wed, 19 Jan 2000 01:14:43 GMT"
References: <38829357.ED633932@home.com> <85u6cs$85d$1@nntp9.atl.mindspring.net> <oqpuv04ixm.fsf@titan.progiciels-bpi.ca> <7e8h4.3861$Ef6.1020706@news.shore.net>
Message-ID: <vndzou11iyl.fsf@betelgeuse.ccs.neu.edu>

"Michael P. Reilly" <arcege at shore.net> writes:

> Floor does funny things to negative numbers.

> >>> int(floor(-3.4))
> -4
> >>> int(floor(-4.5))
> -5

In both cases, int(floor(x)) is giving you the greatest integer that
is not greater than x.  How is this funny?

-Justin

 



From dworkin at ccs.neu.edu  Tue Jan  4 13:05:58 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 04 Jan 2000 13:05:58 -0500
Subject: exception arg
In-Reply-To: mn@altern.org's message of "Tue, 04 Jan 2000 17:38:59 GMT"
References: <slrn8749le.d0.mn@banger.iol.it>
Message-ID: <vndpuvhsoe1.fsf@betelgeuse.ccs.neu.edu>

mn at altern.org (Mirko Nasato) writes:

> >>> 1/0
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ZeroDivisionError: integer division or modulo
> >>> try:
> ...     1/0
> ... except ZeroDivisionError, arg:
> ...     if arg == "integer division or modulo":
> ...             print "This is what I expect."
> ...     else:
> ...             print 'Unexpected. "arg" is "%s".' % arg
> ... 
> Unexpected. "arg" is "integer division or modulo".
> >>>

In older versions of Python, exceptions were string objects.  As of
Python1.5, they are generally class objects.

Perhaps this will make things a little clearer:

>>> try:
...   1/0
... except ZeroDivisionError, arg:
...   if str(arg) == "integer division or modulo":
...     print "This is what I expect."
... 
This is what I expect.
>>> try:
...   1/0
... except ZeroDivisionError, arg:
...   x = arg
... 
>>> x
<exceptions.ZeroDivisionError instance at 80dff60>
>>> str(x)
'integer division or modulo'

>From the standard documentation:

Exception 
       The root class for exceptions. All built-in exceptions are
       derived from this class. All user-defined exceptions should also
       be derived from this class, but this is not (yet) enforced. The 
       str() function, when applied to an instance of this class (or
       most derived classes) returns the string value of the argument
       or arguments, or an empty string if no arguments were
       given to the constructor. When used as a sequence, this
       accesses the arguments given to the constructor (handy for
       backward compatibility with old code). The arguments are also
       available on the instance's args attribute, as a tuple. 


-Justin

 



From tim_one at email.msn.com  Tue Jan 25 02:34:20 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 25 Jan 2000 02:34:20 -0500
Subject: OFFTOPIC: *-ly y'rs, 
In-Reply-To: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com>
Message-ID: <000201bf6706$987a9ba0$632d153f@tim>

[Agent Drek]

Cool handle!

> I know this is a stupid question but at least I'm not
> complaining about delimiters :) here goes ...
>
> I have only ever seen these sign-offs on this list:
>
> *-ly y'rs
>
> Where did they come from

Classified records at the Center for Disease Control identify y'rs truly as
the originating vector.  It's merely Fun with Adverbial Phrases -- which is
what all but the paranoid recognize from the start <wink>.

It's really a mongrelization, part related to (but easier than) a Tom Swifty
("I don't think I'll get another lion as a pet," Tom remarked offhandedly;
"The washing machine is going wild!" exclaimed Tom agitatedly; ...); part
gentle mocking of a former colleague's penchant for archaic expression ("y'r
most obdn't and humble servant ..."); and part made-up crap for the hell of
it <wink>.

> and are they patented :)

Luckily for you, I've been indulging this habit for about 20 years, and my
patent recently expired.  Have at it!

> How can one join the signoff-ly club?
>
> these-are-super-fun-ly y'rs,

with-the-novice-we-applaud-effort-not-achievement<wink>-ly y'rs
    - tim





From Christian.Schaller at mchp.siemens.de  Mon Jan 17 08:40:53 2000
From: Christian.Schaller at mchp.siemens.de (Christian Schaller)
Date: Mon, 17 Jan 2000 14:40:53 +0100
Subject: Python -> EXE?
Message-ID: <38831BE5.76B189A8@mchp.siemens.de>

Hello...

  Is there a tool similar to Perl2Exe that transforms a Python program
to a MS .EXE programc that can be executed without installing Python on
a machine?

  Thanks in advance!

bye
  Chris...


From mhammond at my-deja.com  Wed Jan 19 13:07:13 2000
From: mhammond at my-deja.com (mhammond at my-deja.com)
Date: Wed, 19 Jan 2000 18:07:13 GMT
Subject: PythonWin and Dialogs in DLLs
References: <38839d87$0$22473@senator-bedfellow.mit.edu> <388418ec.8757803@news.ricochet.net> <k5_g4.3484$iy5.225935@typ12.deja.bcandid.com> <3884a59a$0$22486@senator-bedfellow.mit.edu>
Message-ID: <864ug2$fln$1@nnrp1.deja.com>

In article <3884a59a$0$22486 at senator-bedfellow.mit.edu>,
  "Robert" <soundhack at hotmail.com> wrote:

> now one more question. I have gotten the subclassed dialog to show
> up, but am having problems with assigning message handlers to buttons.
> Following one of the demos, I did something like the following
>
> (in the dialog's InitDialog method)
>     self.button=self.GetDlgItem(IDC_CMD_BUTTON)
>     self.button.HookNotify(self.OnButton,win32con.BN_CLICKED)
>
> When i click on the button, the OnButton handler isnt getting called.
> I tried self.button.HookCommand() too, with no results. The one thing
> that sorta worked was
> self.button.HookMessage(self.OnButton,win32con.WM_LBUTTONDOWN), which
is not
> exactly what
> I wanted (sometimes the button click animation isnt shown)

Notification messages are handled by the parent, not the object - ie,
you simply do "self.HookMessage" - hook it from the dialog, not the
parent.

If you grep the Pythonwin sources you should find a number of
HookNotify examples - it is used all over the place...

Mark.


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


From tim_one at email.msn.com  Sat Jan  8 17:28:06 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 8 Jan 2000 17:28:06 -0500
Subject: RegExp, Python and strings
In-Reply-To: <857hhb$3kmi$1@www.univie.ac.at>
Message-ID: <000301bf5a27$a28f7c00$242d153f@tim>

[Matthias Huening]
> I'm new to PYTHON and I am trying to find out if PYTHON is the
> right language for work with strings (which is what I need most).

In my experience, the best languages for working with strings are SNOBOL4
and Icon (http://www.cs.arizona.edu/icon/).  They're outstanding for
strings, but not so hot for many other things.  Python is not outstanding
for strings, but pretty darn good at darned near everything (incl. strings).

> Therefore I am playing around with the RE-module.
>
> In PERL I can do things like this (in one line):
> $A = "Rossum, Guido van; Harms, Daryl; Python, Franz-Josef";
> $A =~ s/([A-Z])[\w]+(?![ \w\-]+,)([ ;-]|$)/$1.$2/g;
> This RegExp results in: "Rossum, G. van; Harms, D.; Python, F.-J."

As Andrew showed, there's an almost direct translation of this into Python.
If fitting things "in one line" is an important goal, you'll be happier with
Perl; Python doesn't strive to be a one-liner kind of language.

String manipulation is also (in general) quicker using Perl, although that's
subject to change in future releases (in particular, relative to Python
1.5.2, Perl has special speed advantages in running the same regexp
thousands of times over relatively short (say, < 80 chars) strings).

Regexps in Python are also "just another module"; they're not integrated
deeply into the language as they are in Perl.  It's fair to say that string
manipulation in Python gets more effective the more you get away from trying
to "write Perl in Python".  Python's non-regexp string module functions
(like string.find and string.split) are too often overlooked by new
immigrants.

> Now I am trying to rebuild this in PYTHON, but I can't get it
> to work.  Any hints? Should I keep trying? Or should I stick
> to PERL for those kinds of string-manipulations?

If you're thoroughly happy with Perl, there's no need to switch -- but if
you were I doubt you'd be asking the question <wink>.  Maybe you could say
more about what you're trying to get *away* from, and we can try to guess
whether you'll be happier with Python on those bases.

While you *can* do the regexp substitution above very much the same way you
do it in Perl, a "native Pythoneer" is likely to think about ways to break
it into smaller problems (e.g., for staters using string.split to break the
input on semicolons).  In this specific problem that's mostly a matter of
taste, but Python *encourages* decomposition in ways Perl does not (e.g.,
there are no embedded assignments, and in various other ways Python erects
barriers against trying to "do too much" in one line -- you can fight that
with some success, but the result will be strained).

if-you-"think-in-perl"-perl-is-the-right-language-ly y'rs  - tim





From sameerc1 at yahoo.com  Mon Jan 17 00:44:45 2000
From: sameerc1 at yahoo.com (Sameer Chowdhury)
Date: Sun, 16 Jan 2000 21:44:45 -0800
Subject: python and ASP
Message-ID: <002501bf60ae$1773bc20$b674fea9@desktop>

How do Sessions work when using python with ASP.   I, for example, have
working Python based ASP code, I add the following line: Session("userID") =
userID
after defining userID.  After I add this line, I get the following error
when I test it.
Python ActiveX Scripting Engine error '80020009'

Unknown

?

Unknown

Can anyone suggest a remedy.  Thanks in advance.


__________________________________________
NetZero - Defenders of the Free World
Get your FREE Internet Access and Email at
http://www.netzero.net/download/index.html



From niels at endea.demon.nl  Thu Jan  6 09:54:26 2000
From: niels at endea.demon.nl (Niels Diepeveen)
Date: Thu, 06 Jan 2000 15:54:26 +0100
Subject: Changes to UserList.py
References: <20000105183309.1485.qmail@web3006.mail.yahoo.com>
Message-ID: <3874ACA2.5DBC9A0@endea.demon.nl>


Jakob Schiotz schreef:
> I feel that the optimal solution would be a change in UserList.py, but
> I cannot figure out how the optimal solution should be.  

This looks like the best solution to me:
    def __init__(self, initial=None):
        if initial is None:
            self.data = []
        elif isinstance(initial, UserList): 
            # avoid going through __getitem__
            self.data = list(initial.data)
        else:
            self.data = list(initial)

> Perhaps
> UserList.__init() should be changed to be similar to __cmp__, and then
> the build-in function list() should be extended to work on any object
> having a __getitem__ method.  The latter change should probably be done
> anyway in a later version of Python.

It has already been done, in 1.5.2.

-- 
Niels Diepeveen
Endea automatisering



From bitbucket at isomedia.com  Fri Jan  7 17:18:06 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Fri, 07 Jan 2000 22:18:06 GMT
Subject: python with ASP
References: <020e01bf5970$041d46e0$556afea9@desktop>
Message-ID: <38766552.517310182@news.isomedia.com>

I agree that not being able to embed the HTML inside indentation but
outside of the code tags is awkward at first, but it's a tradeoff.
VBScript doesn't have true multiline strings.  Nor does it have
built-in dictionaries.  Or classes.  Or exceptions.  Or the ability to
create an array and fill it with a single line of code.

So in the meantime, I prefer to use Python and just be creative about
working around this limitation.  And in the end, a multi-line string
really isn't that different from the HTML that's outside of the code
blocks.  They're both text.

      -Eugene

On Fri, 7 Jan 2000 16:33:40 -0800, "Sameer Chowdhury"
<sameerc1 at yahoo.com> wrote:

>>When you copy-and-pasted my code, it appears you lost the tabbing.
>>I've attached the code in case your newsreader isn't displaying the
>>tabs happily.
>>    -Eugene
>
>Yes, it seems to work now.  But restricting to one line of code, or a
>multiline string is kind of limitting.  This kind of limitation doesn't
>exist with JScript, or VBScript.  I hope this will be fixed in the future.
>
>__________________________________________
>NetZero - Defenders of the Free World
>Get your FREE Internet Access and Email at
>http://www.netzero.net/download/index.html
>

import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From nobody at nowhere.nohow  Sun Jan 23 19:49:38 2000
From: nobody at nowhere.nohow (Grant Edwards)
Date: Mon, 24 Jan 2000 00:49:38 GMT
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com> <DAJ5pFA0awi4EwIe@gol.com> <86f4of$k5b$1@thoth.cts.com>
Message-ID: <slrn88n8d2.cbl.nobody@isis.visi.com>

In article <86f4of$k5b$1 at thoth.cts.com>, Will Rose wrote:

>Ditto.  Python has its faults (cries of no!, no!) but it's better than
>all the alternatives.

To paraphrase Michael Elkins:

     "All programming languages suck. This one just sucks less."

-- 
Grant Edwards                   grante             Yow!  TONY RANDALL! Is YOUR
                                  at               life a PATIO of FUN??
                               visi.com            


From gmcm at hypernet.com  Thu Jan  6 19:32:52 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 6 Jan 2000 19:32:52 -0500
Subject: newbie embedding questions
In-Reply-To: <cs9d4.42068$Ke.26436@news21.bellglobal.com>
Message-ID: <1264917322-7279010@hypernet.com>

Jason Maskell wrote:

[Will Ware recommends extending with SWIG]

> Actually, I was looking at that this afternoon, and right now.
> Unfortunately, there don't seem to be any binaries available, and
> compiling it in BCB is looking to be a huge chore. I suppose I
> could install VC again, but that seems like a big hassle just to
> compile this tool and uninstall it.

The real point is that embedding and extending are not 
mutually exclusive. In your situation, doing both makes perfect 
sense. If you follow the normal extending procedure, you'll 
have a .dll. Establishing direct communication between the .dll 
and your embedding app (if you need it) may involve a bit of 
trickery; you should probably let Python load the .dll and use 
CObjects to pass a C pointer through Python back to your 
embedding app.

If you want them all in one, you could look at the inittab stuff, 
which would let the embedding app add a new builtin to 
Python.

You've got a ton of options. If it starts to get too complex, look 
around for another way to try it.

- Gordon



From alex at somewhere.round.here  Sat Jan  8 22:53:41 2000
From: alex at somewhere.round.here (Alex)
Date: 08 Jan 2000 22:53:41 -0500
Subject: fastest way to merge lists.
References: <roy-D565B0.21255908012000@netnews.nyu.edu>
Message-ID: <etdbt6vgat6.fsf@w20-575-109.mit.edu>

If speed is a big issue for you, might consider using Aaron Watters'
kjbuckets extension.  It provides a set type, and is supposed to be much
faster than using dictionaries.  (I believe the claim, but I've never
tested it.  I use it mainly so I don't have to write d[k]=None all the
time.)  I use it all the time for just the sort of thing you describe.

See
http://www.pythonpros.com/arw/kjbuckets/

Also, if you don't go that route, there may be some mileage in using
operator.setitem.  I suppose I should time it myself, since I'm
suggesting it, but I can't be bothered.  Something like

import operator
d={}
l=list1+list2+list3+list4+list5 
map(operator.setitem,len(l)*[d],l,len(l)*[1])

...might be worth looking into.

Oh, and another thing you could do is sort the whole lot, and run over
them to pick out the first instance of each string:

l=list1+list2+list3+list4+list5 
l.sort()
last=None
merged=[]
for element in l:
  if element!=last:
    merged.append(element)
    last=element

Alex.


From wjk at wjk.mv.com  Sat Jan 22 02:23:02 2000
From: wjk at wjk.mv.com (Wm. King)
Date: Sat, 22 Jan 2000 02:23:02 -0500
Subject: A modest proposal
References: <865bmq$dum$1@news.ycc.yale.edu> <FonA21.B26@world.std.com>
Message-ID: <38895AD5.9F5A3108@wjk.mv.com>

A compromise is in order.
I think centered would be cool.

Will Ware wrote:

> Paul M (paul.magwene at yale.edu) wrote:
> : Dear Pythoneers...
>
> A great evil has been perpetuated in the design of the Python
> language. It grieves me to be the one to bring this to the world's
> attention but it must not continue to go unacknowledged. Persons of
> right indentation have once again been marginalized and excluded
> because _every_editor_ever_written_ indents from the LEFT SIDE OF THE
> PAGE. The brutal, systematic, uniform, codified, societally
> legitimized vicious assault on right indentation has gone unnoticed,
> despite my many pleas to Greenpeace and Amnesty International. I am
> forced to bring this issue to the very feet of the Python community
> itself, and beg, nay, demand, that the humanity of right-indenters be
> recognized and respected. Let the bloodshed end here.
>
> Right-indenters are not a vindictive lot. We do not insist that the
> last fifty years of oppression be compensated overnight. We do not
> require that CPython's left indentation be removed and replaced with
> right indentation. We are a peaceful people, eager for understanding
> and amicable coexistence. In the near term, we can get by with the
> following script, which translates left indentation to right
> indentation.
>
>                             # to-right.py
>                    import sys, string, re
>
>               before = re.compile('^\ +')
>
>                                maxlen = 0
>                               lines = [ ]
>           for L in sys.stdin.readlines():
>                            L = L[:-1]
>                   if len(L) > maxlen:
>                   maxlen = len(L)
>                       lines.append(L)
>
>                           for L in lines:
>                  s = before.search(L)
>                         if s != None:
>                  n = s.regs[0][1]
>                         L = L[n:]
>                                 else:
>                             n = 0
> print (maxlen - n - len(L)) * ' ' + L
>
> As a temporary retrofit, I also offer the left-indented version:
>
> # to-right.py
> import sys, string, re
>
> before = re.compile('^\ +')
>
> maxlen = 0
> lines = [ ]
> for L in sys.stdin.readlines():
>     L = L[:-1]
>     if len(L) > maxlen:
>         maxlen = len(L)
>     lines.append(L)
>
> for L in lines:
>     s = before.search(L)
>     if s != None:
>         n = s.regs[0][1]
>         L = L[n:]
>     else:
>         n = 0
>     print (maxlen - n - len(L)) * ' ' + L
>
> Right-indenters are a forebearing bunch, and do not instantly
> expect a command-line option allowing for right indentation.
> We can be gracious enough to wait to see it in Python 2.0.
> --
>    People say that everyone has a few skeletons in their closet. Not me.
>    Well, not yet anyway. I mean, the bodies are still decomposing.
> Will Ware               email: wware[at]world[dot]std[dot]com



From tjg at avalongroup.net  Fri Jan  7 01:03:09 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Thu, 06 Jan 2000 22:03:09 -0800
Subject: First try at Tkinter stuff
Message-ID: <3875819D.41ECB8BA@exceptionalminds.com>

Hi,

I'm just starting to play with TK and am completely confused with
packing and gridding, etc.

I've got a window that displays four frames. It does it quite nicely and
they look very nice. However, when I add a label or a button, the frame
to which I add the label or button shrinks to the size of the button.
This is completely undesirable behaviour. I have tried using both .pack
and .grid, but I guess I don't understand how they are supposed to work.

Attached is my little test class.
class Brokerage:
    def __init__(self, master):
        MainFrame = Frame(master, height=600, width=800)
        MainFrame.pack()
        
        SupplierFrame = Frame(MainFrame, height=575, width=300,
                                relief=GROOVE, borderwidth=1)
        SupplierFrame.grid(row=0, column=0, rowspan=2)

        DataFrame = Frame(MainFrame, height=250, width=200,
                                relief=GROOVE, borderwidth=1)
        DataFrame.grid(row=0, column=1)

        ShippingFrame = Frame(MainFrame, height=250, width=200,
                                relief=GROOVE, borderwidth=1)
        ShippingFrame.grid(row=1, column=1)
        
        VendorFrame = Frame(MainFrame, height=575, width=300,
                                 relief=GROOVE, borderwidth=1)
        VendorFrame.grid(row=0, column=2, rowspan=2)

	# Code up to here works as expected, code below does not.

        Label(SupplierFrame, text="Vendor ID:").pack(side=TOP)

        self.hi_there = Button(VendorFrame, text="Hello",
command=self.say_hi)
        self.hi_there.pack(fill=BOTH, expand=1)
    
    def say_hi(self):
        print "hi there, everyone!"

Thanks for any assistance you might be able to provide.

-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From sholden at bellatlantic.net  Wed Jan  5 10:27:03 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Wed, 05 Jan 2000 15:27:03 GMT
Subject: Py2K wishes
References: <38675B72.18A139FF@prescod.net> 	<87ogbcuym5.fsf@den.home.net> <14440.14613.951856.73684@dolphin.mojam.com> <87902fvbm5.fsf@den.home.net>
Message-ID: <387362DD.9FDA3880@bellatlantic.net>


Frank Sergeant wrote:
> 
> Skip Montanaro <skip at mojam.com> writes:
> 
> >     Frank> Just as clearly, 'def' is an abbreviation for the noun 'define'.
> >
> > Was there a missing smiley?  Last time I looked, "define" was a verb...
> 
> No, no, just a missing brain.  I meant to write "definition".
> 
>   -- Frank

Frank, you are being way too modest.  For myself, I thought
you were trying to emulate Humpty Dumpty, from "Through the
Looking Glass":

    `When _I_ use a word,' Humpty Dumpty said in rather a scornful
  tone, `it means just what I choose it to mean--neither more nor
  less.'

    `The question is,' said Alice, `whether you CAN make words mean
  so many different things.'

    `The question is,' said Humpty Dumpty, `which is to be master--
  that's all.'

I guess, for Python, the answer to the last question is "GvR"!

regards
 Steve


From mwh21 at cam.ac.uk  Thu Jan 27 08:58:09 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 27 Jan 2000 13:58:09 +0000
Subject: Keywordargument in string.split
References: <13B8DD29BC@nds-3.azg.nl>
Message-ID: <m31z731ury.fsf@atrus.jesus.cam.ac.uk>

"P.J.W.S. Vrijlandt" <P.J.W.S.VRIJLANDT at INT.azg.nl> writes:

> I wrote:
> 
> > I have this error I don't quite understand.
> > Why can't I give 'sep' as a keyword argument?
> > 
> > It seems that the string module I import, is not the string.py in the 
> > standard distribution but a not-quite-equivalent c-implementation of 
> > it.
> 
> I got replies pointing to 'strop', which is a c-implementation of 
> string, and is imported by string.py
> 
> But what is the header (def) of split in strop? 
> (and why is it different)

Supporting keyword arguments for C extensions requires a little bit of
work on the part of the extension writer. Not much, but a bit. 

This work doesn't seem to have been done for many modules in the
standard distribution. This could be for two reasons:

1) The support for C extensions taking keyword parameters is quite
   new, and not much has caught up yet (no idea whether this is true)

2) The standard C extensions are so commonly used that people don't
   need to use kwargs (I'd never noticed the lack before).

3) PyArg_ParseTupleAndKeywords is significantly slower than
   PyArg_ParseTuple, even in the no kwarg case (I don't believe this
   is the case, but it's possible).

It seems that sha and parser accept keyword arguments, and nothing
else. Would patches to remedy this be accepted? Looks easy enough to
acheive, if a little tedious.

Cheers,
Michael


From helge.hess at mdlink.de  Thu Jan  6 13:38:16 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Thu, 06 Jan 2000 19:38:16 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <14452.56388.442949.799702@weyr.cnri.reston.va.us>
Message-ID: <3874E118.1C062D2C@mdlink.de>

"Fred L. Drake, Jr." wrote:
> Helge,
>   Yes:  It doesn't make sense. 

Well, IMHO it does make sense, see below.

> The job of the __init__() method is to
> initialize the instance which has already been created. 

Why should we have this distinction ? I find it inconsistent to declare
'__init__' to behave differently from other methods. *This* doesn't make
sense to me.

> There is no
> way for a Python program to receive the return value.

In the current interpreter not. That is what annoys me, since it is an
completly articial limitation I don't understand.
As mentioned it took only 5 minutes to 'correct' this limitation.

>   To implement the singleton pattern you appear to be interested in,
> try this:

[singleton using function]
>         def C():
>             global _C_singleton

This example covers only the most basic case and is a hack (eg how do I
do 'o instanceof C' ?). The idea behind returning a value different from
'self' or 'None' is to customize the class based on the parameters.

eg:
  class ImmutableDictionaryClass: # this is abstract
    def __init__(self, keys, values):
      if len(keys) == 0:
        # always return the same empty dict to save memory
        if emptyImmutableDict is None: 
          emptyImmutableDict = EmptyImmutableDict()
        return emptyImmutableDict
      elif len(keys) < 10:
        return SequentialDictionary(keys, values)
      else:
        return HashTableDictionary(keys, values)

This is called a 'class-cluster' in Objective-C and has quite some
advantages. Another useful application for clusters are string-objects
(different classes for different char widths, eg Py8bitString,
PyUnicodeString, ..)

The nice thing is that you only need a single (abstract) class to
interact with. Defining a function like you did above seems
unnecessarily hackish to me.

Notably removing the constraint in the interpreter doesn't affect
existing Python code ! This is what I wonder about. The interpreter was
apparently restricted intentionally in this regard and I wonder what
lead to this decision.

best regards
  Helge
-- 
SKYRIX-OS  Web Application Environment - http://www.skyrix.com
SKYRIX-IAS Groupware Application Suite - http://www.skyrix.com



From ht at cogsci.ed.ac.uk  Tue Jan 11 04:07:58 2000
From: ht at cogsci.ed.ac.uk (Henry S. Thompson)
Date: 11 Jan 2000 09:07:58 +0000
Subject: xmllib has fatal bug.
References: <slrn87kpsu.aq0.kc5tja@garnet.armored.net>
Message-ID: <f5b3ds5dlht.fsf@cogsci.ed.ac.uk>

kc5tja at garnet.armored.net (Samuel A. Falvo II) writes:

> 
> Now, suppose I feed it the following XML data:
> 
> <schema targetNS='http://purl.org/metadata/dublin_core'
> 	version='1.0'
> 	xmlns='http://www.w3.org/1999/XMLSchema">
> 	<element/>
> </schema>

Always check your input with an independent validator:  the above is
ill-formed, having unmatched quote characters in the default namespace 
declaration.

ht
-- 
  Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
     2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
	    Fax: (44) 131 650-4587, e-mail: ht at cogsci.ed.ac.uk
		     URL: http://www.ltg.ed.ac.uk/~ht/


From mwh21 at cam.ac.uk  Tue Jan 18 05:10:22 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 18 Jan 2000 10:10:22 +0000
Subject: Python code for documenting Python classes?
References: <01HW.B4A9003C005E83C70C98670C@news1.radix.net>
Message-ID: <m3g0vvadwx.fsf@atrus.jesus.cam.ac.uk>

T. Bridgman <wtbridgman at radix.net> writes:

> Just so I don't try to re-invent the wheel...
> 
> Does anyone know of a Python program for documenting Python classes?

No, but I've whipped something up. The script at the end of this post
scans through the methods of a class and collects a list of attributes
that are assigned to self.

E.g:

>>> class A:
	def __init__(self,a):
		self.b = a
		self.c = 1
		self.e = "orange"
	def meth2(a,b): # even works if you don't call it self!
		a.confuse = b

		
>>> classdoc.do_class(A)
['b', 'c', 'confuse', 'e']

It uses bytecodehacks, so get that here: (where I've just uploaded a
new version with a couple fixes)

http://starship.python.net/crew/mwh/

(it could be written just using dis, but not as quickly) 

It only does the class in question; it could be trivially modified to
recurse into base classes (but only Python ones, I'm afraid).

This is a fairly maximal set of data members; you could just scan
__init__ for a more minimal set. Trying to work out which members will
always be set is a job for someone else (maybe someone who can solve
the halting problem...).

It also makes no attempt to determine the type of member (that would
be *hard*), or format the output (this should be easy).

enough-caveats-for-one-post-ly y'rs
Michael

--- this is classdoc.py:

from bytecodehacks.code_editor import InstanceMethod
from bytecodehacks.ops import *
import types

def do_method(meth):
    code = InstanceMethod(meth).im_func.func_code
    ops = code.co_code
    varnames = code.co_varnames
    names = code.co_names

    ans = []
    
    for i in range(len(ops)):
        op = ops[i]
        if op.__class__ is LOAD_FAST:
            if op.arg == 0: # self is the first posarg
                if ops[i+1].__class__ is STORE_ATTR:
                    name = names[ops[i+1].arg] 
                    if name not in ans:
                        ans.append(name)

    return ans

def do_class(klass):
    ans = []
    for m in dir(klass):
        a = getattr(klass,m)
        if type(a) is types.MethodType:
            for i in do_method(a):
                if i not in ans:
                    ans.append( i )

    ans.sort()



From dcinege at psychosis.com  Tue Jan 18 14:59:07 2000
From: dcinege at psychosis.com (Dave Cinege)
Date: Tue, 18 Jan 2000 14:59:07 -0500
Subject: Python and MS SQL (Microsoft SQL server)
References: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in>
Message-ID: <3884C60B.BE820782@psychosis.com>

"George Thom at s" wrote:
> 
> Hi Dave,
> The odbc module that comes as part of the win32 extensions bundled with
> Python 1.5.2 for the various flavours (!) of Windows should do just fine.

So you're saying python has to be runnign on NT for this to work?

> Additional functionality (with some tweaking ... ymmv ;-) can be obtained
> with marc Lemburg's mxODBC module.

What about this? we're looking to run on solaris or linux.



From darrell at dorb.com  Sat Jan  8 13:42:08 2000
From: darrell at dorb.com (Darrell)
Date: Sat, 8 Jan 2000 13:42:08 -0500
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <011501bf58a4$7b0db540$0100a8c0@rochester.rr.com> <3877513B.73B762F4@maxtal.com.au>
Message-ID: <004901bf5a08$11a35be0$0100a8c0@rochester.rr.com>

From: "skaller" <skaller at maxtal.com.au>
> What would be 'generally' useful is a genuine switch 
> statement -- a sort of unrolled dictionary whose values
> are pieces of code. Syntax?
> 
> switch x:
> case 1: ..
> case 2: ...
> default: ...
> 
> like C, except that each handler has an implicit
> break at the end of the handler suite (no fall through,
> like in C).
> 
> Comments?

This is a common thing, does this express the idea ? 

def one(arg):
    print 'ARGV:', arg,
    return 'RetVal'

def two(arg):
    print 'ARGV:', arg,
    return 'RetVal'

states={1:one, 2:two, 3:3}

def switch(state, args, stateTable, default=None):
    val=stateTable.get(state,default)
    if callable(val):
        return val(args)
    return val

for x in range(5):
    print 'State:', x, `switch(x, "argVal", states)`

####### Output
State: 0 None
State: 1 ARGV: argVal 'RetVal'
State: 2 ARGV: argVal 'RetVal'
State: 3 3
State: 4 None

--Darrell




From eroubinc at u.washington.edu  Sat Jan  8 02:02:05 2000
From: eroubinc at u.washington.edu (Evgeny Roubinchtein)
Date: Sat, 08 Jan 2000 07:02:05 GMT
Subject: is variable a list?
References: <38729120.7616054D@simplypets.com>
Message-ID: <86hfgp6o7m.fsf@ruby.somewhere.washington.us>

>>>>> "James" == James Punteney <jamesp at simplypets.com> writes:


    James> Is there some way to determine whether a variable is list
    James> or a string?  Basically I have a variable that I would like
    James> to be able to be a string or a list, but I need some way to
    James> tell which it is.

Besides using the type() function you could also use the types module
and the isinstance() function:

>>> l=[]
>>> s=''
>>> import types
>>> isinstance(l, types.ListType)
1
>>> isinstance(l, types.StringType)
0
>>> isinstance(s, types.StringType)
1
>>> isinstance(s, types.ListType)
0

Hmm, I guess TIMTOWDY _sometimes_ applies to Python :-).

-- 
Evgeny.


From georgethomas at pspl.co.in  Tue Jan 25 04:13:36 2000
From: georgethomas at pspl.co.in (George Thom@s)
Date: Tue, 25 Jan 2000 14:43:36 +0530
Subject: Date Time problem in the Win32 extensions
Message-ID: <003501bf6714$767725b0$cf02a8c0@intranet.pspl.co.in>

Hi everyone,
I've managed to put together a rudimentary script that extracts all the
summary information from a storage object using the amazing support for COM
offered by Python. Part of the code was adapted from an example in the beta
chapter of 'Python Programming on Win32' by Mark Hammond & Andy Robinson. I
seem to have encountered a problem in PIDSI_EDITTIME(0x0000000a). Consider
the following code fragment.

import pythoncom
from win32com import storagecon # constants related to storage functions.

# These come from ObjIdl.h
FMTID_UserDefinedProperties = "{F29F85E0-4FF9-1068-AB91-08002B27B3D9}"

#------------- Lots of other COM stuff that can be obtained from the online
beta chapter -------------------

ps = pss.Open(FMTID_UserDefinedProperties)  # pss is a handle to the
IPropertySetStorage interface

props = PIDSI_TITLE, PIDSI_SUBJECT, PIDSI_AUTHOR, PIDSI_KEYWORDS,
PIDSI_COMMENTS, PIDSI_TEMPLATE, PIDSI_LASTAUTHOR, PIDSI_REVNUMBER,
PIDSI_EDITTIME, PIDSI_LASTPRINTED, PIDSI_CREATE_DTM, PIDSI_LASTSAVE_DTM,
PIDSI_PAGECOUNT, PIDSI_WORDCOUNT, PIDSI_CHARCOUNT, PIDSI_THUMBNAIL,
PIDSI_APPNAME, PIDSI_DOC_SECURITY

data = ps.ReadMultiple( props )

On retrieval of the PIDSI_EDITTIME property, a PyTime object is obtained.
It, however, gives me 1601 as the year and consequently the Format() method
returns meaningless information. Although I initially thought this was a Y2K
issue, I later noticed that the other PyTime fields (PIDSI_LASTPRINTED for
instance) continued to return compliant date/time information. I would
really appreciate it if someone could help and let me know if I'm missing
something obvious here.

regards,
george
--------------------------------------------------------------------------
George Thomas
Resident,Planet Earth,Third Rock from the Sun
georgethomas at pspl.co.in


Flat No. FI-5, Fifth Floor,
Himagirinath Co-op. Hsg. Society,
486A Elphinstone Road,
Kirkee, Pune - 411003,
Maharashtra State,
India.
--------------------------------------------------------------------------
Faced with the choice between changing one's mind and proving
there is no need to do so, almost everybody gets busy on the proof.
                                                                 - John
Kenneth Galbraith
--------------------------------------------------------------------------





From mikael at isy.liu.se  Thu Jan 20 11:07:54 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 20 Jan 2000 17:07:54 +0100 (MET)
Subject: A modest Error (I mean Proposal)
In-Reply-To: <uHFh4.1875$L4.54845@newsread2.prod.itd.earthlink.net>
Message-ID: <XFMail.000120170754.mikael@isy.liu.se>

On 20-Jan-00 Arinte wrote:
 >  
 >  "Mikael Olofsson" <mikael at isy.liu.se> wrote in message
 >  news:XFMail.000120142944.mikael at isy.liu.se...
 >  
 >  On 20-Jan-00 Arinte wrote:
 >   >  I am getting this weird error???
 >   >  Traceback (innermost last):
 >   >    File "Beer", line 17, in ?
 >   >    File "D:\POSSApps\anotherposs app\PossDevice.py", line 37, in IOCTL
 >   >      if type(command)==type(2):
 >   >  TypeError: len() of unsized object
 >  
 >  Maybe you have accidentally redefined type to len...
 >  
 >  I would entertain that thought, but this problem only occurs on the 2nd
 >  call to this function???

OK, so we've narrowed it down some. Maybe that's what happens between 
those two calls. 

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    17:03:57

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From herzog at online.de  Tue Jan 18 11:27:02 2000
From: herzog at online.de (Bernhard Herzog)
Date: 18 Jan 2000 17:27:02 +0100
Subject: Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us> <m12AMt3-00007rC@vortigen.demon.co.uk> <al8u2kbycss.fsf@sirppi.helsinki.fi>
Message-ID: <m3embfz6p5.fsf@greebo.nodomain.de>

Markus Stenberg <mstenber at cc.Helsinki.FI> writes:

> Edward Welbourne <eddy at vortigen.demon.co.uk> writes:
[snip]
> > Three new kinds of argument come into being (with my given spelling):
> >   * safe tunnel: after both * and **
> >   * keyword-only: after * but not **
> >   * positional-only: after ** but not *
> 
> Ugh.. *puke* this looks like ugly hack from hell designed to counter
> Python's original debatable deficiency in namescape/scoping department.
> All examples I've seen to date have just provided ways of countering
> current scoping, and I'd rather see Python's scope system changing
> (possibly something along the lines mr. Skaller of Viper fame proposed)
> than language getting more and more cryptic with very little gain to show
> for it. 

I'll second that. Turning a kludge that works because of an
implementation detail into an officially supported feature isn't a good
idea, IMO. Support for lexical scoping is the correct approach.

[snip]
> > You want to let your function see some value you can compute, but you
> > want it to be able to take arbitrarily many positional args; or you need
> > to be sure your caller can't mess with one of your args.

lexical scoping would take care of that, unless I'm missing something. 

[snip]
> def filename(*fragments, root=os.getcwd()): ?
> 
> It's not allowed right now, but looks pretty clear to me, and it's logical
> extension to current form.

IMO, this particular case -- keyword args after the * -- should be used
to declare arguments that have to be passed as keyword args, just like
Edward suggested.

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


From Rod.Haper at sage.nelesautomation.com  Fri Jan 21 17:30:20 2000
From: Rod.Haper at sage.nelesautomation.com (Rod Haper)
Date: Fri, 21 Jan 2000 16:30:20 -0600
Subject: Looking for Solaris 2.6/Python 1.5.2 make test fix
References: <86ajtj$8f8$1@srv38.cas.org>
Message-ID: <3888DDFC.38F7BBC@sage.nelesautomation.com>

I'm also getting a segmentation fault on Red Hat Linux 6.1 at the same
point in 'make test' but no core file is generated:

rm -f ./Lib/test/*.py[co]
PYTHONPATH= ./python ./Lib/test/regrtest.py 
make: *** [test] Segmentation fault
PYTHONPATH= ./python ./Lib/test/regrtest.py 
make: *** [test] Segmentation fault




lvirden at cas.org wrote:
> 
> When I configure then do a make all and then make test, Python core dumps:
> 
> rm -f /ldatae/Misc/python/Python-1.5.2/Lib/test/*.py[co]
> PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py
> Bus Error - core dumped
> make: [test] Error 138 (ignored)
> PYTHONPATH= ./python /ldatae/Misc/python/Python-1.5.2/Lib/test/regrtest.py
> Bus Error - core dumped
> make: *** [test] Error 138

-- 
+----------------------------------------------------------------------+
| Rod Haper, Senior System Analyst       |  Neles Automation           |
| Phone: 713-346-0652 - main             |  SCADA Solution, Inc.       |
|        713-939-0393 - fax              |  7000 Hollister             |
| Rod.Haper at sage.nelesautomation.com     |  Houston, Texas 77040, USA  |
+----------------------------------------------------------------------+


From peter.stoehr at sdm.de  Tue Jan 25 05:31:17 2000
From: peter.stoehr at sdm.de (Peter Stoehr)
Date: Tue, 25 Jan 2000 11:31:17 +0100
Subject: delete files?
References: <86jtm2$7cs$1@news1.sunrise.ch>
Message-ID: <388D7B75.9A7F85A7@sdm.de>

Hi Nathan

nathan schrieb:
> 
> Hello
> somebody knows how can I quickly delete files (without using os.system..) ?
> Working under WinNT, v.1.5.2.
> Thank you
> 
> nat
what's wrong with
	os.remove(...)

    Peter
-- 
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 tismer at tismer.com  Sat Jan 29 10:05:42 2000
From: tismer at tismer.com (Christian Tismer)
Date: Sat, 29 Jan 2000 10:05:42 -0500
Subject: concurrency in Python
References: <86utmv$1va$1@nnrp1.deja.com>
Message-ID: <389301C6.F07951D8@tismer.com>


gvwilson at my-deja.com wrote:
> 
> Hi.  I'm interested in hearing from anyone who's ever built a
> concurrency toolkit for Python that used a model other than
> threads + locks --- futures, tuple spaces, active expressions,
> concurrent aggregates, actors, or anything else.

I think you should have a look at Stackless Python that
comes with full-blown first class continuations.

http://www.tismer.com/research/stackless/

ciao - chris.stackless

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From sei at sunse.jinr.ru  Wed Jan 19 16:06:57 2000
From: sei at sunse.jinr.ru (Evguenii Smogailov)
Date: Thu, 20 Jan 2000 00:06:57 +0300 (MSK)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <slrn88c8i0.ja9.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>
Message-ID: <Pine.SOL.3.91.1000120000501.20920A-100000@sunse>

On 19 Jan 2000, Stephen Coursen wrote:
> When did this become a trait of Perl?  Gods, what's so hard about knowing
> that == is the equality operator?

Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.

(and you mean i need remember these '!=' and '<>' also :)

evguenii





From loewis at informatik.hu-berlin.de  Thu Jan  6 14:55:16 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 06 Jan 2000 20:55:16 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de>
 <14452.56388.442949.799702@weyr.cnri.reston.va.us>
 <3874E118.1C062D2C@mdlink.de>
Message-ID: <p6qwvpnotzv.fsf@informatik.hu-berlin.de>

Helge Hess <helge.hess at mdlink.de> writes:

> In the current interpreter not. That is what annoys me, since it is an
> completly articial limitation I don't understand.
> As mentioned it took only 5 minutes to 'correct' this limitation.

I assume you also took care of decrefing the 'garbage' object as well,
right?

Anyway, if your change is accepted into the language, you get a
serious incompatibility:

class A:
  def __init__(self, list = None):
    self.value = 0
    if list is None:
      return
    for item in list:
      self.value = self.value + item

Today, A() would give you a new object. With your change, A() gives
you None (because the return is implicitly 'return None').

What's worse, if you have

class A:
  def __init__(self):
    print "Here"

means that you cannot create instances of A at all, as A() will return
with an implicit 'return None'. That could be fixed by telling the
compiler to put 'return self' into the constructors by default, but
that juse makes __init__ different from the other functions, again.

This should explain why we can't change the language like that. That
does not explain why the language is the way it is: I guess that
__init__ was designed not really as a constructor (it does not create
the object), but as an initializer - just like C++ or Java, where you
can't return something different from the "constructor", either.

Regards,
Martin



From gklix at hassler.de  Mon Jan 24 12:54:41 2000
From: gklix at hassler.de (Gerald Klix)
Date: Mon, 24 Jan 2000 18:54:41 +0100
Subject: Stackless Python 1.0 + Continuations 0.6
References: <mt2.0-27365-948450827@news.informatik.uni-bonn.de> <86f8uu$3ds$1@kopp.stud.ntnu.no> <388BDEFA.8D57A66F@tismer.com>
Message-ID: <388C91E1.500B272A@hassler.de>

Christian Tismer wrote:

> Magnus Lie Hetland wrote:
> >
> > Cool!
> >
> > :)
> >
> > Without *really* having read all the documentation, I venture a question...
> > By using S.P. and continuations etc. -
> > would it be possible to stop and pickle a thread? This might be very
> > interesting for Mobile Agent Technology
> > etc. (I.e. a running program can migrate over a network and resume its
> > execution etc...)
>
> This can of course be done. A little more structure will be
> necessary to register a particular interpreter for pickling.
> Continuation objects would need to learn how to save their
> state as well. If there is interest, I will look into it.
>
> Not so clear at the moment is how I would deal with real threads,
> so this issue might be a little harder to solve.
>
> ciao - chris
>
> --
> Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
> Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
> D?ppelstr. 31                :    *Starship* http://starship.python.net
> 12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
> PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
>      we're tired of banana software - shipped green, ripens at home

Chris, please add this point to your arguments for stackless python!

Gerald



From python at scoobysnacks.com  Sat Jan  1 11:04:08 2000
From: python at scoobysnacks.com (Marshall)
Date: Sat, 01 Jan 2000 16:04:08 GMT
Subject: pickle problem:cant reload object saved with binary option
Message-ID: <386e1de1.998928692@news-server>

I created a class instance (self.info) to hold all my variables and
I'm trying to save it for use in later sessions.  The binary file is
much smaller but if I set the binary flag while saving, it blows an
error when I try to load it.  I'm trying this on Win98 and I'm calling
it from a Menu widget using command=self.save and I change the default
bin=0 to bin=1 and it won't load anymore.  The load error is at the
bottom.  Any direction is greatly appreciated.  Thanks.

The documentation says ...
The Unpickler class does not have an argument to distinguish between
binary and text pickle formats; it accepts either format. 


	def save(self,file='test.sav',bin=0):
		blah=open(file,'w')
		pickle.dump(self.info,blah,bin)
		blah.close()
	def load(self,file='test.sav'):
		blah=open(file,'r')
		self.info=pickle.load(blah)
		blah.close()
		self.buildlist()

Exception in Tkinter callback
Traceback (innermost last):
  File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 764, in
__call__
    return apply(self.func, args)
  File "new.py", line 249, in load
    self.info=pickle.load(blah)
  File "C:\Program Files\Python\Lib\pickle.py", line 826, in load
    return Unpickler(file).load()
  File "C:\Program Files\Python\Lib\pickle.py", line 495, in load
    dispatch[key](self)
  File "C:\Program Files\Python\Lib\pickle.py", line 509, in load_eof
    raise EOFError
EOFError:

P.S. python at scoobysnacks.com is valid email


From tim_one at email.msn.com  Fri Jan 14 18:22:57 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 14 Jan 2000 18:22:57 -0500
Subject: class.__getattr__ behavior:  feature or bug?
In-Reply-To: <200001142224.QAA27912@ws89.krumm0.com>
Message-ID: <000a01bf5ee6$4aacf080$a0a2143f@tim>

[Huayin Wang]
> I used to think that I know what's a bug and what's a
> feature, now I am confused.

Well, you're playing with low-level hooks now!  They're both powerful and
dangerous.  *Expect* to spend a lot of time debugging when you work at this
level; and stay away from this level unless you truly need this kind of
power.

> I do not understand why I would want to use __getattr__
> special method to control/define other special methods
> like '__repr__' when I can just define it if I want to.

__getattr__/__setattr__ make no distinction between "special" methods and
"other" methods -- they wouldn't be powerful enough to do their jobs if they
didn't apply to *all* methods.  By defining __getattr__, you're telling
Python "I don't like the way *you* look up attributes, so I'm going to do it
myself".  In return for the power, you also have to take full responsibility
for doing everything Python does for you when you don't use __getattr__.

BTW, part of what Python normally does for you is raise AttributeError  when
it can't find an attribute.  As another followup said, if your __getattr__
had raised AttributeError instead of returning None when looking up
__repr__, your program would have worked (well, it would have run a little
longer <wink>).

> By the way, does any of you tested you code/idea ;) and
> is infinite __getattr__ loop a bug?

>> your-next-post-will-be-wondering-whether-an-infinite-
>>    __getattr__-loop-is-a-bug<0.5-wink>-ly y'rs  - tim

> never mind :)

Heh heh.  If a __getattr__ hook is defined, you have to be very sure that
self.x is defined before __getattr__ is called, for every x such that the
__getattr__ implementation references self.x (else __getattr__ will invoke
itself again, ad infinitum).  That's a mouthful too, but it's obvious after
the first six times you debug it <wink>.

__setattr__ is even subtler, because it's called even for attributes that
*are* already defined.  This is why you'll sometimes see bizarre code like

    self.__dict__[somename] = somevalue

in methods of classes that define __setattr__; it's a way of spelling

    setattr(self, somename, somevalue)

that *doesn't* trigger __setattr__.

the-surest-way-to-avoid-these-surprises-is-to-avoid-
    these-methods-ly y'rs  - tim





From pbible at home.com  Mon Jan 17 19:18:47 2000
From: pbible at home.com (Paul E. Bible)
Date: Tue, 18 Jan 2000 00:18:47 GMT
Subject: PYTHONPATH newbie question
References: <00091c0e.fbb4d1d4@usw-ex0102-012.remarq.com>
Message-ID: <HjOg4.836$iL4.7074@news.rdc1.tx.home.com>

Under Windows NT, go into the Control Panel and select the System icon.
When it comes up, select the Environment tab, which will display the
environment variables for your system.  Click on the User Variables for
<username> listbox, then enter PYTHONPATH in the Variable text box and the
actual path in the Value text box.  Then click on the Set button and you're
done; except that you have to reboot the system for this variable to get
set.

Sorry in advance if this is too much detail...
Paul


"johnM" <john.mcginnis at usa.net> wrote in message
news:00091c0e.fbb4d1d4 at usw-ex0102-012.remarq.com...
> I have just recently been twiddling with Python and have to say this is
> a great scripting language.
>
> My Question:
>
> How does one make changes to PYTHONPATH permanent? I can import module
> sys and call sys.path.append("dir") no problem. I can call sys.path and
> see that dir has been added.
>
> But when I exit the shell and reenter the shell the dir entry in
> PYTHONPATH has disappeared. I am running Python 1.5.2 on a Windows NT
> 4.0 SP4 system.
>
> Any hints or am I missing something?
>
> Thanks in advance.
>
> JohnM, the highly impressed Python newbie....
>
>
>
> * 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 mwh21 at cam.ac.uk  Wed Jan 19 06:55:18 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Jan 2000 11:55:18 +0000
Subject: Python Rocks! - get rid of colons
References: <860ftk$bas@news.or.intel.com> <8610or$nsu$1@news.udel.edu> <861jm5$nrj$1@mach.vub.ac.be> <Pine.WNT.4.21.0001191125160.243-100000@wks_3.e-shopwork.com>
Message-ID: <m3bt6i9syh.fsf@atrus.jesus.cam.ac.uk>

Sposhua <sposhua at my.pc> writes:

> I'm one of the "prefers indentation to {}" ppl, and I'd just like to
> add: whe the heck do we need colons (:) at the end of if's 'n loops
> anyway? The only loss I see is to these 1-liners:
> 
> if a==b: eric=idle
> 
> What's wrong with:
> 
> if a==b
> 	eric=idle

I should probably wait for someone more authoritative, but I seem to
recall that this is the result of research done at CWI before Python
was invented; people find it easier to read with the colons.

I'm not sure what `people' they used, but I think I agree with them.
 
> (oh, and wouln't it make life easier to change == to = and let
> python work out when it is assingment and when it is comparison?)

What so the string `a = 1' would mean different things if it was 

if <here>:
    pass

than if it was

if 1:
    <here>

? No thanks, that sounds like ... perl.

Not going to happen, anyway.

Cheers,
Michael


From fredrik at pythonware.com  Fri Jan 21 06:52:16 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 21 Jan 2000 12:52:16 +0100
Subject: A modest Error (I mean Proposal)
References: <MODh4.1808$D4.62243@newsread1.prod.itd.earthlink.net>
Message-ID: <000901bf6405$f7692360$f29b12c2@secret.pythonware.com>

Arinte <shouldbe at message.com> wrote:
> I am getting this weird error???
> Traceback (innermost last):
>   File "Beer", line 17, in ?
>   File "D:\POSSApps\anotherposs app\PossDevice.py", line 37, in IOCTL
>     if type(command)==type(2):
> TypeError: len() of unsized object

> Here is the code for the function it craps out on...

[snip]

is this the entire program, or are any user-written
C extensions involved in this?

a great way to get utterly weird errors is to have
a C extension function which ignores an internal
exception, but returns to Python without clearing
the error state (PyErr_Clear).

if you cannot fix the C code, you can clear the
error state from Python, using something like:

    mymodule.badfunction()
    hasattr(None, "none")

just after the C call.

</F>





From jam at quark.emich.edu  Tue Jan 25 08:16:07 2000
From: jam at quark.emich.edu (Jeff)
Date: Tue, 25 Jan 2000 08:16:07 -0500
Subject: Embedding Python in C-code??
In-Reply-To: <388D9FB5.7389D8E1@bibsyst.no>; from thomas@bibsyst.no on Tue, Jan 25, 2000 at 02:05:57PM +0100
References: <388D9FB5.7389D8E1@bibsyst.no>
Message-ID: <20000125081607.B21863@quark.emich.edu>

On Tue, Jan 25, 2000 at 02:05:57PM +0100, Thomas Weholt wrote:
> Hi,
> 
> Just wondered, and it`s probably stupid, if it`s possible to use python
> code/objects/modules from a C++/C-program. I mean, Glade and alot of
> other GUI-builders are aimed at C/C++ development. If we could somehow
> use Python as foundation and just access that from a gui built using
> C/C++ it would be cool.
> 
> Thomas

you can embed the python interpreter in C/C++ programs.. take a look at the
'extending and embedding' manual for tips:

http://www.python.org/doc/current/ext/ext.html

-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From gerrit.holl at pobox.com  Tue Jan 18 17:00:42 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Tue, 18 Jan 2000 23:00:42 +0100
Subject: Python Rocks!
In-Reply-To: <388466D4.CC797CFC@earth.ox.ac.uk>; from nickb@earth.ox.ac.uk on Tue, Jan 18, 2000 at 01:12:52PM +0000
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk>
Message-ID: <20000118230042.B2833@stopcontact.palga.uucp>

Nick Belshaw wrote on 948197572:
> >
> > Get used to it, and don't use tabs.
> >
> 
> ! - whats wrong with tabs?
> 
> :-)))))))))))))

Some newsreaders (MS outlook) turn them into spaces.

Jeesh, I need to get to bed!

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From tim_one at email.msn.com  Fri Jan 21 04:38:08 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 21 Jan 2000 04:38:08 -0500
Subject: What about REBOL (www.rebol.com)
In-Reply-To: <ngQh4.17853$pb2.1435429@tw11.nn.bcandid.com>
Message-ID: <001401bf63f3$3a6eb5c0$ec2d153f@tim>

[Jayr Magave]
> ... I would like to know if someone evaluated the new
>  script language REBOL.  It looked  very powerful and high
> level to me.  But lacks library support (specially
> database support).
>
> I am about to take my time to learn an useful multiplataform
> script language and Python looked very fine, until I took
> at look at REBOL.

I picked up REBOL when it was first released, and was semi-active on the
REBOL mailing list for about a year.  It's a seriously cool language.  So is
Python.  Neither use semicolons much, so what's to dislike about either of
'em <wink>?

Which is a backwards way of saying that, in the absence of you identifying
what's important to you, there's no way short of a large book to "compare &
contrast" the aspects of interest to you.  As you've already discovered,
REBOL (at least in part because it's much newer) has far fewer libraries
available.  What else do you care about?

One large & basic difference is that Python is primarily object-oriented,
while REBOL is primarily function-oriented.  I count REBOL as a member of
the lexically scoped branch of the Lisp tree (despite that it *looks*
nothing like Lisp, a REBOL "block" is a lexical closure, and the semantics
are virtually identical to Scheme's).  Python hangs off the OO branch of the
Algol tree.

So if you have an OO-vs-functional bias, you'll definitely feel more
comfortable at first with one of them.

But enough guessing -- learning languages is good for you.  Learn both!
Work your way thru the Python tutorial for an hour, then switch to REBOL's,
& so on.  Judge for yourself.  You wouldn't buy a car without taking it for
a test drive, right?  Picking a language is more like choosing a mate
<wink>.

bring-them-both-flowers-and-call-'em-the-morning-after-ly
    y'rs  - tim





From grant at nowhere.  Tue Jan 18 12:08:44 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 18 Jan 2000 17:08:44 GMT
Subject: olemsg32 module used in win32com exchange server example?
Message-ID: <slrn8897gs.3cm.grant@grante.comtrol.com>

I'm trying to run the exchange server test script that I found
in the win32com sources.  I've two questions:

 1) Where does the "olemsg32" module come from?  I've installed
    python 1.5.2 and win32all and I don't seem to have olemsg32.

 2) Does the win32com stuff talk directly to the Exchange
    server, or does it go through a local copy of Outlook?

-- 
Grant Edwards                   grante             Yow!  I'm in DISGUISE as a
                                  at               BAGGAGE CHECKER....I can
                               visi.com            watch the house, if it's
                                                   ORANGE...


From grant at nowhere.  Thu Jan  6 12:37:11 2000
From: grant at nowhere. (Grant Edwards)
Date: Thu, 06 Jan 2000 17:37:11 GMT
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu>
Message-ID: <slrn879km7.n3d.grant@grante.comtrol.com>

In article <852g8h$75 at gap.cco.caltech.edu>, Phil wrote:

>_Python Essential Reference_ is just what it says; a reference.
>If, like me, you are an experienced (which says nothing about the
>quality of my code ;^)) coder, you can use it to learn Python.

Seconded.  If you already know a few languages, and already now
OOP, then PER is fine to learn from.

-- 
Grant Edwards                   grante             Yow!  Psychoanalysis?? I
                                  at               thought this was a nude
                               visi.com            rap session!!!


From gmcm at hypernet.com  Fri Jan 21 08:49:06 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 21 Jan 2000 08:49:06 -0500
Subject: Python -> EXE?
In-Reply-To: <869aqk$mlt$1@nnrp1.deja.com>
Message-ID: <1263658529-14676682@hypernet.com>

tiddlerdeja at my-deja.com wrote:

> Unfortunatly it doesn't. I did try that, then it failed because it
> couldn't find pythoncom.dll. I have pythoncom15.dll and I then renamed
> that too. I then got my EXE but when I ran it it did nothing. No
> errors, no nothing. It was of some size (I can't remember now).

It doesn't support COM, as the docs explain, because it can't 
reverse engineer the import hacks. If you find your way around 
that, you'll also need to figure out all the registry stuff required 
for COM to work, and do that manually. Mark has promised to 
help, but don't hold your breath; he's a busy man.
 


- Gordon



From thamelry at vub.ac.be  Tue Jan 25 04:56:43 2000
From: thamelry at vub.ac.be (Thomas Hamelryck)
Date: 25 Jan 2000 09:56:43 GMT
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <slrn88qq8g.3la.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <86js0r$l2s$2@mach.vub.ac.be>

Remco Gerlich <scarblac-spamtrap at pino.selwerd.nl> wrote:

: And many of them have become abominations. Python is over a decade old.
: Whitespace indentation is an important feature, that makes a lot of sense,
: and that makes code much clearer, but is different from other languages.
: If you don't like that, there is pindent.py.

I _do_ use pindent.py, but not always. It's not very convenient.

: That just can't change anymore and won't, so if you really can't live with
: it, "use another language" is sane advice, isn't it?

You are not getting my point. I _can_ live with python, even with the burdon
of obligatory indentation. I only start complaining about it again when I see
a newbie being buried with "obligatory indentation is the greatest invention 
since sliced bread, every body loves it, if you don't love it you're stupid" 
messages. 
Also note that I am not opposed to indentation itself of course. It makes programs a 
lot more readable. But _obligatory, overly rigid_ indentation makes your programs
less readable. The fact that I cannot do the following in python is a good 
illustration:

glBegin(GL_LINES)
    glVertex3f(self.x1, self.y1, self.z1)
    glVertex3f(self.x2, self.y2, self.z2)
glEnd()

Not to mention the weird error messages when you end up with a stray white space in
a python program.

: C++ is a good example of an abomination.

C++'s small user base as compared to python's is probably due to the lack
of forced indentation. :-)

Cheers,

---
Thomas Hamelryck    Institute of Molecular and Structural Biology
Aarhus University   Gustav Wieds Vej 10C
DK-8000 Aarhus C    Denmark                                                                        


From tseaver at aftershock.neosoft.com  Tue Jan  4 12:30:50 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 4 Jan 2000 11:30:50 -0600
Subject: CPAN for Python?
References: <38720F70.A5DC6F33@ttm.com> <84t52c$1gg$1@nntp9.atl.mindspring.net>
Message-ID: <C28D1A44F2165E48.B5C4E34C4CD957A5.44F995461BA2F6ED@lp.airnews.net>

In article <84t52c$1gg$1 at nntp9.atl.mindspring.net>,
Aahz Maruch <aahz at netcom.com> wrote:
>In article <38720F70.A5DC6F33 at ttm.com>,
>Scott Anderson  <sanderson at ttm.com> wrote:
>>
>>Is there an equivalent of Perl's CPAN for Python?
>
>Well, there's Parnassus (http://www.vex.net/parnassus/), but there's not
>yet an overarching authority.

www.python.org blesses it, which is probably about as close as we'll see.
I may misrecall (blessedly little Perl experience) but doesn't CPAN allow
for browsing from the interpreter prompt, and downloading?  That would be
a cool extension of a wonderful resource.

>OTOH, Python's "built-in" libraries are
>more extensive than Perl's; see the Library Reference for details.

parnassus.py,-anyone?'ly

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From ScrewMe at mypad.com  Sat Jan 15 03:24:38 2000
From: ScrewMe at mypad.com (TechSpood)
Date: Sat, 15 Jan 2000 02:24:38 -0600
Subject: Any Newbies?
Message-ID: <s2Wf4.974$sv3.3886@newsfeed.slurp.net>

Anyone besides me out there almost a complete newbie to all programming and
is learning python for a first language?






From davem at magnet.com  Thu Jan 27 13:57:40 2000
From: davem at magnet.com (Dave Mitchell)
Date: Thu, 27 Jan 2000 13:57:40 -0500
Subject: outtake from the Python Conference?
Message-ID: <Pine.SGI.4.05.10001271356040.18406-100000@lemur.magnet.com>

 What, is Guido running for prez now under the name Jim Taylor? :P

 http://www.nandotimes.com/newsroom/ap/ntn/images/politics/politics1.html

 -dave

 




From gmcm at hypernet.com  Wed Jan 12 09:11:02 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 12 Jan 2000 09:11:02 -0500
Subject: Threading & garbage collection
In-Reply-To: <m3embnpffi.fsf@havenrock.com>
Message-ID: <1264436227-5853368@hypernet.com>

Matt Gushee wrote:

> I'm trying to write a subclass of threading.Thread where each
> thread:
> 
>   1) is invoked by a GUI event handler
>   2) collects some data (to be exact, it downloads a web page
> 	and extracts some information from it), and finally
>   3) invokes a GUI method to display the data

It would probably be wise (and with some GUIs, absolutely 
vital) to make sure that all the GUI calls are made on the GUI 
thread. In other words, the worker thread should probably post 
the data to the GUI thread somehow. Queue.py is a likely 
candidate for passing the data. Depending on the GUI, you 
may be able to notify the GUI thread when the data is 
available, or the GUI thread may have to periodically check for 
it.


> ...Do the standard
> rules about circular references apply to threads? I.e., should I
> add something like:
> 
> 	self.win = None
> 
> to the above?

Sure thing.

- Gordon (who lives 2.5 hrs downeast from you)




From hnowak at cuci.nl  Fri Jan 14 16:40:11 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Fri, 14 Jan 2000 22:40:11 +0100
Subject: class.__getattr__ behavior:  feature or bug? 
In-Reply-To: <200001142112.PAA27786@ws89.krumm0.com>
Message-ID: <200001142140.WAA12189@dionysus.fw.cuci.nl>

On 14 Jan 00, at 15:12, Huayin Wang wrote:

> >>> class S:
> ...    def __getattr__(self, name):
> ...       if name == 'c':
> ...          self.a = self.a + 1
> ...          return self.a
> ...       else: return None
> ... 
> >>> s=S()
> >>> s
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> TypeError: call of non-function (type None)
> >>>

Feature. I think it works something like this: Typing the name of the 
instance (s) in interactive mode causes Python to look for the __repr__ 
method (and/or __str__, possibly), because it needs a printable 
representation... but it doesn't find it (since it's not defined, 
obviously). So it tries to find an attribute __repr__ using __getattr__... 
which returns None. Python thinks this result value should be a function 
(like __repr__ is... OK, so it's actually a method... same difference), and 
tries to call it. In this case, it ends up calling None, though, which 
leads to an error since None isn't a function.

This should work:

class S:
    def __getattr__(self, x):
        return None
    def __repr__(self):
        return str(id(self))

I must admit this error is kinda obscure. ^_^

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
You call me a masterless man. You are wrong. I am my own master.



From erno at iki.fi  Tue Jan 18 20:16:22 2000
From: erno at iki.fi (Erno Kuusela)
Date: 19 Jan 2000 03:16:22 +0200
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk>
Message-ID: <87ya9muahl.fsf@fun77.kivikko.hoas.net>

>>>>> "Nick" == Nick Belshaw <nickb at earth.ox.ac.uk> writes:

    Nick> ! - whats wrong with tabs?

Nothing at all, since no sane editor would ever use a tab
width other than 8, an no sane person would ever want anything
else than 8-character indents.

  -- dr. 8


From drj at mail.bulkley.net  Sun Jan 30 01:31:47 2000
From: drj at mail.bulkley.net (Peter Schopfer (aka Dr. J. - magician))
Date: Sat, 29 Jan 2000 22:31:47 -0800
Subject: opening files in windows and putting seperators in windows
Message-ID: <3893DAD2.B29FF140@mail.bulkley.net>

Hi

I am trying to fassion a word processor of sorts, and i was wondering if
there is any way of calling the Windows 9x open "thing". it is that mini
browser that you get when you open files in a lot of windows
applications. it there anyway to call this from within Tkinter? If so,
how would you do this?
Also, how do you add seperators in a root window?

keagan schopfer



From jasont at wellmed.com  Thu Jan 27 11:29:50 2000
From: jasont at wellmed.com (Jason Taylor)
Date: Thu, 27 Jan 2000 08:29:50 -0800
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com>
Message-ID: <MPG.12fa1256cc30caf0989682@or.news.verio.net>

I just installed this new DLL, and it's a night-and-day difference.  
There's almost no wait the first time, and it's as fast as Perl the 
second and subsequent times.  By switching back to the old DLL, the wait 
reappears.

Thanks guys again for your help!!!


From hanche at math.ntnu.no  Mon Jan 24 12:16:58 2000
From: hanche at math.ntnu.no (Harald Hanche-Olsen)
Date: 24 Jan 2000 12:16:58 -0500
Subject: In defence of the two-namespace rule
References: <E12BIkF-0003Zz-00@lsls4p> <m3n1q0fzg7.fsf@greebo.nodomain.de> <869uel$suj$1@nntp1.atl.mindspring.net> <slrn88hnpb.brp.quinn@auto.ugcs.caltech.edu> <slrn88n40a.2qh6.cjc26@bridget.mindriot.net> <948708715.22839@news.adtranzsig.de>
Message-ID: <pco1z775r05.fsf@math.ntnu.no>

+ "Dirk-Ulrich Heise" <hei at adtranzsig.de>:

| I don't really know what a closure is, not being
| functional-programming-minded, but could one say that a closure is a
| sort of specialization of a function, i mean, a variant of a
| function where a certain value replaces a "variable"s value?

I think you have the right idea, but you should probably replace
"value" by "binding".  Now I am no computer scientist myself, so this
could be slightly wrong (but hopefully not completely so).  The way I
understand it, a closure is a function together with a specific
binding for each of the non-local variables of the function.  A
binding associates a location with each variable, the locations being
where values are stored.  (In a purely functional language you don't
notice the distinction between bindings and values, since you cannot
change the contents of locations; you can only create new ones and
fill them with values.  This is one of the reasons why it is easier to
reason about programs in functional languages.)

| And does the word "closure" play an important role outside of the
| functional programming world?

Sure.  In Scheme, for example, you can create a simple counter as
follows:

(define (new-counter)
  (let ((count 0))
    (lambda ()
       (set! count (+ 1 count))
       count))))

which you would use as follows:

(define counter (new-counter))

Then the first time you call (counter) you get 1.  The second time you
get 2, and so on.  What happens is that (new-counter) returns a
closure, consisting of a function (the inner lambda) together with a
binding associating the variable count with a specific location.  Each
time the closure is called, it updates the counter and returns the new
value.

| And what does it have to do with scopes?

Everything and nothing.  The above example would look somewhat like
this in python:

def new_counter():
    count=0
    def counter():
        count = count + 1
        return count

except that would not work because of python's scoping rules.  You'd
have to make it a class instead:

class Counter:
    def __init__(self):
        self.count = 0
    def next(self):
        self.count = self.count + 1
        return self.count

You create a new counter using

counter = Counter()

and call it as counter.next().  [You could simplify the interface by
renaming the next method as a __call__ method, but that is beside the
point.]

You could still think of counter.next as being a sort of closure,
python terminology notwithstanding[*], in which the self variable is
bound to a specific Counter instance.  And, indeed, if it referenced
any global variables, those variables would have bindings to, in the
global namespace in which the Counter class was defined.

[*] Pythonesque for this is "bound method".

But because of python's scoping rules, as we have seen the notion of
closure is not as generally useful when discussing pythonic
constructs.

Insert-clever-witticism-here-<wink>-ly y'rs,
-- 
* Harald Hanche-Olsen     <URL:http://www.math.ntnu.no/~hanche/>
- "There arises from a bad and unapt formation of words
   a wonderful obstruction to the mind."  - Francis Bacon


From malcolmt at smart.net.au  Mon Jan 24 05:53:59 2000
From: malcolmt at smart.net.au (Malcolm Tredinnick)
Date: Mon, 24 Jan 2000 21:53:59 +1100
Subject: Tkinter and wxPython
In-Reply-To: <388B060B.D13F7B64@exceptionalminds.com>; from Timothy Grant on Sun, Jan 23, 2000 at 05:45:47AM -0800
References: <388B060B.D13F7B64@exceptionalminds.com>
Message-ID: <20000124215359.A1166@Ridcully.home>

On Sun, Jan 23, 2000 at 05:45:47AM -0800, Timothy Grant wrote:
> OK, After just spending three weeks learning Tkinter on-the-fly, I have
> to ask this question, what advantages/disadvantages are there to each
> Tkinter and wxPython?

I'll have a stab at this and will no doubt be stomped on by greater minds
tomorrow (although most of the Greater Minds are probably all at the
conference, so I may be safe) ...

Tkinter:
+ It is truly cross-platform: The same program will run without changes on
machines with MS Windows, MacOS or any version of Unix that can run a version
of X Windows. You can also be "reasonably certain" that if Python is installed
on a platform, Tkinter will be there, too (or, if not, can be easily
obtained).
+ It is reasonably well documented and getting more so. There are already
articles around on how to "translate" Tcl/Tk man pages into Tkinter commands
and I, personally, have found some stuff on Perl/Tk to be helpful here, too.
An O'Reilly publications book on Tkinter is due out later this year.
- It is a bit slow to start up and run (since Tk is written in Tcl, which is a
fairly slow language -- at least by Python standards).
- Many people feel it looks a bit "clunky", although this seems to vary from
person to person. I don't mind it (and the amount of time I spend playing
PySol means I spend a lot of time looking at a Tkinter interface :-) ).

wxPython:
+ A bit more "feature laden" than Tkinter in some respects (this may be a "-"
in some books). More things are customisable (sort of -- see below). I believe
there are more "ready to work" gadgets in wxPython, but I haven't sat down and
counted and I don't use Tkinter much, so I probably don't know what is really
available there.
+ Although I haven't benchmarked stuff, it certainly feels faster than
Tkinter. Noticeably so when starting up a program, but also when a program
with a lot of windows is running. Tkinter tends to bog down a bit when lots of
windows need to be updated. wxPython is a step or two closer to the native
C-code (well, C++ code in this case), so this may explain it.
+ Updated regularly. Robin Dunn (Mr. wxPython) seems to fix bug reports about
two minutes before you send him the email and you can download the current CVS
version (or even a compiled snapshot thereof for Microsoft-types) as often as
you like.
+/- Not a lot of documentation (in a nutshell: read the wxWindows docs and
then you know pretty much everything about wxPython. The problem is, the docs
are like a reference manual, rather than a tutorial). However, (good point!)
the wxPython mailing list is an excellent source of help.
- Trickier to install than Tkinter. This is partly because wxWindows, the
product it is wrapping is still very much in development, so there are a
reasonable number of bugs (although I seem to have stopped being bitten by
them in the last three or months, so it is definitely getting better).
- Not included with the standard Python distribution, so if you are
distributing an application, you need to include instructions on how to
download and install wxWindows and wxPython (again, this is getting easier
every day, but it is still non-trivial).
- Not available (yet!) on Macintoshes, although Robin recently made it clear
he will be happy to fix that as soon as a Macintosh machine complete with
development software is made available to him. (wxWindows is available on
Macintoshes, so it's "only" a matter of getting the Python bindings to
compile, as far as I understand it).
- Your applications don't look and feel 100% the same on MS Windows and X
Windows platforms. This is because wxWindows does *not* aim for the lowest
common denominator across all its implementations. Rather, it supplies a
bunch of features and some of them may be no-ops on any given implementation.
The differences, however, are not major (mostly cosmetic things).
- Steeper learning curve than for Tkinter. wxWindows (and hence wxPython) is a
BIG package and can be quite intimidating if you haven't done any sort of
windows class library programming (GTK, Motif/LessTif or MFC, say) before.

Having written all that, my own personal preference is for wxPython for any
moderate sized application and Tkinter for the "quick and dirty" stuff where I
just want a window *now*. I like the look-n-feel of wxWindows applications.
Of course, I only write apps for myself, so my audience is very sympathetic
towards the programmer. :-)

Don't be too put off by the extra learning for wxPython -- you won't be able
to write the next super-app overnight, but with a bit of work (and much
reading of the example/test scripts) it is manageable.

Hopefully this is of some benefit (and not too far from reality).

Cheers,
Malcolm Tredinnick

--
For every action there is an equal and opposite criticism.




From matthias.huening at univie.ac.at  Sun Jan  9 05:19:15 2000
From: matthias.huening at univie.ac.at (Matthias Huening)
Date: Sun, 9 Jan 2000 11:19:15 +0100
Subject: RegExp, Python and strings
References: <000301bf5a27$a28f7c00$242d153f@tim>
Message-ID: <859nio$ocm$1@www.univie.ac.at>

Hi Aahz, Andrew & Tim,

Thank you very much for your answers to my question.
My problem was the "r'...'". I misunderstood its function (as I said: these
are my fist steps in Python).

I also thank you for your comments on Python & strings, Tim. I don't know
yet if I am 'thinking in Perl'. I like it, but I am certainly not a real
Perl adept. I am still looking for the right language for me. As my first
impression of Python is very good, I think I'll be around here for a
while...

Matthias
- - - - -
matthias.huening at univie.ac.at
http://www.ned.univie.ac.at/





From nhv at cape.com  Thu Jan 13 18:54:38 2000
From: nhv at cape.com (Norman Vine)
Date: Thu, 13 Jan 2000 18:54:38 -0500
Subject: Desperate: Linux Sound Module
References: <20000113210535.A2015@stopcontact.palga.uucp>
Message-ID: <85lokb$14bf$1@newsie2.cent.net>

Gerrit Holl wrote

> I'm really desperate: I can't find any module for playing sounds on a
> Linux system! I already searched everywhere: google, python.org,
parnassus.
> I just want to play .au or .wav, but even edna can't *play* *play* *PLAY*
> an mp3!

Open Source cross platform game library in C++
with basic sound support
http://www.woodsoup.org/projs/plib

Will have python interface shortly.

Norman




From aa8vb at yahoo.com  Thu Jan 20 09:44:00 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 20 Jan 2000 09:44:00 -0500
Subject: Looking for the FAQ
In-Reply-To: <lefebvre-2001000912180001@landru.medcor.mcgill.ca>
References: <slrn88bqi2.512.sprawlsr@TacHQ.com> <lefebvre-2001000912180001@landru.medcor.mcgill.ca>
Message-ID: <20000120094400.A95675@vislab.epa.gov>

Luc Lefebvre:
 |sprawlsr at worldnet.att.net wrote:
 |> 
 |>         I'm new to this group. I run Linux and it has Python 1.5. I'm
 |> an ANSI C programmer, learning C++ at my University, and I'm
 |> interested in checking python out. Is there a good book that takes a
 |> person from beginning to at least intermediate level? This is probably
 |> something covered in the FAQ, so if you just point me to the FAQ, I'll
 |> take it from there.
 |> 
 |> Robert
 |
 |There is the book "Learning Python" from O'Reilley which got a good review
 |in Linux Journal.

Before you go spending money on books (life as a college student can be
tough), there are plenty of tutorials on-line to give you a flavor.  Here
are just a few.  The first one will bring you up to intermediate level on
the language in no time:

    http://www.python.org/doc/current/tut/tut.html     (The official one)
    http://www.crosswinds.net/~agauld/
    http://www.troubleshooters.com/codecorn/python/pptut.htm

Then bookmark this link:

    http://www.python.org/doc/

And take a quick scan of the TOC of the Library Reference on this page,
particularly section 2, to see what's out there just waiting to be used in
a default Python install.  There are other modules folks offer separately,
but this set will give you a good start.

-- 
Randall Hopper
aa8vb at yahoo.com



From lobozc at my-deja.com  Mon Jan 17 21:07:23 2000
From: lobozc at my-deja.com (lobozc at my-deja.com)
Date: Tue, 18 Jan 2000 02:07:23 GMT
Subject: [Types-sig] Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk> <200001170221.VAA10441@eric.cnri.reston.va.us>
Message-ID: <860hsj$917$1@nnrp1.deja.com>

In article <200001170221.VAA10441 at eric.cnri.reston.va.us>,
  Guido van Rossum <guido at CNRI.Reston.VA.US> wrote:
> > I intend to push this proposal during IPC8.  I would greatly
appreciate
> > constructive comments in advance, if only to temper my
enthusiasm ;^}
>
> Eddy, it's too late in Python's life for radical proposals.  My head
> hasn't exploded yet, but it sure hurts whenever I try to read your
> proposal carefully.  I'm not a functional programmer, and Python is
> not a functional programming language; and as far as I'm concerned it
> never will be one.
>
> Maybe you'll make the case better in person -- I'll be there to
listen!
==========
Looking at the problem from the outside - not heavy python/fp user -
general comment.

Python is not an fp language. But it shouldn't stop Python from
borrowing some mechanisms from fp (and other paradigms)- especially as
they obviously reduce the coding and make for safer programming.
Examples here are Map function and things like list comprehensions.
Naturally, they should be implemented withoug breaking Python.

Limiting Python to just oo, without borrowing tried and effective
computational mechanisms from other programming paradigms and languages
may stunt its growth in the longer term.




>
> (Practical comment: the syntax <generator> <name> [<bases>]:
> <suite> won't fly; you seem to want <generator> to be any expression,
> and an expression followed by a name is too fragile a construct to be
> comfortably parseable (even if it might be unambiguous).)
>
> --Guido van Rossum (home page: http://www.python.org/~guido/)
>
>


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


From samschul at pacbell.net  Wed Jan 26 16:22:01 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Wed, 26 Jan 2000 21:22:01 GMT
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
References: <816010E2456BD111A48700805FBBE2EE01C60408@ex-quebec-u1.baan.com>
Message-ID: <86nohk$ukl$1@nnrp1.deja.com>

I foreworded this messege to Pacbell, and received the reply from them
that the orgin of the message was from AOL, and that they were using
pacbell customer names to hide the source of the message.

Sam Schulenburg



In article <816010E2456BD111A48700805FBBE2EE01C60408 at ex-quebec-
u1.baan.com>,
  Gaetan Corneau <Gaetan_Corneau at baan.com> wrote:
> I'm really tempted to hack this guy's machine.
> Destroy everything.
> Or at least bust his mailbox.
>
> The kind of spammer I hate the most.
> Sorry, I know this is off-topic.
> ______________________________________________________
>    Gaetan Corneau
>    Software Developer
>    Software Engineering Process Group
>    BaaN  Supply Chain Solutions
>    http://www.baan.com
>    E-mail: Gaetan_Corneau at baan.com
>    Tel: (418) 266-8252
> ______________________________________________________
> "Profanity is the one language all programmers know best"
>
> -----BEGIN GEEK CODE BLOCK-----
> Version: 3.1
> GCS/B/MU d- s+:++ a+ C++
> UL+ P--- W+ N- K- W++ t--
> !5 X- R+ tv-- b++ DI++
> G e++ h---- r+++ y++++
> ------END GEEK CODE BLOCK------
>
> > -----Original Message-----
> > From:
> > 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > [mailto:24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.
> > pacbell.ne
> > t]
> > Sent: Sunday, January 23, 2000 3:45 PM
> > To: python-list at python.org
> > Subject: DARREN MULLIGAN SEEKING MODELS! 89582
> >
> >
> > My name is Darren Mulligan and I am an aspiring photographer
> > from California.
> > I will get right to the meat and gravy of this post, so as
> > not to waste your time!
> >
> > +++++WHAT AM I SEEKING?+++++
> >
> > I am looking for young models (prefer early teen 14-16 year
> > old female) for nude and semi-nude photography. This is a
> > great way for me to update my portfolio, and also a great way
> > for you to get a free portfolio, too! This could lead to
> > employment in the Adult XXX entertainment business. What a
> > great way to make money!
> >
> > Please respond to one of the following e-mail addresses. I
> > run my very own mail server on an ADSL connection so you can
> > be assured to complete discretion and privacy. If one e-mail
> > address does not work, simply try the next!
> >
> > Sorry for the long list! My ADSL connection goes down from
> > time to time, so I posted all the e-mail addresses I can be
> > reached at. If one does not work, just move on to the next!
> >
> > nanae at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > usenet at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > news.admin.net-abuse.email at adsl-63-197-143-
124.dsl.scrm01.pacbell.net
> > spamlinkpage at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> > alt.make.money.fast at nntp.8m.com  nanas at nntp.8m.com
> >   hostmaster at FREESERVERS.COM
> >
> > THIS IS NOT A SCAM! IT IS NOT CHILD PORN!!
> > There is nothing wrong or illegal about the artistic
> > representation of a young body. If you have a problem with
> > this post, please e-mail me and I will not post to your
> > newsgroup again!
> >
> > Sincerely,
> > Darren Mulligan
> >
> > --
> > http://www.python.org/mailman/listinfo/python-list
> >
>
>


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


From thomas at bibsyst.no  Tue Jan 18 06:36:49 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Tue, 18 Jan 2000 12:36:49 +0100
Subject: Using glade to build guis
Message-ID: <38845051.5A9A1A49@bibsyst.no>

Hi,

I`ve tried to use Glade to build guis and run them with the glade.py
module/script. What I don`t understand is how to make it react to
keypressing or other events. Does a manual/tutorial/example exists on
how to prototype a gui in glade then code events-handling and the
functional parts of the gui later ??


Thanks.
Thomas


From kern at caltech.edu  Sun Jan 23 16:50:38 2000
From: kern at caltech.edu (Robert Kern)
Date: 23 Jan 2000 21:50:38 GMT
Subject: Python seems to be growing fast.
References: <01bf65e9$6e2d2160$c9dc343e@dreadzone>
Message-ID: <86ft3e$bar@gap.cco.caltech.edu>

In article <01bf65e9$6e2d2160$c9dc343e at dreadzone>,
	"Dirk Heise" <dheise at debitel.net> writes:
> I mean, the Windows download exe does -
> from 2.3 MByte in 1.5.1 to 5MByte in 1.5.2.
> Having a slow connection here at home, 
> this annoys me a little. Can a 0.0.1
> version difference really explain such a 
> large difference?

It was the decision to also include the TCL/TK installer that
increased the file size.  The major justification for it (that I can
recall) was that newbies who wanted to use IDLE would be perplexed if
TCL/TK wasn't installed along with Python.

> Dirk

Robert Kern
kern at caltech.edu


From nascheme at enme.ucalgary.ca  Wed Jan 19 21:05:56 2000
From: nascheme at enme.ucalgary.ca (Neil Schemenauer)
Date: Thu, 20 Jan 2000 02:05:56 GMT
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com> <865lsm$s3p@news.or.intel.com>
Message-ID: <slrn88crfc.eab.nascheme@adler.dynodns.net>

tye4 <tye4 at yahoo.com> wrote:
>Better yet. Fix the dumb language.
...
>Go ahead and convince these guys to switch do it the "Python Way".
>And Good luck. You will need it.

The Python interpeter has a very liberal license.  Why don't you
make this block terminator change and promote your version.  You
should have no trouble as "thousands of programmers have this
problem" with the current language.

You can also quit posting in comp.lang.python.  People in that
newsgroup are weird and unreasonable.  Besides, those few people
are insignificant compared to the thousands that will love your
new language.

Quick, go and leave the unenlighted behind.  Please?


  ignorance-is-bliss-ly Neil


From culliton at clark.net  Tue Jan 18 16:47:11 2000
From: culliton at clark.net (Tom Culliton)
Date: Tue, 18 Jan 2000 21:47:11 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com>
Message-ID: <zb5h4.1605$Ve7.38526@iad-read.news.verio.net>

In article <862jqd$l9s at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>
>Tom Culliton <culliton at clark.net> wrote in message
>news:YZRg4.1292$Ve7.28796 at iad-read.news.verio.net...
>> In article <860ftk$bas at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
>> Have you got extensive experience at designing and implementing
>> computer lanaguages like Guido?  Did you spend years working on a
>> major language design/development project (ABC) including lots of
>> usability studies like Guido?  Have you had the benefit of Tim Peters,
>> a veritable font of language implementation experience and good taste,
>> whispering (or occasionally shouting) in your ear?  I rather doubt it.
>>
>Are you insinuating that this is not a problem since I'm not as
>experienced as Guido?

First if had I intended insinuation it would have been more subitle
than that.  Second your experience relates to how much weight we're
going to assign to your _opinion_.  Indentation isn't a problem for me
or for most of other people around here, whose opinions and experience
I and the other folks here hold in much higher regard.

>I don't need to design a language to detect flaws in it.

Your opinion or prejudice doesn't make it a flaw.  Having designed a
widely used language would add weight to your opinion.  BTW - If you
want to wave Nick Wirth at us, I think he tends toward baby talk in
his designs.


From mskott at image.dk  Sun Jan 23 17:16:17 2000
From: mskott at image.dk (Martin Skøtt)
Date: 23 Jan 2000 23:16:17 +0100
Subject: Simple socket loop
References: <ln5gqz5p.fsf@jumbo.skott.dk> <86fqpc$5q0$1@nntp4.atl.mindspring.net>
Message-ID: <hfg4qvri.fsf@jumbo.skott.dk>

aahz at netcom.com (Aahz Maruch) writes:

> Instead of trying to recreate the wheel yourself, take a look at the
> asyncore library.

Well, some time I am so blind that I cant see the solution when its right in
front of my nose :-)

-- 
Many regards
Martin Sk?tt
mskott at image.dk

"Every time I learn something new it pushes some old suff out of my brain"
Homer Simpson


From gerrit.holl at pobox.com  Thu Jan 20 11:27:35 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 20 Jan 2000 17:27:35 +0100
Subject: Looking for Python IDE ?
In-Reply-To: <Ncyh4.6734$3b6.33129@ozemail.com.au>; from tylerpeter@hotmail.com on Thu, Jan 20, 2000 at 05:51:10PM +1100
References: <Ncyh4.6734$3b6.33129@ozemail.com.au>
Message-ID: <20000120172735.A4046@stopcontact.palga.uucp>

tyler Peter wrote on 948387070:
> Hi
> I used to work in VB6 and am planning on porting an application to Python
> but would like to have a nice IDE like VB6 to work in.
> 
> Are there any good IDE's available for Python running on NT and Linux

Idle. It's in the Tools/ dir. of the dist.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From dworkin at ccs.neu.edu  Thu Jan 20 14:06:40 2000
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 20 Jan 2000 14:06:40 -0500
Subject: How to store a function pointer in class?
In-Reply-To: "Sami Hangaslammi"'s message of "Thu, 20 Jan 2000 20:12:27 +0200"
References: <867j5p$qvq$1@mordred.cc.jyu.fi>
Message-ID: <vndaem01s1r.fsf@betelgeuse.ccs.neu.edu>

"Sami Hangaslammi" <shang.remove_edu at st.jyu.fi.edu> writes:

> >>> class Test:
>         class_attribute = None
> 
> >>> def test_func(x):
>         print "test_func called with",x
> 
> >>> test_func
> <function test_func at b4e620>
> >>> Test.class_attribute = test_func
> >>> Test.class_attribute
> <unbound method Test.test_func>
> >>> Test.__dict__['class_attribute']
> <function test_func at b4e620>
> 
> Now I have a script where I need to store pointers to functions (among
> other things) to class attributes and retrieve them as ordinary
> functions. How do I do this

Well, 'pointers to functions' are not really a very Pythonic way to
think about such things.  A class or instance variable may be a
reference to a function object, though.

Since I'm not sure what you want to _do_ with your function pointers,
I'm not sure if I am answering your real question.  This may be helpful:

>>> class Test: 
...     class_attribute = None 
... 
>>> my_test = Test()
>>> def test_func(x): 
...     print "test_func called with",x 
... 
>>> Test.class_attribute = test_func # or my_test.class_attribute =
test_func
>>>                                  # depending on what you really
want
... 
>>> my_test.class_attribute()
test_func called with <__main__.Test instance at 80dff10>

Note that Python is not designed to have class methods, but rather
instance methods.  All functions defined inside a class expect to
receive an instance of the class as an implicit first argument.

>>> Test.class_attribute()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: unbound method must be called with class instance 1st argument
>>> Test.class_attribute(my_test)
test_func called with <__main__.Test instance at 80dff10>
>>> my_test.class_attribute()
test_func called with <__main__.Test instance at 80dff10>

The first call here is an error.  The second and third are equivalent
to each other.

-Justin

 





From claird at starbase.neosoft.com  Sun Jan 16 08:32:15 2000
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 16 Jan 2000 07:32:15 -0600
Subject: Does python support long longs?
References: <x8_f4.26$tX3.648@vixen.cso.uiuc.edu> <85qa3c$hk$1@nntp9.atl.mindspring.net>
Message-ID: <722C7DA3C5245C0A.D784A1B5D0E7D6C6.EABECEEA7DD01A77@lp.airnews.net>

In article <85qa3c$hk$1 at nntp9.atl.mindspring.net>,
Aahz Maruch <aahz at netcom.com> wrote:
>In article <x8_f4.26$tX3.648 at vixen.cso.uiuc.edu>,
>steve s chen  <sschen at ux8.cso.uiuc.edu> wrote:
>>
>>1) Does Python support long long datatypes?  
>
>Not really.  But it does have its *own* long type, which can literally
>be as large as the memory of your machine will support (i.e. millions of
>digits).
>
>>2) Does Python support unsigned 32-bit ints?
>
>Nope.
>
>Note that to correct both of these deficiencies, you could in theory
>build your own Pythonic-C extension if this is "mission critical" for
>you.
Perhaps the questioner realizes this is only a
formal deficiency, something like goto-lessness.
It's possible that what he *really* wants is
arithmetic with certain capabilities, and, as
we know, Python's up to that, and more.
>
>>3) A less significant issue -- is there a utility that analyzes 
>>	 a segment of python code verifying the dynamic name-bindings?
>>	 NameError exceptions weren't being reported until run-time;
>>	 does that mean we must QA each control path to ensure 'correct'
>>	 execution?
>
>Yup.  The way to build large Python applications is to componentize and
>loosely-couple the hell out of everything.
A lot more than just Python, of course.

I've been working on ways to explain this one.
Conscientious developers (and especially managers)
coming from C++ (say) to languages as dynamic as
Python or even Smalltalk "freak out":  "You mean
it doesn't even have the sense to verify by 'link-
time' that all required functions are present?!?"
As they're already comfortable in the glass house
of languages with "pointers" that might be, and
frequently are, unassigned, there's little gain
in casting stones (to mix images rather flamboy-
antly).  What I'm after instead is a positive
discipline we can present which addresses these
objections.

<URL://http://www.sunworld.com/sunworldonline/swol-01-2000/swol-01-regex.html>
hints at such a presentation.  I need to make it
much more accessible to the scripting-averse,
though.
			.
			.
			.
-- 

Cameron Laird <claird at NeoSoft.com>
Business:  http://www.Phaseit.net
Personal:  http://starbase.neosoft.com/~claird/home.html


From gerrit.holl at pobox.com  Sun Jan 16 13:49:14 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 16 Jan 2000 19:49:14 +0100
Subject: SUGGEST: os.getuser()
In-Reply-To: <000b01bf6049$36532140$f29b12c2@secret.pythonware.com>; from fredrik@pythonware.com on Sun, Jan 16, 2000 at 06:43:28PM +0100
References: <20000116174004.A5995@stopcontact.palga.uucp> <000b01bf6049$36532140$f29b12c2@secret.pythonware.com>
Message-ID: <20000116195710.A7700@stopcontact.palga.uucp>

Fredrik Lundh wrote on 948044608:
> Gerrit Holl wrote:
> > may I suggest this function for in the posix module:
> > 
> > def getuser():
> >     import pwd
> >     return pwd.getpwuid(getuid())[0]
> 
> you'll find a more complete implementation of that
> function in the getpass module:

The code of that getuser() is:

def getuser():
        """Get the username from the environment or password database.

        First try various environment variables, then the password
        database.  This works on Windows as long as USERNAME is set.

        """

        import os

        for name in ('LOGNAME', 'USER', 'LNAME', 'USERNAME'):
                user = os.environ.get(name)
                if user:
                        return user

        # If this fails, the exception will "explain" why
        import pwd
        return pwd.getpwuid(os.getuid())[0]

So the user can mislead the program. I think the other way would be better:

...
try:
    import pwd
    return pwd.getpwuid(os.getuid())[0]
except ImportError:
    import os
    for name in ('LOGNAME', 'USER', 'LNAME', 'USERNAME'):
        user = os.environ.get(name)
        if user:
            return user
    return None

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From mwh21 at cam.ac.uk  Thu Jan 27 03:47:55 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 27 Jan 2000 08:47:55 +0000
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
References: <m12DjXD-000CxiC@artcom0.artcom-gmbh.de>
Message-ID: <m34sbz2950.fsf@atrus.jesus.cam.ac.uk>

pf at artcom-gmbh.de (Peter Funk) writes:

> Look at the following coding alternatives often seen in generic python code:
> 
> class A:
>     def load(self): 
>         # ... do something e.g. load a file
>         try:
>             self.update()
>         except AttributeError:
>             pass
> 
> class B:
>     def load(self):
>         # ... do something e.g. load a file
>         if hasattr(self, 'update'):
>             self.update()
> 
> Which coding idiom is better?  The second form is one line shorter than
> the first, but it has the disadvantage, that the method (or attribute) must
> be named twice.  Both idioms are used in the standard libary.

The usual response is that you use the former if you only expect the
attribute look up to fail rarely ("exceptionally") and the latter if
you expect it to fail regulalarly.

There's also this approach:

class C:
    def __init__(self):
        up = getattr(self,"update",None)
        if up is not None: up()

This depends on Python 1.5.2 (I think) so that could be one reason it
doesn't crop up much in the standard library. It looks a bit odd to
me, not sure why.

I tend to use one of the first two depending on my mood.

Cheers,
Michael


From gnpotts at email.uophx.edu  Sun Jan  2 18:31:17 2000
From: gnpotts at email.uophx.edu (Gary N. Potts)
Date: Sun, 2 Jan 2000 18:31:17 -0500
Subject: New To Python
Message-ID: <1cRb4.10160$zp.149897@news3.mia>

I'm brand new to Python and really to programming.  I've dabbled a bit by
programming muds and the like, but nothing real.  I've downloaded Python
1.5.2 and all the documentation to go with it.  I've found several help
sites that are able to teach me the language, however, I have not been able
to find a decent instructional guide on how to actually use the program
itself.  Is there any good tutorials for that purpose out there and if so,
were can they be found?

--
Gary N. Potts
gnpotts at email.uophx.edu
garynp2 at aol.com





From flight at mathi.uni-heidelberg.de  Mon Jan 10 06:32:09 2000
From: flight at mathi.uni-heidelberg.de (Gregor Hoffleit)
Date: Mon, 10 Jan 2000 12:32:09 +0100
Subject: Python in Corel Linux
In-Reply-To: <87ya9ymdjp.fsf@psyche.evansnet>
References: <947294639.1002448618@news.islandnet.com> <87ya9ymdjp.fsf@psyche.evansnet>
Message-ID: <20000110123209.B3258@mathi.uni-heidelberg.de>

On Mon, Jan 10, 2000 at 11:22:42PM +1300, Carey Evans wrote:
> Corel Linux is based on Debian GNU/Linux <URL:http://www.debian.org>,
> so you should just be able to point apt at a Debian mirror and install
> Gregor Hoffleit's excellent Debian packages.
> 
> There may be some catches to this; I haven't done it myself, so I
> don't know what else you might need to install.

I haven't tried that myself and would be interested to hear reports, but
I would expect that it just works.

You might also want to try and install the Python 1.5.2 upgrade for slink
from http://www.debian.org/~flight/python/slink. Sorry, but I forgot the
apt line ;-).

    Gregor



From jeff.macdonald at inforonics.com  Mon Jan  3 15:56:08 2000
From: jeff.macdonald at inforonics.com (Jeff Macdonald)
Date: Mon, 03 Jan 2000 15:56:08 -0500
Subject: JPython Projects
References: <386D0703.EB0807AA@prescod.net>
Message-ID: <38710CE8.D67E36CD@inforonics.com>

GJC?


Paul Prescod wrote:
> 
> Here are some interesting JPython projects for someone with a time on
> their hands:
> 
>  1. Use the new GJC to compile JPython and CPython into the same binary.
> 
>  2. Set up a way of calling Python modules written for JPython (with
> Java library dependencies) from CPython and Python modules with C
> dependencies from JPython. Some things might work a little odd because
> of the two interpreters but I think most things would work...but I'm
> just guessing.
> 
>  3. Figure out how to cut one interpreter out of the loop. Get CPython
> to be able to directly call Java stuff or JPython to directly call
> C-coded modules. The former would probably be easier but the latter
> would allow us to leverage more of the JVM and JIT work that is being
> done.
> 
> I think that GJC is an important step in integrating the (thus far
> incompatible) C and JVM worlds and Python is the natural glue language
> for the new integrated environment.
> 
>  Paul Prescod

-- 
Jeff Macdonald
e-Applications Developer 
Inforonics, Inc. 
Voice mail: (978) 698-6342 
http://www.inforonics.com


From sanderson at ttm.com  Wed Jan 26 21:23:13 2000
From: sanderson at ttm.com (Scott Anderson)
Date: Wed, 26 Jan 2000 21:23:13 -0500
Subject: Creating Objects
References: <6AA37AF8C553513785256873000B44D3.000B456F85256873@ttm.com>
Message-ID: <388FAC11.F9A5CBAB@ttm.com>

Small suggestion:

Create an object for each system that will translate to and from a
neutral internal representation.

Then, for each message do a translate from the incoming system to the
neutral rep, then from the neutral rep out to the outgoing system:

msgObject = incoming.translateToNeutral(message)
outgoing.translateFromNeutral(msgObject)

I've used this method before with multiple, non-standard communications.

Any similarities between the format for the individual systems is up to
you to factor, as you haven't given any detail. :-)

Regards,

-scott

calishar@*NOSPAM*home.com wrote:
> 
> Hi folks,
> 
> I'm working on solving a problem that my boss put to me the other
> day, and I have run into a solution that I don't know how to implement
> in Python (or any other language for that matter)
> 
> Basic description of problem (particulars not included)
> 
> Work out a system that can import information from (any number) of
> systems, and write the infomration out (meaningfully) to any system in
> the set.
>



From jam at quark.emich.edu  Sun Jan 16 14:12:23 2000
From: jam at quark.emich.edu (Jeff)
Date: Sun, 16 Jan 2000 14:12:23 -0500
Subject: IMAP - fetch using imaplib
In-Reply-To: <85ssc4$8jk$1@snipp.uninett.no>; from apederse@siving.hia.no on Sun, Jan 16, 2000 at 05:42:06PM +0100
References: <85ssc4$8jk$1@snipp.uninett.no>
Message-ID: <20000116141223.C26923@quark.emich.edu>

On Sun, Jan 16, 2000 at 05:42:06PM +0100, Asle Pedersen wrote:
> I have adressed this question to the group previoussly without any success
> so I try once more :
> I need help with the following IMAP command:
> DF56 FETCH 1:3 (RFC822.SIZE BODY[HEADER.FIELDS (FROM SUBJECT)]).
> I have tried using the imaplib for other fetch commands without trouble, but
> this one I can't seem to get working utilising the imaplib.
> 
> -Asle Pedersen
> 

if you could perhaps post code and/or the specific error messages you are
seeing, it may help to diagnose the problem.

regards,
J
-- 
|| visit gfd <http://quark.emich.edu/>
|| psa member -- <http://www.python.org/psa/> 
|| New Image Systems & Services, Inc. <http://www.newimage.com/>



From ivanlan at callware.com  Wed Jan 19 09:04:12 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 19 Jan 2000 07:04:12 -0700
Subject: Tkinter hangs again under Windows
References: <20000119064523.A15259@foxboro.com>
Message-ID: <3885C45C.F97A887B@callware.com>

Hi All--

Hans-Joachim Widmaier wrote:
> 
> Niels Diepeveen wrote:
> 
> > This seems to be the problem. If you show any dialogs before entering
> > mainloop, you won't receive events anymore.
> 
> Apparently this is the case. Is this a (known?) bug or just an undocumented
> feature, or is it even documented? The problem seems to only exist under
> Windows.
> 
> Thanks a lot. I'll find a way around it, probably a self.after() that
> shows the dialogue shortly after entering the mainloop.
> 

Use the tkMessageBox functions:
----------------------------------
root=Tk()
..
..
tkMessageBox.showwarning("Bad filename",
            ("Can't open file %s" % (sys.argv[1])),parent=root)
----------------------------------

I.e., go ahead and create your main window, and _then_ put up a dialog
box.  If you specify the "parent=root" option, everything will work just
fine.  If you don't specify it, you're hosed.

<some-hoses-stop-the-flow>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From jblazi at netsurf.de  Thu Jan 20 10:39:28 2000
From: jblazi at netsurf.de (Janos Blazi)
Date: Thu, 20 Jan 2000 16:39:28 +0100
Subject: newbie question on modules
Message-ID: <38872a70_4@goliath.newsfeeds.com>

I have a module a in which a function m_add is defined. Now I want (as the
last statement of module a) load and execute a module b. Now it seems that b
does not know the function m_add. How can I change that?

I'd like to access everything defined in a in the module b that is the
(variable) second part of a.

What can I do?

Janos Blazi




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----


From stadt at cs.utwente.nl  Wed Jan  5 07:58:29 2000
From: stadt at cs.utwente.nl (Richard van de Stadt)
Date: Wed, 05 Jan 2000 13:58:29 +0100
Subject: Dr. Dobb's Python-URL! - weekly Python news and links (Dec 21)
References: <7D99E9F5B20B4BA9.D1FD33F5C753DF5C.FCBA193C7D9DF095@lp.airnews.net>
Message-ID: <38733FF5.6BFC06A5@cs.utwente.nl>

Gordon McMillan wrote:
> 
>     People interested in optional static typing in Python should
>     look at the types SIG
>         http://www.python.org/pipermail/types-sig/
>     which is experiencing a flood of activity,  (including
>     unauthorized use of Guido's time-machine).

How come there are messages from April 2000 in the archive?

Richard.


From tiddlerdeja at my-deja.com  Sat Jan 22 15:03:14 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Sat, 22 Jan 2000 20:03:14 GMT
Subject: COM and CreateObject and inheritence
Message-ID: <86d2dr$a9b$1@nnrp1.deja.com>

I managed to get the HelloWorld COM Python example to work using
CreateObject in VB. I could not get a Reference to Python.TestServer so
that I could:

Dim Test and HelloWorld

Instead I had to:

Dim Test As Object

Is it not possible to get a reference to it?

Also, I tried to create my own test:

foo.py
-----

class foo:
    def __init__(self):
        print 'init'

    def bar(self):
        return 99

comfoo.py

from foo import *
class comfoo(foo):
    _reg_clsid_ = "{3F90DE60-D0F5-11D3-96E4-9A059544BB30}"

    _reg_progid_ = "My.Test"

    _reg_desc_ = "COM Test"

    _public_methods_ = ['bar']


    def __init__(self):
        print 'cominit'


When I try to call 'bar' in VB it crashes. comfoo is registered and both
files are in my python path.


Can someone tell me what I'm doing wrong?

Any help appreciated.


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


From neelk at brick.cswv.com  Wed Jan  5 19:38:17 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 6 Jan 2000 00:38:17 GMT
Subject: Simple cartesian product
References: <iacpuvhqwdp.fsf@vier.idi.ntnu.no> <iacn1qlqvw7.fsf@vier.idi.ntnu.no> <iack8lpqvoi.fsf@vier.idi.ntnu.no>  <iachfgtqvj9.fsf@vier.idi.ntnu.no>  <slrn8755l1.gvl.neelk@brick.cswv.com>
Message-ID: <slrn877ovr.hbv.neelk@brick.cswv.com>

Neel Krishnaswami <neelk at brick.cswv.com> wrote:
> 
> Here's my entry:
> 
> def product(a, b):
>     u = map(lambda i: a[i/len(a)], range(len(a)*len(b)))
>     v = map(lambda j: b[j%len(b)], range(len(a)*len(b))) # Same as b*len(a)
>     return map(None, u, v)

I realized this can be condensed even further:

def product(a, b):
    return map(lambda n,a=a,b=b: (a[n/len(b)], b[n%len(b)]),
               range(len(a)*len(b)))

It's still slower and more memory-hungry than the obvious for-loop,
though.


Neel





From kc5tja at garnet.armored.net  Tue Jan 11 18:04:03 2000
From: kc5tja at garnet.armored.net (Samuel A. Falvo II)
Date: 11 Jan 2000 23:04:03 GMT
Subject: XML-RPC Server
References: <85faal$fg4$3@newsserver.rrzn.uni-hannover.de>
Message-ID: <slrn87ne3i.qla.kc5tja@garnet.armored.net>

>I know it should be hard to do, someone must have done it.
>Just a couple of simple objects answering towards XML-RPC requests.

In fact, it's insanely hard to do.  I can't tell you the tears I had to go
through just to download the dang thing.  ;-)

Look for the "xmlrpclib" package for Python.  You can get to it through the
www.xmlrpc.com website.  That package includes the client software as well
as two types of servers: one for Medusa, and one for the out-of-the-box
SocketServer package.  It takes a little bit to get used to, but not much --
you can have a fully functional XML RPC server set up in minutes.

>And I also appreciate and update on SOAP client and server python
>implementation efforts.

Indeed -- I too would like more information on this.  I checked out
www.develop.com, since they seem to be the biggest mover and shaker in the
SOAP world, but I haven't found anything just yet.  Maybe I'm just not
looking in the right areas?

-- 
KC5TJA/6, DM13, QRP-L #1447
Samuel A. Falvo II
Oceanside, CA


From sei at sunse.jinr.ru  Wed Jan 19 16:22:43 2000
From: sei at sunse.jinr.ru (Evguenii Smogailov)
Date: Thu, 20 Jan 2000 00:22:43 +0300 (MSK)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <38862948.5AEE964C@roguewave.com>
Message-ID: <Pine.SOL.3.91.1000120002040.3347A-100000@sunse>

On Wed, 19 Jan 2000, Bjorn Pettersen wrote:
> > On 19 Jan 2000, Stephen Coursen wrote:
> > > When did this become a trait of Perl?  Gods, what's so hard about knowing
> > > that == is the equality operator?
> >
> > Well, just the same reason as for replacing '||', '&&' with 'or', 'and'.
> 
> I see... so you don't really want eq, gt, and ne; you want equal, greater, and
> not equal.

Yes, something like this;

evguenii




From charliehubbard76 at yahoo.com  Wed Jan  5 11:14:59 2000
From: charliehubbard76 at yahoo.com (charliehubbard76 at yahoo.com)
Date: Wed, 05 Jan 2000 16:14:59 GMT
Subject: Portable Authentication? WinNT and UNIX
Message-ID: <84vqlj$2e8$1@nnrp1.deja.com>

Hello,

I'm writing a simple, portable server for both WinNT and Unix in python,
and I need to authenticate my users with the system users and passwords.
I know that Unix provides these mechanisms, but what about WinNT?  Does
anyone know how I can authenticate WinNT users thru Win32 api?

It's important that clients encrypt the password and login before
sending it across the network so any info on how I could do that within
python would be helpful.

thanks
charlie


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


From aahz at netcom.com  Tue Jan 18 18:17:12 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 18 Jan 2000 23:17:12 GMT
Subject: **kwargs
References: <1263886656-955174@hypernet.com>
Message-ID: <862s9o$12v$1@nntp6.atl.mindspring.net>

In article <1263886656-955174 at hypernet.com>,
Gordon McMillan <gmcm at hypernet.com> wrote:
>
>BTW, while foo-abuse is common, I've never seen it carried to 
>such extremes.

As the saying goes, "If the foo shits...."
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From moeller at networksplus.net  Sun Jan 23 01:20:19 2000
From: moeller at networksplus.net (Derek Moeller)
Date: Sun, 23 Jan 2000 00:20:19 -0600
Subject: Small `du' reformatter :-)
In-Reply-To: <oqk8l1ef4a.fsf@titan.progiciels-bpi.ca>; from pinard@iro.umontreal.ca on Sat, Jan 22, 2000 at 08:44:21PM -0500
References: <oqogagfidt.fsf@titan.progiciels-bpi.ca> <8766womanm.fsf@baal.winnegan.de> <oqk8l1ef4a.fsf@titan.progiciels-bpi.ca>
Message-ID: <20000123002019.A27888@networksplus.net>

On Sat, Jan 22, 2000 at 08:44:21PM -0500, Fran?ois Pinard wrote:
> Siggy Brentrup <bsb at winnegan.de> ?crit:
> 
> Yes, I took a look, thanks!
> 
> May I offer to replace the distributed script with mine?  You will get a
> script which is shorter and more elegant Python.  Its output might also
> be a bit better pruned and sorted, in my opinion.  Before:
> 
>         172 .
>           |  21 src
>           |  16 lib
>           |  10 data
>           |   |     4 .deps
> 
> After:
> 
>             172 \_ .
>              10    \_ data
>               4    |  \_ .deps
>              16    \_ lib
>              21    \_ src
> 

I must say, it is quite a bit nicer. Requires most recent version of
python, I believe, but considering that it would ship with it it's not a
problem.

-- 
Derek Moeller
  moeller at networksplus.net
  http://dophnic.yi.org



From fredrik at pythonware.com  Wed Jan 19 15:58:42 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 19 Jan 2000 21:58:42 +0100
Subject: Python Rocks!
References: <20000119152127.A6919@foxboro.com> <8654iq$fcl@news.or.intel.com> <38861CA6.F99F7C65@callware.com>
Message-ID: <007c01bf62bf$fa5c1ec0$f29b12c2@secret.pythonware.com>

Ivan Van Laningham <ivanlan at callware.com> wrote:
> You're saying to people who know and love Python, "Python is a great
> language except that it's Python.  Let's fix it."  We don't think it's
> broken.  Your insistence on swimming upstream on this issue is not only
> wrong-headed but boring.

oh, you mean I was the only one here who found TYE4's
last post (the one you replied to) to be among the most
hilarious things I've ever seen on this newsgroup?

    "immense amount of effort"
    "not a single programming language"
    "thousands of programmers have a problem"

gotta be living on another planet [1], that guy...

</F>

1) or maybe not:
http://www.uib.no/aasland/Swi-Complex.Html
(I suddenly realize why pressing "tab" requires an
immense amount of effort for TYE4 aka. SNF5 ;-)




From darrell at dorb.com  Sun Jan 30 13:49:29 2000
From: darrell at dorb.com (Darrell)
Date: Sun, 30 Jan 2000 13:49:29 -0500
Subject: Stackless Python 1.0 + Continuations 0.6
References: <mt2.0-27365-948450827@news.informatik.uni-bonn.de> <86f8uu$3ds$1@kopp.stud.ntnu.no> <388BDEFA.8D57A66F@tismer.com> <388C91E1.500B272A@hassler.de> <388EB066.3DACCDEE@tismer.com>
Message-ID: <012001bf6b52$bf5d3430$0100a8c0@rochester.rr.com>

"""
Great job Christian.
It sure can speed up deep recursive functions.

I'd like to see what others have figured out about how to use
Stackless.

What are the chances this will make it into python 1.6 ?

The following is excerpts from the Stackless distribution.
With some comments and mods to help me understand.
Let me know if I'm way off or if this should have been an attachment.
"""
import continuation, time

get_caller  = continuation.caller

def nS2(n):
    """
    Not limited by stack depth.
    """
        # Causes a GP if no parms passed.
        # If given a 1 cont would point up on in the call stack/chain ?
    cont=get_caller(0)
        # I feel an ignorant statement coming on.
        # Using "cont" instead of "call" is an optimization.
        # cont assumes that when called it doesn't have to
        # return to it's caller. So a deep recursion doesn't have
        # to keep a stack to return to each level. The final return
        # at some depth, is the final returned value.
    call=cont.call
        # Each call to cont arrives at this point.
        # And k is assigned to the current value of n with update(n)
        # If you had used n it would always be the original value.
    k=cont.update(n)
    if k:
        # This won't return so "print msg" is never called
        # if "call(k-1" had been used it would have printed the msg.
        cont(k-1)
        msg="""
        This isn't printed when cont is used. Using cont
        doesn't experience stack depth limits.

        This is printed when call is used.
        Using call runs slower and has the same limits
        as a normal recursive function.
        """
        print msg
    else:
        del cont.link
    return k

def s1(n):
    """
    Limited by stack depth
    """
    if n:
        n=s1(n-1)
    return n


def timer(n,func):
    t1=time.time()
    for x in range(1000):
        func(n)
    print time.time()-t1

if 0:
    """
    Stackless is much faster for the right recursive function
    """
    timer(100,nS2)
    timer(100,s1)


# ======================================================================
#         coroutine
# ======================================================================
"""
This was hard to understand.
"""

class coroutine:

    current = None
    caller_of_this = continuation.caller

    def __init__ (self, f, *a, **kw):
        self.state = lambda v,f=f,a=a,kw=kw: apply (f, a, kw)
        self.caller = None

    def resume (self, value=None):
        caller = coroutine.current
        caller.state = self.caller_of_this()
        self.caller=caller.state
        self.caller = caller
        coroutine.current = self
            # If you tracing this call will runaway.
        self.state (value)

    def kill (self):
        self.__dict__.clear()


def resume_caller (value):
    me = coroutine.current
    me.caller.resume (value)

def resume_main (value):
    _main.resume (value)

_main = coroutine (None)
coroutine.current = _main

if __name__ == '__main__':
    # counter/generator

    def counter (start=0, step=1):
        n = start
        while 1:
            print 'A:',n
            resume_caller (n)
            n = n + step
            print 'B:', n


    """
    It's surprising that cc.resume() returns a value.
    Guess it's related to the lambda in the __init__ ??
    Need an experiment.
    Prints:
    A: 0
    0
    B: 1
    A: 1
    1
    B: 2
    A: 2
    2

    The second resume takes us back to the "print 'B:'"
    Then back to the "print A:"
    It's as though the counter functions are in separate threads.
    And resume or resume_caller is a blocking message pass.
    """
    cc = coroutine(counter)
    print cc.resume()
    print cc.resume()
    print cc.resume()


        # same-fringe
    def _tree_walker (t):
        if type(t) is type([]):
            for x in t:
                _tree_walker (x)
        else:
            """
            Passes control back to same_fringe
            tree_walker and _tree_walker are in the same pseudo thread.
            """
            resume_caller (t)

    def tree_walker (t):
        _tree_walker (t)
        print 'Returns None causing the caller to exit'
        resume_caller (None)
        print 'Never prints'

    def same_fringe (t1, t2):
        co1 = coroutine (tree_walker, t1)
        co2 = coroutine (tree_walker, t2)
        while 1:
            leaf1 = co1.resume()
            leaf2 = co2.resume()
            """
            tree_walker searches finds a result and calls resume_caller(t)
            which passes back the result. It's printed then cox.resume()
            continues the search. tree_walker is waiting to continue the
            search.
            """
            print 'leaf1: %s leaf2: %s' % (leaf1, leaf2)
            if leaf1 == leaf2:
                if leaf1 is None:
                    return 1
            else:
                return 0

    print same_fringe (
    [0,1,[2,3],[4,[5,[6]],7,[[[[8]]]],9]],
    [[[0,1],2,[3,[4],[5]],[[6],[7]],8],9]
    )






From jfzhou at 188.net  Mon Jan 31 00:40:57 2000
From: jfzhou at 188.net (Öܹ«)
Date: Mon, 31 Jan 2000 13:40:57 +0800
Subject: Why Perl do But Python can't?
References: <200001310128.UAA02858@poverty.bloomington.in.us>
Message-ID: <8737c1$76d$1@news.cz.js.cn>

Thank you help.
But How I do it to solve the follow quertion:

1:through the WWW to manage user,the CGI need root uid.
2:how to read or write files that need some privilege ie. root ,admin etc...

some one tell me
1:use another programme to communite with CGI useing socket,but it is
complex.
2:use expect mod ,but python dosn't have.

is there better solution?


Kelly Lynn Martin wrote in message
<200001310128.UAA02858 at poverty.bloomington.in.us>...
>On Mon, 31 Jan 2000 08:48:09 +0800, "????" <jfzhou at 188.net> said:
>
>>Someone know the difference of Perl and Python in the CGI running
>>mode?
>
>Scripts (of any sort) are inherently unsafe when run setuid or setgid.
>See the perlsec manpage for why.
>
>Kelly
>




From sp00fD at yahoo.com  Mon Jan 24 10:12:03 2000
From: sp00fD at yahoo.com (sp00fD)
Date: Mon, 24 Jan 2000 15:12:03 GMT
Subject: stupid FORM problem
Message-ID: <86hq3u$gmu$1@nnrp1.deja.com>

I'm having a problem with a cgi program.  Basically some html calls a
<FORM METHOD="POST" ACTION="/my.cgi?action=view&table=thistable">
...
<SELECT NAME="thread">
<OPTION VALUE="1">foo
....

None of my params {action : "view", table : "thistable", thread : 3(or
whatever)} go through.  I've tried this using GET as the METHOD as
well, but no luck.  Here's the thing though, if I type the path in as
the url, it works (i.e http://myhost/my.cgi?
action=view&table=thistable&thread=3)

What am I doing wrong.

This POST goes to a "traffic cop" that creates an instance of a class
depending on the action, and passes the params, then displays the
output of that class instance.

Thanks,


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


From brunomadv at ciudad.com.ar  Mon Jan  3 19:48:44 2000
From: brunomadv at ciudad.com.ar (Bruno Mattarollo)
Date: Mon, 3 Jan 2000 21:48:44 -0300
Subject: CGI Script using Python
In-Reply-To: <D1bc4.40$hJ1.1161@news.rdc1.tx.home.com>
Message-ID: <LPBBIMGIKCBKOFOCBMODGENDCJAA.brunomadv@ciudad.com.ar>

Hello.

	Well, I have tryied the same script on my Solaris with Apache 1.3.9 and Python 1.5.2 and it worked. Does it work running it from the shell command line?

	Anyway I think (but I am not sure) that it's not valid HTML... But it has nothing to do with the error you are reporting...

	Hope this helps.


/B

Bruno Mattarollo <brunomadv at ciudad.com.ar> 

-----Mensaje original-----
De: python-list-admin at python.org [mailto:python-list-admin at python.org]En
nombre de Paul E. Bible
Enviado el: Monday, January 03, 2000 9:21 PM
Para: python-list at python.org
Asunto: CGI Script using Python


Hi there,

I'm currently developing a CGI script using Python under Apache Web Server.
I've written a simple script, placed it in the /cgi-bin subdirectory and
perform a chmod 755 on it.  Here is the script:

-----------------------------------------------------------
#!/usr/local/bin/python
import os, string, sys
from cgi import *

def Main():
    print "Content-type: text/html\n\n"
    print "<HTML>"
    print "<HEAD>"
    print "This is a test"
    print "</HEAD>"
    print "<BODY></BODY>"
    print "</HTML>"
    sys.exit(0)

Main()
-----------------------------------------------------------

When the above script is executed on Apache, I get the following message:

-----------------------------------------------------------
500 Internal Server Error
Internal Server Error
The server encountered an internal error or misconfiguration and was unable
to complete
your request.

Please contact the server administrator, webmaster at domain.com and inform
them of the time the error occurred,
and anything you might have done that may have caused the error.
More information about this error may be available in the server error log.
-----------------------------------------------------------

The Apache error log indicates a 'Premature end of script headers:
/usr/local/apache/cgi-bin/test.py'.

My platform is:  Redhat Linux 6.2, Apache Web Server 1.3.9, Python 1.5.2

Any help with this matter would be greatly appreciated.

Thanks in advance,
Paul

----------------------------------------
Paul E. Bible
pbible at littlefishsoftware.com
www.littlefishsoftware.com
----------------------------------------





-- 
http://www.python.org/mailman/listinfo/python-list

From skip at mojam.com  Tue Jan 25 15:46:43 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 25 Jan 2000 14:46:43 -0600 (CST)
Subject: Multidimensional Dictionaries: Possible?
In-Reply-To: <388DE028.ED24A3B1@nipltd.com>
References: <slrn88ppsg.49t.w6uv@voyager.home.com>
	<oqpuuqx6s1.fsf@titan.progiciels-bpi.ca>
	<388DE028.ED24A3B1@nipltd.com>
Message-ID: <14478.2995.807755.217066@beluga.mojam.com>

    Chris> What's the quickest way to do the equivalent of the following
    Chris> Perl in Python, again, using the above example:

    Chris> $dict{$key1}{$key2}++

    Chris> I can only think of the following, I hope there is a better way!:

    Chris> a =  dict[key1, key2]
    Chris> a = a + 1
    Chris> dict[key1, key2] = a

Well, you could skip the intermediate variable:

    dict[key1, key2] = dict[key1, key2] + 1

and/or shorten the key reference up a tad:

    k = (key1, key2)
    dict[k] = dict[k] + 1

Depends on what you're after.

Also, I didn't see the entire thread, so perhaps this was already covered,
but from what little I know of Perl, "$dict{$key1}{$key2}" references a
nested dict sort of thing.  If so, the Python construct "dict[key1, key2]"
is not the same thing.  It references a single dictionary with a key that's
the tuple (key1, key2), e.g.:

    >>> dict = {}
    >>> dict[1, 2] = 'foo'
    >>> dict.keys()
    [(1, 2)]

If what you really want are nested dicts, then you should use something like

    >>> dict = {}
    >>> dict[1] = {}
    >>> dict[1][2] = 'foo'
    >>> dict.keys()
    [1]
    >>> dict[1].keys()
    [2]

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098


    



From tlu at bmo.com  Thu Jan 20 10:40:45 2000
From: tlu at bmo.com (Tao Lu)
Date: Thu, 20 Jan 2000 10:40:45 -0500 (EST)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
Message-ID: <200001201540.KAA16944@trm.bmo.com>

> 
> >
> > Yes, i'm saying that 'eq', 'ne' are more readable then '==' '!=', '<>',
> > 'next' shorter then 'continue' and '..' operator can be a good thing to 
> > consider.
> > 
> 	Yet after over 30 years of ".ne.", ".eq.", ".gt.", etc. in
> FORTRAN, Fortran 90 has just added "<>", "=" (or is it "=="), ">", etc.
> to permit one to get away from those awful text strings.
> 
> 	The only languages I've ever seen that use a text string for
> those operators are FORTRAN and VMS DCL.
> 

FYI: Another P* language also uses 'eq', 'ne' as operators. However, 'eq' and 
'==' are not the same in this P*** language: 'eq' for string and '==' for 
numbers.


Cheers,
_______________________________________________________________________
<name>Tao Lu</name><email>tlu at bmo.com</email><phone>867-6722</phone>






From skip at mojam.com  Tue Jan 18 11:24:19 2000
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 18 Jan 2000 10:24:19 -0600 (CST)
Subject: DBI spec and parameter passing
In-Reply-To: <1263967432-11066499@hypernet.com>
References: <3883A0CA.AB0D4854@exceptionalminds.com>
	<1263967432-11066499@hypernet.com>
Message-ID: <14468.37811.321605.587902@beluga.mojam.com>


    >> Instead of doing the parameter interpolation yourself, try letting
    >> your database module do it for you, e.g.:
    >> 
    >> c = db.cursor()
    >> c.execute("""select id from city where"""
    >> """ city = %s and"""
    >> """ state = %s and"""
    >> """ country = %s""", (city, state, country))

    Gordon> but Skip _meant_:

    Gordon> c.execute(""" select id from city where city = ? and state = ?
    Gordon> and country = ?""", (city, state, country))

I suppose so, in the correct context.  Not being an ODBC person I didn't
know what was required in the context of the original question, so all I
could do was cut/paste from my MySQLdb code, which wants the %s thingies...

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/
847-971-7098




From rjroy at takingcontrol.com  Sun Jan 16 11:36:10 2000
From: rjroy at takingcontrol.com (Robert Roy)
Date: Sun, 16 Jan 2000 16:36:10 GMT
Subject: Breaking out of walk()
References: <38811854.F2BC197E@rl.gov>
Message-ID: <3881f20c.844046140@news1.on.sympatico.ca>

On Sat, 15 Jan 2000 17:01:08 -0800, "Roberts, Robert J"
<ROBERT_J_ROBERTS at rl.gov> wrote:

>Is there a way to gracefully break out of os.path.walk()?
>
names[:]=[]

Make sure you modify the list in-place. Don't do names=[] since that
creates a reference to a different object. You can also choose not to
recurse through certain directories.

if 'cvs' in names: names.remove('cvs')

will skip over directories named 'cvs'




From sposhua at my.pc  Mon Jan 10 12:41:56 2000
From: sposhua at my.pc (Sposhua)
Date: Mon, 10 Jan 2000 17:41:56 +0000
Subject: Filenames of cgi-uploaded files
Message-ID: <Pine.WNT.4.21.0001101735310.249-100000@wks_3.e-shopwork.com>

Newbie CGI script...

Uploading a file from a form - I need to know the filename (to ignore all
non-gif/jpg uploads) and filesize. Can't find anything in the tutorials on
theis :-( Then I guess I upload the pic using

form = cgi.FieldStorage
fileitem = form['picture']
if fileitem.file:
	pic_upload = fileitem.file.read()

(adapted from the tutorial, though it doesn't make much sense to me - expecially
the fileitem.file bit)



From che at debian.org  Fri Jan 21 15:57:50 2000
From: che at debian.org (Ben Gertzfield)
Date: 21 Jan 2000 12:57:50 -0800
Subject: Bug in Calendar Module
References: <22E71DAEC504D111B78100805FFE9DC72F03CAD4@pfs21.ex.nus.edu.sg>
Message-ID: <yttaelz2ldd.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Liew" == Liew Pol Yee <polyp at cyberway.com.sg> writes:

    Liew> Hello! Does a fix for the Calendar Module calendar.py exist? 
    Liew> Currently, i get errors when I do the following: import
    Liew> calendar new = calendar.monthcalendar(2038,2)

    Liew> It seems to be unable to accept dates after January,
    Liew> 2038. Thanks for any help!

This is because you are on a system where the time is stored as a
32-bit integer, as the number of seconds since 1970.

In 2038, the number of seconds since 1970 overflows the 32-bit limit,
and a lot of programs will break.

There isn't a good solution, except to use a system that has a 64-bit
counter for time.

Ben

-- 
Brought to you by the letters F and Q and the number 5.
"It makes my nipples hard!"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/


From dheise at debitel.net  Sat Jan 29 21:48:57 2000
From: dheise at debitel.net (Dirk Heise)
Date: 29 Jan 2000 21:48:57 MEST
Subject: Imagemagick
References: <wzzotoal2k.fsf@sunshine.cs.uu.nl> <20000129201210.A4953@stopcontact.palga.uucp>
Message-ID: <01bf6a99$fdfe3860$LocalHost@dreadzone>


Gerrit Holl <gerrit.holl at pobox.com> schrieb im Beitrag <20000129201210.A4953 at stopcontact.palga.uucp>...
> Das waere leicht.
Das war leicht.

Tschuess

Dirk



From josht at iname.com  Sun Jan  9 15:15:51 2000
From: josht at iname.com (Josh Tompkins)
Date: Sun, 09 Jan 2000 20:15:51 GMT
Subject: Modifying Global Variables from Within a Function
References: <0ce801bf5adb$9eaf75e0$01ffffc0@worldnet.att.net>
Message-ID: <8EB692B19joshtinamecom@209.208.0.7>

That worked perfectly.  Thanks alot for your help.
 
Josh "Josue" Tompkins
___________________________________________________________________________
____

E-Mail:  josht at iname.com ICQ:  21219667 Web:  
http://www.crosswinds.net/~josht
___________________________________________________________________________
____


From fredrik at pythonware.com  Mon Jan 17 09:52:38 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 17 Jan 2000 15:52:38 +0100
Subject: Python -> EXE?
References: <38831BE5.76B189A8@mchp.siemens.de>
Message-ID: <002801bf60fa$80b4b440$f29b12c2@secret.pythonware.com>

Christian Schaller <Christian.Schaller at mchp.siemens.de> wrote:
>   Is there a tool similar to Perl2Exe that transforms a Python program
> to a MS .EXE programc that can be executed without installing Python on
> a machine?

    http://www.python.org/doc/FAQ.html#4.28
    "How can I create a stand-alone binary from a Python script?"

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! -->




From matt at resonation.co.uk  Sat Jan 22 22:54:23 2000
From: matt at resonation.co.uk (Matthew Russell)
Date: Sun, 23 Jan 2000 03:54:23 -0000
Subject: How to differentiate Win98 from NT ?
References: <2B1262E83448D211AE4B00A0C9D61B03BA7070@MSGEURO1>
Message-ID: <86du41$lv0$1@gxsn.com>

try :
    if os.environ['windir']=="C:\\winnt":
            # do whatever
    else:
            ....




From aahz at netcom.com  Sat Jan  1 16:56:00 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Jan 2000 21:56:00 GMT
Subject: Python not a Very High-Level Language?
References: <000001bf54a1$805e6140$c52d153f@tim>
Message-ID: <84lt5g$rrs$1@nntp9.atl.mindspring.net>

In article <000001bf54a1$805e6140$c52d153f at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>Combine that with generators and list comprehensions, and you'd be amazed at
>what you can accomplish in three lines of code.  Indeed, it may take you
>three weeks to figure out what three lines of code do <0.9 wink>.

Yup.  That's what I think is fundamentally wrong about such
functionality rants.  From my POV, the way to productivity lies with
simple expression of complex ideas, not ways of creating complex
expressions.  Unless one is working alone or can enforce training,
complex idioms detract from productivity.
--
                      --- Aahz (@netcom.com)

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

TEOTWAWKI -- today!


From johnjensen at my-deja.com  Thu Jan  6 20:47:11 2000
From: johnjensen at my-deja.com (John Jensen)
Date: Thu, 06 Jan 2000 17:47:11 -0800
Subject: XML to dictionary?
Message-ID: <FGALACFJFLICBBAA@my-deja.com>

Yes, I recognize that the keys have to be unique.  The sample I provided isn't a "real" one -- the data I am actually working with is kind of ugly.  In the data I'm working with the child tags are always unique, so it would be great for this kind of solution.

I'm thinking that the best thing to do in this situation is just write a quick customized script, and not worry about using "fancy" libraries like xmllib, pyxie, etc.

JOhn

 
--

On Thu, 06 Jan 2000 16:09:17   jim kraai wrote:
>There's a little problem:
>  dictionary keys must be unique.
>  XML can have more than one child node of the same name.
>
>Depending on the DTD:
>	<Newsgroup>
>	  <comp.lang.python>
>	    <Message>
>	      <Subject>Hi there</Subject>
>	      <Body>This is a test</Body>
>	    </Message>
>	    <Message>
>	      <Subject>Re: Hi there</Subject>
>	      <Body>This is a reply to:\n> This is a test</Body>
>	    </Message>
>	  </comp.lang.python>
>	</Newsgroup>
>is a valid hunk of xml.
>
>If you're controlling this aspect of the DTD, that's fine, but I 
>don't know how you'd write a DTD to encompass:
><Newsgroup>
>	<Message1>...</Message1>
>	<Message2>...</Message2>
>	<Message...>...</Message...>
>	<MessageN>...</MessageN>
></Newsgroup>
>
>This is a 'bad idea' in XML, but then again, I haven't written 
>that many DTDs.
>
>--jim
>
>johnjensen at my-deja.com wrote:
>> 
>> I have a relatively simple XML question, but I can't seem to figure out
>> how to do it simply and easily with the xmllib module.
>> 
>> Assume I have some XML (stylized below):
>> 
>> <Newsgroup>
>>   <comp.lang.python>
>>     <Message1>
>>       <Subject>Hi there</Subject>
>>       <Body>This is a test</Body>
>>     </Message1>
>>   </comp.lang.python>
>> </Newsgroup>
>> 
>> How can I quickly and easily stuff this into a nested dictionary, so
>> that:
>> 
>> print dict['Newsgroup']['comp.lang.python']['Message1']['Subject']
>> spits out 'Hi there'
>> 
>> The XML I am working with is perfect for this type of solution.
>> 
>> Any help would be appreciated!
>> 
>> John
>


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



From hei at adtranzsig.de  Fri Jan 28 10:25:53 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Fri, 28 Jan 2000 16:25:53 +0100
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be> <38918A29.8AB61742@electricorb.com>
Message-ID: <949076889.194238@news.adtranzsig.de>

Glen Starchman schrieb in Nachricht <38918A29.8AB61742 at electricorb.com>...
>Personally, there is nothing I hate more in the curly-brace languages than
the fact
>that developers don't have to line up their braces to show groupings of
code. I
>guess if every C/C++/Java, etc... developer wrote code like:
>
>for (i=0; i < 100; i++)
>{
>    do something;
>} /* notice how the braces line up and the body is indented!*/
>
>I would be happy with that approach. However, there is nothing better than
forcing
>clean code... in my very humble opinion.

What bothers me even more, they don't mind writing functions without the
function comment headers that shall be used in the project.
You end up with functions containing not a single char of comment.
It would be another improvement to Python if it would force you to
write a doc string for each function of method.
--
Dipl.Inform. Dirk-Ulrich Heise
hei at adtranzsig.de
dheise at debitel.net




From moonseeker at my-deja.com  Tue Jan 11 05:21:29 2000
From: moonseeker at my-deja.com (moonseeker at my-deja.com)
Date: Tue, 11 Jan 2000 10:21:29 GMT
Subject: shelve looking for classes?
Message-ID: <85f071$qdt$1@nnrp1.deja.com>

Hi!

If I try to recreate an object saved with shelve (maybe with pickle too)
shelve looks for the class of the object in __main__. Thats not too good
because I have to keep all classes in __main__ which doesn't help me to keep
my program as modular as possible.

Anyone a suggestion how to make shelve looking somewhere different for the
class?

Thanks,
Mike


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


From fdrake at acm.org  Mon Jan 31 16:04:54 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 31 Jan 2000 16:04:54 -0500 (EST)
Subject: Dynamic class construction?
In-Reply-To: <001301bf6c16$81e08ac0$3acbd9c2@peridot.optichrome.com>
References: <3895B829.39067306@math.okstate.edu>
	<001301bf6c16$81e08ac0$3acbd9c2@peridot.optichrome.com>
Message-ID: <14485.63734.974901.109872@weyr.cnri.reston.va.us>

Adrian Eyre writes:
 > Ah. Be aware this module was only introduced in 1.5.2 IIRC.

  The "new" module has been there for quite a whilte, but the
documentation *is* fairly new.  Thanks to Moshe Zadka for that!


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From fdrake at acm.org  Tue Jan  4 12:58:32 2000
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Tue, 4 Jan 2000 12:58:32 -0500 (EST)
Subject: exception arg
In-Reply-To: <slrn8749le.d0.mn@banger.iol.it>
References: <slrn8749le.d0.mn@banger.iol.it>
Message-ID: <14450.13512.821472.996954@weyr.cnri.reston.va.us>

Mirko Nasato writes:
 > >>> 1/0
 > Traceback (innermost last):
 >   File "<stdin>", line 1, in ?
 > ZeroDivisionError: integer division or modulo
 > >>> try:
 > ...     1/0
 > ... except ZeroDivisionError, arg:
 > ...     if arg == "integer division or modulo":
 > ...             print "This is what I expect."
 > ...     else:
 > ...             print 'Unexpected. "arg" is "%s".' % arg
 > ... 
 > Unexpected. "arg" is "integer division or modulo".
 > >>>
 > 
 > The same seems to occur with every kind of exception, so maybe
 > there's an explanation...

  'arg' is a object for which the __str__ method returns "integer
division or modulo".


  -Fred

--
Fred L. Drake, Jr.	  <fdrake at acm.org>
Corporation for National Research Initiatives



From thomas at xs4all.net  Fri Jan 28 11:49:41 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Fri, 28 Jan 2000 17:49:41 +0100
Subject: Installing quirk...
In-Reply-To: <389007A5.6B22105C@emw.ericsson.se>; from emwpeso@emw.ericsson.se on Thu, Jan 27, 2000 at 09:53:57AM +0100
References: <389007A5.6B22105C@emw.ericsson.se>
Message-ID: <20000128174941.B16093@xs4all.nl>

On Thu, Jan 27, 2000 at 09:53:57AM +0100, Per Soderstam wrote:

> I'm trying to build the Python1.5.2 release on a SPARC5 workstation
> running Solaris5 and i want another functionality for my pyhon promt
> (tsch type, with the up/down/left/right arrow positioning and command
> history as well as backspace instead of delete). Setting the SHELL
> (../tcsh) variable does not seem to work, but I'm no guru :). Do anyone
> understand my question? If so, do anyone have an answer?

If I understand your question properly, you want tcsh-style 'readline'
behaviour instead of bash-style ? Or do you mean you want readline behaviour
period, but you aren't getting any ? :)

If you're already using readline, and it's not what you want, I think you're
out of luck. As far as I know there is no drop-in replacement for readline
that makes it behave exactly like tcsh. On the other hand, you can probably
reconfigure readline's keybindings to behave _mostly_ like tcsh, and
providing your own completer function you can do most of tcsh's extremely
funky completion. But that last bit is probably a bit overkill for the
interactive interpreter.

The readline module still lacks good documentation, but you can learn the
most important bits from the readline(3) and the bash(1) manpage.

(If you're not using readline, that is, history/command-line editing don't
work, you need to recompile Python using readline, or compile it shared and
put it in your lib-dynload directory... See Modules/Setup in your python
source tree.)

tcsh-is-still-better-than-bash<wink>-ly y'rs
-- 
Thomas Wouters <thomas at xs4all.net>

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



From AKiss at GEOMETRIA.hu  Thu Jan 27 06:11:02 2000
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Thu, 27 Jan 2000 12:11:02 +0100
Subject: SourceForge question
Message-ID: <D148AB3A9D8FD311A99A008048C79D1104C254@MAIL>

Hi,

Dirk-Ulrich Heise <hei at adtranzsig.de> wrote in message
news:<948972527.962905 at news.adtranzsig.de>...
> Hi! I'm a clueless Windows user and i want to start a
> tiny project in SourceForge.
> I have a CVS binary and the files for my first release.
> I need: Advice on where to get a SSH binary for Windows

Try this:
http://www.doc.ic.ac.uk/~ci2/ssh/

Arpad

> and how to set up the initial CVS tree. It doesn't need
> to be perfect, this will not become a big team effort,
> i suppose. Somebody who already has set up a
> SourceForge project out there? Tanks for any
> advice.
> 
> --
> Dipl.Inform. Dirk-Ulrich Heise
> hei at adtranzsig.de
> dheise at debitel.net
> 
 



From nickliz at t-online.de  Sat Jan  1 14:14:28 2000
From: nickliz at t-online.de (Eide)
Date: Sat, 1 Jan 2000 20:14:28 +0100
Subject: Problems with Tkinter!
Message-ID: <84ljr3$htr$1@news00.btx.dtag.de>

Tk once worked, but does no longer.
I run W98 with python at C:\program files\python
and tcl at C:\program files\tcl

I tried the autoexec.bat fix as an earlier post suggested
at python.org to no avail.  Here is what I get when trying
the Tkinter._test()
--------------------------------------------------
Traceback (innermost last):
    File "<stdin>", line1, in ?
    File "C:\PROGRA~1\PYTHON\LIB\LIB-TK\Tkinter.py", line 1947, in _test
        root = Tk()
    File "C:\PROGRA~1\PYTHON\LIB\LIB-TK\Tkinter.py", line 886, in __init__
        self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
    {} ./lib/tcl8.0 C:/tcl8.0/library {C:/Program Files/library}

This probably means that Tcl wasn't installed properly.
--------------------------------------------------

My autoexec.bat reads:

PATH
C:\WINDOWS;C:\WINDOWS\COMMAND;C:\PROGRA~1PYTHON;C:\PROGRA~1\PYTHON\LIB;C:\PR
OGRA~\TCL\BIN

and I have tried every possible combination of "..\tcl\lib\" or
"..\tcl\lib\tcl8.0" etc and again, to no avail.  I also uninstalled it all,
cleaned my registry and reinstalled it.  I have no clue at this point.
Please advise.

Nick





From frank.sergeant at redneck.net  Mon Jan 10 20:58:44 2000
From: frank.sergeant at redneck.net (Frank Sergeant)
Date: 10 Jan 2000 19:58:44 -0600
Subject: Python not a Very High-Level Language?
References: <000801bf5802$dcca7d60$0aa0143f@tim> <is66x6vng7.fsf@localhost.ps.uni-sb.de>
Message-ID: <87u2kljrmz.fsf@den.home.net>

Denys Duchier <Denys.Duchier at ps.uni-sb.de> writes:

> As one of mozart's designers and implementors, as well as a python fan
> since the early days, I'd be happy to field questions, if there is any
> interest.

I think I saw that you have an interface to Tk.  Could you discuss
GUI building in Mozart a little?


  -- Frank
  frank.sergeant at pobox.com



From tiddlerdeja at my-deja.com  Sun Jan 30 11:23:15 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Sun, 30 Jan 2000 16:23:15 GMT
Subject: COM register ok. Unregister??????
Message-ID: <871ohi$tfl$1@nnrp1.deja.com>

Could someone tell me how I unregister my COM stuff in Python?

I used:
import win32com.server.register
win32com.server.register.UseCommandLine(COMBid)

To register it.


Any help appreciated.


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


From ynyhrb at w3r9oyas2b3.com  Sun Jan 16 04:02:46 2000
From: ynyhrb at w3r9oyas2b3.com (ynyhrb at w3r9oyas2b3.com)
Date: 16 Jan 2000 09:02:46 GMT
Subject: LEGAL C`A`B`L`E TV D`E-S`C`R`A`M`B`L`E`R  6869
Message-ID: <85s1fm$cb4$5304@ctb-nnrp2.saix.net>

LEGAL C`A`B`L`E TV D`E-S`C`R`A`M`B`L`E`R

Want to watch Sporting Events?--Movies?--Pay-Per-View??....FREE!!!!

*This is the Famous R-O Shack TV D-e-s-c-r-a-m-b-l-e-r 
You can assemble it from Radio Shack parts for about $12 to $15.

We Send You: 
1, E-Z To follow Assembly Instructions. 
2. E-Z To read Original Drawings. 
3. Total Parts List.

**** PLUS SOMETHING NEW YOU MUST HAVE! ****
Something you can't do without.

THE UP-TO-DATE 6 PAGE REPORT: 
USING A D-E-S-C-R-A-M-B-L-E-R LEGALLY

Warning: You should not build a TV D-e-s-c-r-a-m-b-l-e-r 
without reading this report first.

You get the complete 6 page report
and instruction package including
the instruction plans, the easy to 
follow diagram, and most important 
of all the "Using a D`e`s`c`r`a`m`b`l`e`r
LEGALLY Report all for just--$10.00 

Fill out form below and send it,
along with your $10.00 payment to:

C.a.b.l.e.t.r.o.n FREE-TV
12187 S. Orange Blossom Trail #116
Orlando Fl 32837
 
(Cash, Check or Money Order.)
(Florida residents include 7% Florida State Sales Tax)
(All orders outside the U.S.A. add $5.00)


PRINT YOUR:

NAME______________________________________________________

ADDRESS___________________________________________________

CITY/STATE/ZIP____________________________________________

E-MAIl ADDRESS____________________________________________ 

We are NOT ASSOCIATED in any way with RADIO SHACK. 
Neither the design nor instructions were developed 
by, are sold by, or are endorsed by Radio Shack. 
Parts for this fine-tuning device are available 
at many electronics stores (including Radio Shack) 
This is not a Radio Shack product.
fmkpnuvhulslxippitsydtoixiizhgwlvdyoqimrqtfnikqpwdfxmgnlqoepdnmjwrwtjjwyzrffbrhclxqdfwdvyjqlpfdnkekifhwvzcbsqpipxbivueecwpdqgsnpklvkrmfxpwyeqonwlkodukmwkevjnllgxjlfjkrvvxhrpxqozxqftcu



From cheselka at tucana.noao.edu  Thu Jan  6 10:33:25 2000
From: cheselka at tucana.noao.edu (Matt Cheselka)
Date: 6 Jan 2000 15:33:25 GMT
Subject: calling C functions from Python
Message-ID: <852ck5$7da$1@noao.edu>

Hello, 

I'm fairly new to Python and I'm wondering if there is a way to call C
functions from within python?  For instance (very simple) I have an object file
that contains a function to print "hello world" to the screen.  In the python
code I create a button to which I bind a call to this print function.  Does
something like this exist?

TIA,

Matt Cheselka
cheselka at noao.edu


From ionescu at pik-potsdam.de  Tue Jan 25 02:57:53 2000
From: ionescu at pik-potsdam.de (Cezar Ionescu)
Date: Tue, 25 Jan 2000 08:57:53 +0100
Subject: In defence of the two-namespace rule
References: <000601bf6452$c123ada0$caa0143f@tim>
Message-ID: <388D5781.79D885F8@pik-potsdam.de>

[Tim Peters]
> for all x.  Haskell has several shorthand notations for currying of various
> sorts.  For example,
> 
>     (- 1)
> 
> is the function that subtracts one from its argument, while
> 
>     (1 -)
>

Actually, (-1) is an exception since it refers to the negative integer
-1 (i.e. 0 - 1)

Of course, since the timbot is never wrong, this must have been a test!
Do I pass, huh, huh?

Best,
Cezar.


From helge at mdlink.de  Thu Jan  6 16:31:06 2000
From: helge at mdlink.de (Helge Hess)
Date: Thu, 06 Jan 2000 22:31:06 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <852jbk$9tt$1@nntp9.atl.mindspring.net>
Message-ID: <3875099A.2CF9333C@mdlink.de>

Aahz Maruch wrote:
> In article <3874CFE5.765A40EA at mdlink.de>,
> Helge Hess  <helge.hess at mdlink.de> wrote:
> >recently I was heavily wondering why __init__ does not allow return
> >values ! Is there any special reason not to do this ?
> 
> Yup.  The return value *has* to be the object instantiated by Python.

No, the return value has to be None, which is the default return value
as well. The only thing enforcing this from a technical point of view is
a simple

  if (ret != Py_None)
    throw exception

in the PyInstance_New() function. The 'fix' is just

  if ((ret == Py_None) || (ret == self))
    decref(ret), return self
  else
    decref(self), return ret

This doesn't break compatibility to any valid Python code.

> Why do you want to do this?

There are some applications all of which could be more or less solved by
using factories at the expense of a more complex & inconvenient
interface to the functionality (you have to learn two classes instead of
just one).

I can give just three:

1. singletons
  
  _C = None
  class C:
    def __init__(self):
      if _C is None: _C = self
      return _C
  a = C()
  b = C()

  a and b are the same objects, the user of C() doesn't need to know
about that

2. interns

  str2object = {}
  class String:
    def __init__(self, str):
      if not str2object.has_key(str):
          self.value = str
          str2object[str] = self
      return str2object[str]
  a = String("abc")
  b = String("def")
  c = String("abc")

  a and c will be identical.

3. class clusters (as they are called in Objective-C)

  class String: # abstract
    def __init__(self, data, encoding="python"):
      if encoding == "utf8": return Utf8String(data)
      elif encoding == "utf16": return Utf16String(data)
      elif encoding == "python": return data
      else: raise UnsupportedEncoding

  a = String("hhuu", encoding="utf16")
  v = String("sdfhjsh")

  The nice thing is that the user only has to care about String,
  the __init__ function will return an appropriate subclass on
  demand. What an 'appropriate subclass' is, is abstracted from
  the user pov.

My major feeling why __init__ should be allowed to return values
anyway is, that __init__ is a usual method and should be treated
as such.

Greetings
  Helge


From boud at rempt.xs4all.nl  Wed Jan 19 15:02:42 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 19 Jan 2000 20:02:42 GMT
Subject: DB API
References: <1263817020-5143711@hypernet.com>
Message-ID: <865592$obc$1@news1.xs4all.nl>

Gordon McMillan <gmcm at hypernet.com> wrote:

> Perhaps some bozo code "FOR UPDATE" in the select 
> statement? <sound of fingers breaking>. He won't do that 
> again.

> Actually, if the SQL is coded right (no ORDER BY or GROUP 
> BY) and the stuff in the middle is decent, the DB can be 
> returning rows before the query has even finished executing. If 
> the query is parameterized (pre-parsed) and simple, the 
> bottleneck will be the network and the load on the DB will be 
> insignificant. Unfortunately, that's a lot of "if"s...
>  

Well, all I can say - before this turns into a 'it is. No it ain't'
match, that I've been working with several fairly large databases
(above a million rows, above a thousand tables and with some quite
serious processing), and that according to my experience he best plan
is to do most of your processing as close as possible to the database
server - preferably inside. The network never has enough bandwidth,
and the client PC never has enough memory, at least in my experience.

I've seen some horrible GROUP BY's and ORDER BY's - on the other hand,
postprocessing the rows on the client can be impossible, making an ORDER
BY useful (especially if there's an ordered index to go by). Using
unparametrized queries in a production application is simply silly,
since it will overload Oracle's SQL area - I assume other rdbms's will
have an equivalent. I'm not sure what the stuff in the middle should
be - there are such a lot of possibilties, and even when you've got a
nice transaction manager, you'll find that the manager is trying out
his Cognos toy in daytime ;-).

You need a lot of space on the server if you want to handle large queries
- all the stuff will be stored someplace, even if the database can push
the first rows to the client before it's found the last rows (and that's
often not possible - complex joins can make that impossible) - but it
will in all probability sweep the buffer cache clean in the process,
making it inefficient after all. At least, that's what happened when
our last large application had an architecture where most queries
processed a megabyte of data. Forty clients brought the database to
a standstill.

I still think that a relation database is for getting small amounts of
compact data from a large (but well-indexed) mass of data. Dependent
on the nature of the application, the size of the server and the number
of concurrent users, a few thousand rows could be classified as a small
amount, but even on a large server, you wouldn't want hundreds of users
all demanding a few thousand rows every other moment. And I'm sure that
for handling large blobs other solutions are more suitable, like object
databases, but I haven't tried them.

Sure, it's possible that there is a need for data-slurping, but you'd
better be prepared for some questions about performance.

-- 

Boudewijn Rempt  | http://www.valdyas.org


From fredrik at pythonware.com  Thu Jan 20 07:17:13 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 20 Jan 2000 13:17:13 +0100
Subject: extended slice
References: <139104006C4@nds-3.azg.nl>
Message-ID: <004501bf6340$4b3bf2e0$f29b12c2@secret.pythonware.com>

P.J.W.S. Vrijlandt <P.J.W.S.VRIJLANDT at INT.azg.nl> wrote:
> In the reference manual under slice I find hints of an extende slice 
> syntax. It goes like a[2:3:step], a[2:4, 6:8] or a[... , 2:4].
> 
> Indeed 'ellipsis' is a builtin and seems to denote ...
> 
> However, I can find no way of using extended slicing. Do we have this 
> feature?

it's used by the numerical python extension:
http://sourceforge.net/project/?group_id=1369

as of 1.5.2, no built-in sequence types (tuples, list,
arrays, strings) supports the extended syntax.  hope-
fully, that will change in future versions.

</F>

<!-- (the eff-bot guide to) the standard python library:
http://www.pythonware.com/people/fredrik/librarybook.htm
&mdash; see it live on the python conference! &smiley; -->




From mikael at isy.liu.se  Thu Jan 20 09:54:47 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Thu, 20 Jan 2000 15:54:47 +0100 (MET)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <1263743270-9579629@hypernet.com>
Message-ID: <XFMail.000120155447.mikael@isy.liu.se>

On 20-Jan-00 Gordon McMillan wrote:
 >  Grant Edwards wrote:
 >  
 > > As long as we're at, I think we should try to make Python code
 > > not only more readable but also more polite. Rather than yet
 > > another imperitive language, the world needs more polite
 > > request languages!
 > > 
 > > I'm proposing the mandatory use of the keywords 'please',
 > > 'excuse_me', 'I'm_sorry' and 'thank_you'.  
 >  
 >  I have only one quibble. "Excuse me" is, in fact, imperative. 
 >  This is something that English hides, but becomes obvious in 
 >  German.
 >  
 >  Oh, and instead of "except please", it would be more polite to 
 >  say "thank you very much but please". And don't forget to 
 >  throw an exception unless sys.argv[1:5] == ['how', 'are', 'you', 
 >  'today?'].

Now we're talking!

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    20-Jan-00
Time:    15:54:23

This message was sent by XF-Mail.
-----------------------------------------------------------------------



From alex at magenta.com  Thu Jan  6 06:44:48 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 6 Jan 2000 12:44:48 +0100
Subject: Python and ASP
References: <_ji34.8719$523.343739@news.chello.at> <bBk34.1288$Cl3.6689@news-server.bigpond.net.au> <slrn84t6n0.q3j.aa126@fan1.fan.nb.ca> <BZA34.2308$Cl3.11615@news-server.bigpond.net.au> <387351CA.1EA77B30@gssec.bt.co.uk> <UDTc4.11594$lv5.51278@news-server.bigpond.net.au>
Message-ID: <008301bf583b$d2c98480$c02b2bc1@martelli>

Mark Hammond writes:


> Microsoft have defined an API called "Active Scripting", that allows
> applications to host scripting languages, without actually knowing the
> languages they are hosting. Microsoft provide a number of Active
> Scripting applications - ASP, IE4 and 5, WSH and one or 2 others. They

Further, they provide the "Microsoft Script Control", which makes
it trivially easy for ANY application to host any Active Scripting
compliant language (I have a reasonably small example _in C_
using the Script Control as a part of my Win32 programming
tutorial at http://aleax.supereva.it/ -- in Italian only, sorry --
I wouldn't normally use C to write *any* COM application, of
course, but I do in the tutorial for didactical reasons, and it's
hard to find any smaller, useful application of COM than hosting
said control...!-).


> Python supports active scripting as soon as you install
> win32all-???.exe. Then, any Active Scripting application 
> also supports Python.

...making Win32All+ScriptControl arguably the easiest way
of all to add Python scripting to your Win32 applications,
with the side benefit (?-) of easily allowing heathens to
use Perl, VB, or whatever, in lieu of Python!-)  You do have
to provide an Automation compliant object model, to let the
scripts reach back and do something to your app, but that's
reasonably easy with most development environments.


Alex





From grant at nowhere.  Tue Jan 11 10:47:33 2000
From: grant at nowhere. (Grant Edwards)
Date: Tue, 11 Jan 2000 15:47:33 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com> <005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <38774FBC.EB8281CF@maxtal.com.au> <slrn87k49l.pup.grant@grante.comtrol.com> <86so055xky.fsf@g.local>
Message-ID: <slrn87mk4l.1d1.grant@grante.comtrol.com>

In article <86so055xky.fsf at g.local>, Gareth McCaughan wrote:
>Grant Edwards wrote:
>
>> At least with the compilers I've used, that depends on the case
>> values.  If the switch case values are sparse, the code
>> generated is the same as it would be for If/elif.
>> 
>> For example:
>..
>> May generate table-lookup code that is O(1), but
>> 
>>   switch (c)
>>     {
>>     case      0: [...] break;
>>     case   3450: [...] break;
>>     case -49393: [...] break;
>>     case 233450: [...] break;
>>     case -14000: [...] break;
>>     case  -9834: [...] break;
>>     }
>>     
>> Would hopefully generate O(n) linear-search code.
>
>It ought to generate code that does something more
>like a binary chop. Looking at the output of gcc -O2
>on a big random switch, that's certainly what gcc
>does.

If it's big enough, then that's what it should do -- then you
can get O(log n) [I wonder how it handles cases that
fall-through...] For a small number of cases I wouldn't be
surprised if linear code performs better.  In either case, it's
not O(1) like a table-lookup.

-- 
Grant Edwards                   grante             Yow!  .. does your DRESSING
                                  at               ROOM have enough ASPARAGUS?
                               visi.com            


From jaap at zora.nl  Mon Jan 10 16:21:16 2000
From: jaap at zora.nl (jaap groot)
Date: Mon, 10 Jan 2000 21:21:16 +0000
Subject: Did you all see this??
Message-ID: <387A4D4C.3D1F8330@zora.nl>

lwn.net/daily shows a link to a site
(http://www.software-carpentry.com/)where an implementation 
contest with big prices is announced.
Four projects to replace for instance Make.
And all have to be implemented primarily in Python.

Jaap Groot, Holland


From brian at brian.cbs.umn.edu  Mon Jan 24 11:12:47 2000
From: brian at brian.cbs.umn.edu (Brian Langenberger)
Date: 24 Jan 2000 16:12:47 GMT
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <3dzouke2v7.fsf@amarok.cnri.reston.va.us> <200120002033497688%zippy@cs.brandeis.edu> <388869AB.7EF93B0E@zoo.co.uk>
Message-ID: <86htlv$2cp$1@news1.tc.umn.edu>

Jethro Cramp <jsc at zoo.co.uk> wrote:
: I have both books: 'Programming Python' and 'Learning Python'. I value the
: 'Learning Python' book way and above 'Programming Python'.

I never bothered with 'Programming Python', but I tried
'Learning Python' and had great success with it.  Suddenly all was
clear.  All I needed after that was the 'Pocket Reference' and the
lib.pdf (or just browsing the source scripts) and I had everything
I needed for day-to-day Python usage.

For new users, I recommend 'Learning Python' quite highly, myself.



From eqeq at lll.com  Tue Jan  4 14:47:28 2000
From: eqeq at lll.com (eqeq at lll.com)
Date: Tue, 04 Jan 2000 19:47:28 GMT
Subject: !!!!!!1500 SEXY PICS FREE TO DOWNLOAD NOW!!!!!
Message-ID: <k7sc4.14603$bj7.11052@typ12.deja.bcandid.com>

http://fly-in.to/youngteens



From phil at ricochet.net  Tue Jan 18 04:27:36 2000
From: phil at ricochet.net (phil at ricochet.net)
Date: Tue, 18 Jan 2000 09:27:36 GMT
Subject: Update: Trouble with Python and ASP
References: <860gpo$882$1@nnrp1.deja.com> <860of5$dh3$1@nnrp1.deja.com>
Message-ID: <38843132.14971337@news.ricochet.net>

On Tue, 18 Jan 2000 03:59:40 GMT, larsde at my-deja.com wrote:

>In article <860gpo$882$1 at nnrp1.deja.com>,
>  larsde at telepost.com wrote:
>> I'm experiencing disturbing problems with using Python as an ASP
>> scripting engine. I've been working on a project for quite some time,
>> and so far everything has been peachy.
>
>I also sometimes even get the following error message:
>
>Python ActiveX Scripting Engine error '80020009'
>
>Traceback (innermost last): KeyboardInterrupt:
>
>This also happens only when I try to access the application with two or
>more webclients at the same time. This can't possibly be right, right?
>
I've never used server side scripting, but it sort of sounds like it
might be a threading model issue of which i apparently have had a few
(none of which have anything to do with Python). The model by default
looks to be "both".  You could try setting the model to "Single" and
see if that fixes things and if it does to "Apartment" and see if that
still works.

phil.


The world is divided into two sorts of people:
Those that believe the world is divided into
two sorts of people and those that don't.


From josht at iname.com  Sun Jan  9 14:40:26 2000
From: josht at iname.com (Josh Tomkins)
Date: Sun, 09 Jan 2000 19:40:26 GMT
Subject: Modifying Global Variables from Within a Function
Message-ID: <8EB69CD0Ejosuetompkins@209.208.0.7>

Hello Everyone!

This is, perhaps, a newbie type of question (I am new to Python), but
here goes: 

Can you modify a global variable from inside a function?  Basically what
I want to do is this: 

a = 0 #global variable

def func1(None):
    	a = 5 #or whatever
    	return

def func2(None):
    	#do stuff here
    	func1(None)
    	#do more stuff with the newly modified a
    	return

this is a contrived example.  In reality, I will need to modify several
variables, changing them to zero.  Is this possible? 

Thanks,

Josue


From jxmvlq at sdfsdf.com  Sun Jan  9 09:15:16 2000
From: jxmvlq at sdfsdf.com (jxmvlq at sdfsdf.com)
Date: Sun, 09 Jan 2000 14:15:16 GMT
Subject: C`A,B.L`E DE-SCRAMBLER INSTRUCTIONS......FREE C`A-B,L.E`.............................  2245
Message-ID: <UJ0e4.27200$J%4.49640@dfiatx1-snr1.gtei.net>

LEGAL C`A`B`L`E TV D`E-S`C`R`A`M`B`L`E`R

Want to watch Sporting Events?--Movies?--Pay-Per-View??....FREE!!!!

*This is the Famous R-O Shack TV D-e-s-c-r-a-m-b-l-e-r 
You can assemble it from Radio Shack parts for about $12 to $15.

We Send You: 
1, E-Z To follow Assembly Instructions. 
2. E-Z To read Original Drawings. 
3. Total Parts List.

**** PLUS SOMETHING NEW YOU MUST HAVE! ****
Something you can't do without.

THE UP-TO-DATE 6 PAGE REPORT: 
USING A D-E-S-C-R-A-M-B-L-E-R LEGALLY

Warning: You should not build a TV D-e-s-c-r-a-m-b-l-e-r 
without reading this report first.

You get the complete 6 page report
and instruction package including
the instruction plans, the easy to 
follow diagram, and most important 
of all the "Using a D`e`s`c`r`a`m`b`l`e`r
LEGALLY Report all for just--$10.00 

Fill out form below and send it,
along with your $10.00 payment to:

C.a.b.l.e.t.r.o.n FREE-TV
12187 S. Orange Blossom Trail #116
Orlando Fl 32837
 
(Cash, Check or Money Order.)
(Florida residents include 7% Florida State Sales Tax)
(All orders outside the U.S.A. add $5.00)


PRINT YOUR:

NAME______________________________________________________

ADDRESS___________________________________________________

CITY/STATE/ZIP____________________________________________

E-MAIl ADDRESS____________________________________________ 



rtg45t34t43t34t34t43t34t34t34t34t34t34t34
xylzkcufgnzbjlovevfptjtqsnrdvuebkljlodxwwfugjvejjtuqixmectwqyjlfwdtqrqzytouxcgddfbqdrhfythpvhspzrgfznfcgdtpmjgmyyuhmrzxzcej



From aahz at netcom.com  Wed Jan  5 17:31:14 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 5 Jan 2000 22:31:14 GMT
Subject: htmllib question - Is this the best tool for my task?
References: <84ecvh$mh2$1@nnrp1.deja.com> <84efnr$rgl$1@nntp9.atl.mindspring.net> <84g50u$rtf$1@nnrp1.deja.com> <yu1so0cw4p7.fsf@black-racer.atg.com>
Message-ID: <850gni$pk$1@nntp1.atl.mindspring.net>

In article <yu1so0cw4p7.fsf at black-racer.atg.com>,
chris patti  <cpatti at black-racer.atg.com> wrote:
>feoh at cosmic.com writes:
>> 
>> I have a task where I need to figure out if certain HTML tags (Rather
>> Java Server Pages tags, but they're still HTML, kinda :) contain certain
>> things in an HTML document I'd feed to the script.
>> 
>> I'm thinking htmllib is the right tool for this task, am I wrong?
>> 
>> So as far as I understand things thus far, I should subclass HTMLParser,
>> define start_java and end_java tag functions.. But where from there?
>
>To answer my own question - I ended up using regexps for this.
>htmllib was overkill.

Butting in before the Timbot, if htmllib was overkill, regexps are
probably the wrong solution.  string.find() is probably faster.
--
                      --- Aahz (@netcom.com)

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

The problem with an ever-changing .sig is that you have to keep changing it.


From gerrit.holl at pobox.com  Fri Jan 21 11:30:13 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 21 Jan 2000 17:30:13 +0100
Subject: stupid question !
In-Reply-To: <001601bf6425$5fccfb60$3fe8bd89@-p---v---e-->; from s986001@mailserv.cuhk.edu.hk on Fri, Jan 21, 2000 at 11:37:06PM +0800
References: <001601bf6425$5fccfb60$3fe8bd89@-p---v---e-->
Message-ID: <20000121173013.A1465@stopcontact.palga.uucp>

Kenny wrote on 948494226:
> 
> -----???l?l??-----
> ?H????: Gerrit Holl <gerrit.holl at pobox.com>
> ??????: Kenny <s986001 at mailserv.cuhk.edu.hk>
> ????: 2000?~1??21?? PM 08:05
> ?D??: Re: stupid question !
> 
> I have checked the permission of the directory, and I ensure the permission
> of
> the /cgi-bin is full permission. The result on the browser is still the text
> I typed in the program.
>   I am using Win-98 version and implementing it in Personal Web server.

I don't know about Personal Web server, but with Apache, you have to tell
*Apache* that /cgi-bin has the permissions, not the directory permissions.
Can you execute any other programs in /cgi-bin using <form ...>?
If not, read the documentation on Personal Web server.

> Should I use Linux to do it? Thank you!

I don't know.

>     And one more thing, do you know which free web hosting site supports
> python?     thank you!

I know, but they don't give it away to everyone. They only host sites they
like. It's a Linux machine. They mainly host opensource projects and they
want to see the site before the decide to host it.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From anders.eriksson at morateknikutveckling.se  Fri Jan 28 04:57:14 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Fri, 28 Jan 2000 10:57:14 +0100
Subject: Cookies
References: <9eR9OHHc=LB237DA=k6HW46V+bO9@4ax.com> <14461.65478.595789.251070@ghost.bbn.com> <3890ACC6.24815613@fastpointcom.com>
Message-ID: <bGKROHO9yop+OHnOT1LjnTAgM4wO@4ax.com>

On Thu, 27 Jan 2000 12:38:30 -0800, brad marshall
<bradmarshall at fastpointcom.com> wrote:

>Tim, I can't find your Cookie module anywhere to download.  COuld you
>point me in the right direction?

Well I'm not Tim but...

There is a link from 'Vault of Parnassus' just do a search on cookie

// Anders
PS There seem to be some trouble with connecting to 'parnassus' at the
moment so you might have to try later.


---
English isn't my first or second language, 
so errors or anything that you find offending is not there on 
purpose it's just due to the translation.


From lefebvre at medz.mcgill.ca  Thu Jan 20 09:12:44 2000
From: lefebvre at medz.mcgill.ca (Luc Lefebvre)
Date: Thu, 20 Jan 2000 14:12:44 GMT
Subject: Looking for the FAQ
References: <slrn88bqi2.512.sprawlsr@TacHQ.com>
Message-ID: <lefebvre-2001000912180001@landru.medcor.mcgill.ca>

In article <slrn88bqi2.512.sprawlsr at TacHQ.com>, sprawlsr at worldnet.att.net wrote:

> Hello,
> 
>         I'm new to this group. I run Linux and it has Python 1.5. I'm
> an ANSI C programmer, learning C++ at my University, and I'm
> interested in checking python out. Is there a good book that takes a
> person from beginning to at least intermediate level? This is probably
> something covered in the FAQ, so if you just point me to the FAQ, I'll
> take it from there.
> 
> Robert

There is the book "Learning Python" from O'Reilley which got a good review
in Linux Journal.

-- 
***     Luc Lefebvre
NOTE:   if your reply via email
        please remove the z from my return address
***


From mgushee at havenrock.com  Sun Jan  9 09:38:55 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 09 Jan 2000 09:38:55 -0500
Subject: Newbie here, a few questions
References: <20000108193621.01129.00000731@ng-fk1.aol.com>
Message-ID: <m3ln5zux6o.fsf@havenrock.com>

grndayfan6 at aol.com (GrnDayFan6) writes:

> I'm new to programming in general, and I thought, after looking around a while,
> that Python would be a good place to start.

And I'm sure many here would agree with you. Welcome!

> loops. Nothing fancy yet. But I was wondering how I would turn the code that I
> use for my programs into stand alone programs, so I don't have to use the
> interpreter to execute it.

Have a look at http://starship.python.net/crew/gmcm/distribute.html

> And just in case it matters, I have version 1.4, not 1.5.2.

Don't think it matters -- though 1.4 is generally getting a bit musty.

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From s.schwarzer at ndh.net  Sat Jan 29 07:33:21 2000
From: s.schwarzer at ndh.net (Stefan Schwarzer)
Date: Sat, 29 Jan 2000 13:33:21 +0100
Subject: existence of a variable
References: <86t4fu$s6u$1@nnrp1.deja.com>
Message-ID: <3892DE11.971B805@ndh.net>

Hello Martin,

martindvorak at my-deja.com schrieb:
> What is the best way to determine existence
> of a variable in Python (I need to check
> if a variable is defined and if so,
> delete it)?

If you know in which module (class etc.) the variable would be,
you may use the builtin hasattr function:

[warpy|d:/]echo a = 7 >test.py
[warpy|d:/]python
Python 1.5.2 (#0, Jun 27 1999, 11:23:01) [VisualAge C/C++] on os2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import test
>>> hasattr( test, 'a' ), hasattr( test, 'b' )
(1, 0)
>>> class testclass:
...     def f(): pass
...
>>> hasattr( testclass, 'f' ), hasattr( testclass, 'g' )
(1, 0)

<Disliking-the-inconsistency-of-hasattr-vs.-has_key>-ly y'rs
Stefan


From m.faassen at vet.uu.nl  Thu Jan  6 21:33:56 2000
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 7 Jan 2000 02:33:56 GMT
Subject: Python plugin
References: <3874ECFF.64095C00@iqsoft.hu>
Message-ID: <853jak$iuq$3@newshost.accu.uu.nl>

Nemeth Miklos <nemeth at iqsoft.hu> wrote:
> Is there a (webbrowser) plugin for executing Python scripts?
> I'm having a lot of problem with Java, and I am seeking alternatives.

What about JPython?

http://www.jpython.org

Regards,

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


From tseaver at aftershock.neosoft.com  Mon Jan  3 13:26:25 2000
From: tseaver at aftershock.neosoft.com (Tres Seaver)
Date: 3 Jan 2000 12:26:25 -0600
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <3870DB37.761034EA@roguewave.com> <75B5A3A6EB0EDE04.FCE8CED8D9B414D5.95006D8FA0ECEB31@lp.airnews.net> <3870E485.998FF2FD@roguewave.com>
Message-ID: <A52FE169FF2DDD09.52412C585E0B2951.4D93D8DA88B9FC5B@lp.airnews.net>

In article <3870E485.998FF2FD at roguewave.com>,
bjorn  <bjorn at roguewave.com> wrote:
>
>
>Tres Seaver wrote:
>
>> In article <3870DB37.761034EA at roguewave.com>,
>> bjorn  <bjorn at roguewave.com> wrote:
<snip>
>> >So what you're saying is that dicts are broken since they can't take
>> >lists as keys <wink>.
>>
>> What reasonable use can you propose for an associative container which
>> allows mutable keys?  You might as well revert to a list of (key,value)
>> pairs, since you're going to do a linear search on each lookup anyway
>> if the keys are allowed to change.
>
>If you look at any hashtable implementation for C, you'll see that the keys
>can >be char* (or any other mutable type for that matter).  There are really
>only two >requirements for the keys in an associative container (i) they must
>be hashable, and (ii) they must remain unchanged for the duration of the
>container (or more >specifically, the hash value must remain unchanged).

Strings are immutable in Python, and thus make ideal keys.  

>It's might be nice (for a suitable definition of nice) to have the language
>hold your hand through this, but it is far from neccessary.  Sometimes it
>really does suffice to tell the programmer that if he changes a key he'll
>get undefined behavior.

And indeed, constructing a tuple of mutable elements can generate just such
behavior, even in Python (although it requires some trickery to make it happen,
as Python knows about the obvious cases, and complains of "unhashable type").

However, dictionaries, and the proper working thereof, are essential to
Python's own implementation -- they do not exist merely as "user-space" data
structures.  Easing unsafe practices here is asking for trouble.

Corned-beef-or-turkey-hash?'ly

Tres.
-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com


From tjg at avalongroup.net  Mon Jan 10 21:02:51 2000
From: tjg at avalongroup.net (Timothy Grant)
Date: Mon, 10 Jan 2000 18:02:51 -0800
Subject: PyGreSQL and Win?
Message-ID: <387A8F4B.383761FF@exceptionalminds.com>

Has anyone ever compiled and successfully used the PyGreSQL module on a
Windows box?


-- 
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<



From gerrit.holl at pobox.com  Thu Jan 13 15:05:35 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 13 Jan 2000 21:05:35 +0100
Subject: Desperate: Linux Sound Module
Message-ID: <20000113210535.A2015@stopcontact.palga.uucp>

Hello,

I'm really desperate: I can't find any module for playing sounds on a
Linux system! I already searched everywhere: google, python.org, parnassus.
I just want to play .au or .wav, but even edna can't *play* *play* *PLAY*
an mp3!

I don't have solaris, I don't use Windows. Can't I play sounds?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my mail!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From ralph at inputplus.demon.co.uk  Mon Jan 24 17:59:55 2000
From: ralph at inputplus.demon.co.uk (Ralph Corderoy)
Date: 24 Jan 2000 22:59:55 -0000
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <3dzouke2v7.fsf@amarok.cnri.reston.va.us> <200120002033497688%zippy@cs.brandeis.edu> <B4B0FA22.196%matt_winterroth@hotmail.com>
Message-ID: <86ilhb$hc6$1@inputplus.demon.co.uk>

> lets not forget that lutz's "programming python" is quite old -
> dating back to october of 1996...i'd much rather wait for a new
> edition to be printed up, but i'm not holding my breath with
> o'reilly.

I was referring to _Programming Python_ by Lutz, published by ORA in
1999.  It isn't marked as 2nd Ed., IIRC.  Perhaps it is just re-print
with corrections?

The 1996/1999 difference confused me a little.


Ralph.



From mhammond at skippinet.com.au  Wed Jan  5 22:14:33 2000
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 06 Jan 2000 03:14:33 GMT
Subject: Compiling on Alpha-Win2K/WinNT (Python 1.5.2)
References: <850p0u$q6a$1@nnrp1.deja.com>
Message-ID: <tMTc4.11600$lv5.51353@news-server.bigpond.net.au>

<tmick at mail.com> wrote in message news:850p0u$q6a$1 at nnrp1.deja.com...
> I am trying to compile Python 1.5.2 (as extracted from py152.tgz on
> www.python.org) for Win2000 on an Alpha. Has anyone done this? I
> presume that it should be similar to compiling for Alpha/WinNT.

Not for some time AFAIK.

> My first problem is that all of the build icons/menu-entries in
> DevStudio are grayed out (i.e. I can't build the project). Does
anyone
> know what this indicates?

Almost certainly that the correct "platform" does not exist in the
project.  You will need to add a new configuration for your platform.
This also happens with Windows CE.

> No configuration specified. Defaulting to python15 - Win32 Debug.
>
> cl.exe /nologo /MDd /W3 /Gm /GX /Zi /Od /I "..\Include" /I "..\PC"
/D "_
> DEBUG" /D "USE_DL_EXPORT" /D "WIN32" /D "_WINDOWS" /Fp".\temp-
> debug\python15.pch
> " /YX /Fo".\temp-debug\\" /Fd".\temp-
> debug\\" /FD /c ..\Modules\fpectlmodule.c
> fpectlmodule.c
> ..\Modules\fpectlmodule.c(167) : fatal error C1083: Cannot open
include
> file: 'machine/fpu.h': Permission denied
> NMAKE : fatal error U1077: 'cl.exe' : return code '0x2'

Permission denied is a little strange.  I have no idea what that
module even does, you you could try removing it from the project - you
will also need to remove a reference in PC\config.h

> I do not know of a file called "fpu.h" and cannot find one on my
system.

Strange - not one on my intel based system either.

Sorry I cant be more help - we need to convince from Alpha vendors
that some people in the Python community need one or 2 to ensure
support for their machines :-)

Mark.




From thor at localhost.localdomain  Sun Jan 16 08:40:48 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 16 Jan 2000 13:40:48 GMT
Subject: python ring Info
Message-ID: <slrn883m5g.6he.thor@localhost.localdomain>

Python ring Info
-------------------------

This is a piece of information about the existing python ring in:
http://www.ctv.es/USERS/irmina/pythonring.html
(now with Statistics)

What is it?
>From a functional point of view, it's a list of linked web sites
that share a common interest or information, in this case python
related stuff.

How does it work?
The HTML code included in each node of the ring calls a CGI that
performs the movement thru the ring.

How can I join?
In the above web site , fill in your data and that's all.

Why I should ? 
If you've done a web site, I think it's worth you spend 15 minutes
promoting it, after your web site has been included in it you'll get
more hits than ever.
Moreover, your information may be interesting for other people, or
you can be interested in other people's work, a big Webring provides
the community a lot of unusual but somehow interesting data.

How can I get more hits? 
Because the Webring is a democratic way of organization, that is,
all its members tend to receive the same ammount of visits, the
big ones as much as the small ones.

Well, I don't have a Web page.., so?
So  you must (  :) ) start doing one right now. Seriously, we
( at least me,MGA) are interested in people's work, perhaps you've
been collecting data or even your libraries or methods of work are
clearly defined, the ammount of time required for a fast compilation
of that info is not too much. And I remark that you can learn only
if all ( or many) want to show or teach what they know.

Of course we admit commercial Web sites also, *any* Web site about
python.

Thanks for your time and patience reading this.
MGA





From aahz at netcom.com  Sun Jan 16 09:56:57 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Jan 2000 14:56:57 GMT
Subject: Does python support long longs?
References: <x8_f4.26$tX3.648@vixen.cso.uiuc.edu> <85qa3c$hk$1@nntp9.atl.mindspring.net> <722C7DA3C5245C0A.D784A1B5D0E7D6C6.EABECEEA7DD01A77@lp.airnews.net>
Message-ID: <85sm7p$s58$1@nntp6.atl.mindspring.net>

In article <722C7DA3C5245C0A.D784A1B5D0E7D6C6.EABECEEA7DD01A77 at lp.airnews.net>,
Cameron Laird <claird at starbase.neosoft.com> wrote:
>In article <85qa3c$hk$1 at nntp9.atl.mindspring.net>,
>Aahz Maruch <aahz at netcom.com> wrote:
>>
>>The way to build large Python applications is to componentize and
>>loosely-couple the hell out of everything.
>
>A lot more than just Python, of course.

Sure, but Python *encourages* it.  Python makes it *easy*.  Writing
test-harness wrappers so that you can test the component you're
currently building usually takes an hour at the most -- and you can use
that for QA later on if you wish.

I think, though, that it's to Python's advantage that it came late
enough to the game to leverage a lot of component technologies (database
access and DCOM/Corba/XML-RPC and sockets in general).
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From crushen at onvoy.com  Sat Jan 15 05:49:44 2000
From: crushen at onvoy.com (crushen at onvoy.com)
Date: Sat, 15 Jan 2000 10:49:44 GMT
Subject: Woah
Message-ID: <38805105.11391226@news.onvoy.com>

THIS WILL ONLY WORK IF YOU FOLLOW THE SIMPLE INTRUCTIONS AT THE
BOTTOM..DO THIS HONESTLY AND FAIRLY AND IT WILL WORK! 

How to turn $6 into $6,000

READING THIS COULD CHANGE YOUR LIFE! 
I found this on a bulletin board and decided to try it. A little while
back, I was browsing through newsgroups and came across an article
similar to this that said you could make thousands of dollars within
weeks with only an initial investment of $6.00! So I thought, "Yeah
right, this must be a scam", but like most of us, I was curious, so I
kept reading. Anyway, it said that you send $1.00 to each of the 6
names and address stated in the article. You then place your own name
and address in the bottom of the list at #6, and post the article in
at least 200 newsgroups. (There are thousands) No catch, that was it.
So after thinking it over, and talking to a few people first, I
thought about trying it. I figured: "what have I got to lose except 6
stamps and $6.00, right?" Then I invested the measly $6.00. Well GUESS
WHAT!!... within 7 days, I started getting money in the mail! I was
shocked! I figured it would end soon, but the money just kept coming
in. In my first week, I made about $25.00. By the end of the second
week I had made a total of over $1,000.00! In the third week I had
over $10,000.00 and it's still growing. This is now my fourth week and
I have made a total of just over $42,000.00 and it's still coming in
rapidly. It's certainly worth $6.00, and 6 stamps, I have spent more
than that on the !! Let me tell you how this works and most
importantly, why it works....Also, make sure you print a copy of this
article NOW, so you can get the information off of it as you need it.
I promise you that if you follow the directions exactly, that you will
start making more money than you thought possible by doing something
so easy! 
Suggestion: Read this entire message carefully! (print it out or
download it.) Follow the simple directions and watch the money come
in! 

It's easy. It's legal. And, your investment is only $6.00 (Plus
postage) 
IMPORTANT: This is not a - ; it is not indecent; it is not illegal;
and it is virtually no risk - it really works!!!! 
If all of the following instructions are adhered to, you will receive
extraordinary dividends. 
PLEASE NOTE: Please follow these directions EXACTLY, and $50,000 or
more can be yours in 20 to 60 days. This program remains successful
because of the honesty and integrity of the participants. Please
continue its success by carefully adhering to the instructions. 
You will now become part of the Mail Order business. In this business
your product is not solid and tangible, it's a service. You are in the
business of developing Mailing Lists. Many large corporations are
happy to pay big bucks for quality lists. However, the money made from
the mailing lists is secondary to the income which is made from people
like you and me asking to be included in that list. 
Here are the 4 easy steps to success: 
STEP 1: Get 6 separate pieces of paper and write the following on each
piece of paper "PLEASE PUT ME ON YOUR MAILING LIST." Now get 6 US
$1.00 bills and place ONE inside EACH of the 6 pieces of paper so the
bill will not be seen through the envelope (to prevent thievery).
Next, place one paper in each of the 6 envelopes and seal them. You
should now have 6 sealed envelopes, each with a piece of paper stating
the above phrase, your name and address, and a $1.00 bill. What you
are doing is creating a service. THIS IS ABSOLUTELY LEGAL! You are
requesting a legitimate service and you are paying for it! Like most
of us I was a little skeptical and a little worried about the legal
aspects of it all. So I checked it out with the U.S. Post Office
(1-800-725-2161) and they confirmed that it is indeed legal! Mail the
6 envelopes to the following 
addresses: 

#1)Bill Lijewski 807 Keast Hutchinson, KS 67501 

#2) Rodney Fadler p.o.box 244 Herculaneum,mo.63048 

#3)M. Molina 807 Siesta Loop Robstown, TX 78380 

#4) Annie J 716 39th street Des Moines IA 50312 

#5)Alvin Loi 4148 N. Peck RD Apt.D El Monte CA, 91732 

#6) David Wiebler 6440 Elm Street, Corcoran MN, 55340

STEP 2: Now take the #1 name off the list that you see above, move the
other names up (6 becomes 5, 5 becomes 4, etc...) and add YOUR Name as
number 6 on the list. 

STEP 3: Change anything you need to, but try to keep this article as
close to original as possible. Now, post your amended article to at
least 200 newsgroups. (I think there are close to 24,000 groups) All
you need is 200, but remember, the more you post, the more money you
make! 
This is perfectly legal! If you have any doubts, refer to Title 18
Sec. 1302 & 1341 of the Postal laws. Keep a copy of these steps for
yourself and, whenever you need money, you can use it again, and
again. 
PLEASE REMEMBER that this program remains successful because of the
honesty and integrity of the participants and by their carefully
adhering to the directions. Look at it this way. If you are of
integrity, the program will continue and the money that so many others
have received will come your way. 
NOTE: You may want to retain every name and address sent to you,
either on a computer or hard copy and keep the notes people send you.
This VERIFIES that you are truly providing a service. (Also, it might
be a good idea to wrap the $1 bill in dark paper to reduce the risk of
mail theft.) 
So, as each post is downloaded and the directions carefully followed,
six members will be reimbursed for their participation as a List
Developer with one dollar each. Your name will move up the list
geometrically so that when your name reaches the #1 position you will
be receiving thousands of dollars in CASH!!! What an opportunity for
only $6.00 ($1.00 for each of the first six people listed above) Send
it now, add your own name to the list and you're in business! 

---DIRECTIONS ----- FOR HOW TO POST TO NEWSGROUPS------------ 
Step 1) You do not need to re-type this entire letter to do your own
posting. Simply put your cursor at the beginning of this letter and
drag your cursor to the bottom of this document, and select 'copy'
from the edit menu. This will copy the entire letter into the
computer's memory. 
Step 2) Open a blank 'notepad' file and place your cursor at the top
of the blank page. From the 'edit' menu select 'paste'. This will
paste a copy of the letter into notepad so that you can add your name
to the list. 
Step 3) Save your new notepad file as a .txt file. If you want to do
your postings in different settings, you'll always have this file to
go back to. 
Step 4) Use Netscape or Internet explorer and try searching for
various newsgroups (on-line forums, message boards, chat sites,
discussions.) 
Step 5) Visit these message boards and post this article as a new
message by highlighting the text of this letter and selecting paste
from the edit menu. Fill in the Subject, this will be the header that
everyone sees as they scroll through the list of postings in a
particular group, click the post message button. You're done with your
first one! Congratulations...THAT'S IT! All you have to do is jump to
different newsgroups and post away, after you get the hang of it, it
will take about 30 seconds for each newsgroup! 

**REMEMBER, THE MORE NEWSGROUPS YOU POST IN, THE MORE MONEY YOU WILL
MAKE!! BUT YOU 
HAVE TO POST A MINIMUM OF 200** That's it! You will begin receiving
money from around the world within days! You may eventually want to
rent a P.O.Box due to the large amount of mail you will receive. If
you wish to stay anonymous, you can invent a name to use, as long as
the postman will deliver it. **JUST MAKE SURE ALL THE ADDRESSES ARE
CORRECT.** 
Now the WHY part: Out of 200 postings, say I receive only 5 replies (a
very low example). So then I made $5.00 with my name at #6 on the
letter. Now, each of the 5 persons who just sent me $1.00 make the
MINIMUM 200 postings, each with my name at #5 and only 5 persons
respond to each of the original 5, that is another $25.00 for me, now
those 25 each make 200 MINIMUM posts with my name at #4 and only 5
replies each, I will bring in an additional $125.00! Now, those 125
persons turn around and post the MINIMUM 200 with my name at #3 and
only receive 5 replies each, I will make an additional $626.00! OK,
now here is the fun part, each of those 625 persons post a MINIMUM 200
letters with my name at #2 and they each only receive 5 replies, that
just made me $3,125.00!!! Those 3,125 persons will all deliver this
message to 200 newsgroups with my name at #1 and if still 5 persons
per 200 newsgroups react I will receive $15,625,00! With an original
investment of only $6.00! AMAZING! When your name is no longer on the
list, you just take the latest posting in the newsgroups, and send out
another $6.00 to names on the list, putting your name at number 6
again. And start posting again. The thing to remember is: do you
realize that thousands of people all over the world are joining the
internet and reading these articles everyday?, JUST LIKE YOU are now!!
So, can you afford $6.00 and see if it really works?? I think so...
People have said, "what if the plan is played out and no one sends you
the money? So what! What are the chances of that happening when there
are tons of new honest users and new honest people who are joining the
internet and newsgroups everyday and are willing to give it a try?
Estimates are at 20,000 to 50,000 new users, every day, with thousands
of those joining the actual internet. Remember, play FAIRLY and
HONESTLY and this will really work. 



From phd at phd.russ.ru  Wed Jan 19 07:49:36 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 19 Jan 2000 12:49:36 +0000 (GMT)
Subject: Zope and Apache
In-Reply-To: <ta6FOAJep64FzHytVh77WRvo3KA=@4ax.com>
Message-ID: <Pine.LNX.4.21.0001191248270.22898-100000@fep132.fep.ru>

On Wed, 19 Jan 2000, Anders M Eriksson wrote:
> On a Solaris system must I be root to install zope?

   It depends on how your Apache is running, but mostly yes - for some
operations you need to be root.

> Is it possible to configure zope so it will run with a preexisting
> Apache webserver?
> 
> Must I be root to do this?

   Again, it depends on how your server is configured. Probably, yes.

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From mwh21 at cam.ac.uk  Thu Jan 13 12:21:57 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Jan 2000 17:21:57 +0000
Subject: Python Books & improving PyAwareness - How to ?
References: <01bf5597$3dfee1c0$ddc854a6@brancusi.pro-net.co.uk>
Message-ID: <m31z7lkhu2.fsf@atrus.jesus.cam.ac.uk>

"Jason Cunliffe" <jasonic at nomadicsltd.com> writes:

> Hello
> 
> I am very happy to see the new Python books appearing. Some good titles and
> all the new ones I've picked up are excellent [Quick Python, The Essential
> Python Reference and Python-Annotated archives]...my sincere thanks to
> their respective authors.
> 
> I imagine that most people shooping for Pyhton books will tend to follow
> the links from www.python.org pages and purchase through amazon.com or
> similar.
> 
> Meanwhile, I have been making an ad-hoc survey in the New York City
> bookstores over the holiday. Sad to say there is a very low profile of
> python books there. Only a couple of copies at best in most
> stores

To play a different tune, I went book shopping in London the other day
and was quite astonished at the number of Python books I found.
Blackwell's had a couple, Waterstone's had quite a few and Border's
had loads.

Happy days.

Except that I was looking for Lisp books. Oh well. (Border's had PAIP
- which is out of print - but for ?52. Ouch.)

> ...compared to shelf-fulls of the other P and J lanaguage books. Even
> worse [but perhaps significant] is that in most stores the Python books
> were extremely oddly placed - sometimes a volume or two in the multimedia
> section, sometimes a tome in Internet, somtimes a copy wher you would
> expect it a thte end of the Perl shelf. In no store did I find all the
> current published Python books together. 
> 
> I noted that ther was no Python books in Linux sections - which I think is
> rather surprisin and dissappointing, especially given the enthusiasm now
> for Linux sections in these stores.

<gripe> 
I can never find technical books in bookstores. It's the ridiculous
over-categorisation that gets to me. On this day in London I found
shops that had separate sections for "Pure Mathematics" and for "Logic
and Set theory". I mean, what?

I'm not sure you can do much better than alphabetical order by author
surname, and that sucks too. But less.
</gripe>

of-course-I've-never-needed-a-book-on-something-as-simple-Python-ly
y'rs Michael


From test at test.test  Sat Jan  8 13:18:22 2000
From: test at test.test (Test test)
Date: 08 Jan 2000 13:18:22 -0500
Subject: test
Message-ID: <ltiu14qvf5.fsf@test.test.com>

test
-- 
 test
 test at test.test


From fredrik at pythonware.com  Sat Jan 22 03:35:27 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 22 Jan 2000 09:35:27 +0100
Subject: comp.lang.python.announce working??
References: <01bf643a$5039a8a0$3e41c3d1@Cyberus.cyberus.ca>
Message-ID: <005201bf64b5$d299b490$f29b12c2@secret.pythonware.com>

Gordon Williams <g_will at cyberus.ca> wrote:
> I havent had any news feed from comp.lang.python.announce since the end of
> December.  I contacted my ISP and the news came back from the group that
> supplies the news feed that there havent been any new postings to this
> group since this time.  I find that strange.
> 
> Is anyone else able to access this news group?

it's empty on my server at the moment; could be that
the latest batch (posted friday) hasn't made it through
yet.

you can always check the egroups archive:
http://www.egroups.com/group/python-announce-list/

</F>




From engbpa at liquidinformation.com  Wed Jan  5 13:39:39 2000
From: engbpa at liquidinformation.com (Bala)
Date: 5 Jan 2000 18:39:39 GMT
Subject: Which Python book?
Message-ID: <85035b$1sb$1@bignews.shef.ac.uk>

Hi I'm new to Python but not to programming. I currently have:
        Programming Python (Nutshell Handbook)
        by Mark Lutz
But I find it a little too heavy for my liking.

I'm now thinking of getting either
        Python Essential Reference (OTHER NEW RIDERS)
        by David M. Beazley, Guido Van Rossum
or
        Learning Python (Help for Programmers)
        by Mark Lutz, David Ascher, Frank Willison (Editor)
Which one of these two would you recommend?

Thanks.

-Bala


From nhv at cape.com  Thu Jan 13 06:22:31 2000
From: nhv at cape.com (Norman Vine)
Date: Thu, 13 Jan 2000 06:22:31 -0500
Subject: Python and MapObjects (Was: Python and ArcView GIS)
References: <GyPe4.14473$2x3.253287@newscontent-01.sprint.ca> <Pine.SGI.3.95.1000112172541.19328B-100000@clavey.des.ucdavis.edu>
Message-ID: <85kci6$i9l$1@newsie2.cent.net>

Allan Hollander

> The ArcView extension sounds very interesting.

Yes :)

> Along these lines of interfaces to ESRI software, has anybody done any
> work with getting Python to talk to the MapObjects set of components?  I'd
> love to have some pointers on how to do this

It is not ArcView or even ESRI for that matter
but there is the OpenSource mapserver project
that works with shapefiles and has a SWIGed interface

see
http://mapserver.gis.umn.edu/

Norman Vine





From gerrit.holl at pobox.com  Sun Jan 30 14:44:20 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 30 Jan 2000 20:44:20 +0100
Subject: Why Python rules
Message-ID: <20000130204420.A24979@stopcontact.palga.uucp>

Hello,

I finally know why Python rules. It took me almost a year to find out,
but finally, I know it,

the interactive Python shell uses '>>> ' as it's prompt. So a good mail
client shows Python code in email in different colors than the rest of 
the mail. So Python code is always readable.

regards,
Gerrit.

-- 
Please correct any bad Swahili you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From gmcm at hypernet.com  Sun Jan 16 13:21:40 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 16 Jan 2000 13:21:40 -0500
Subject: using threads with python and winnt
In-Reply-To: <85snka$3di@fstgal00.tu-graz.ac.at>
Message-ID: <1264075593-4560755@hypernet.com>

Kreiner Karl or Charly or somebody wrote:

>             does anyone have expiriences with using threads and winnt ? I
> started working with the built-in module "thread" using
> the function "start_new_thread" , which needs two parameters : a function
> and a list of araguments to be passed to the
> function. but no matter what I do , i just get the error-message:
> 
> type error: first arg must be callable
> 
> I consulted the python manual but it didn't provide help to my problem. so,
> does anyone know the correct syntax of this
> function and are there generally any problems with using the whole library
> with windows nt ?

Threads work dandy on NT (believe it or not, one of the better 
thread implementations you'll find). You should probably be 
using "threading" instead of "thread"; the latter is extremely 
low level. As far as your problem goes, you're probably 
spelling it wrong.

def func(arg)
   # blah blah

Now "func" is a callable, but "func(1)" is an invocation which, if 
it returns a callable will work, but that's probably not what 
you're trying to do.

- Gordon



From gmcm at hypernet.com  Wed Jan  5 08:51:37 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 5 Jan 2000 08:51:37 -0500
Subject: zlib and zip archives
In-Reply-To: <84un63$8ku$1@nnrp1.deja.com>
Message-ID: <1265043095-3792298@hypernet.com>

Anne writes:
> 
> It sounds stupid but I really don't have a clue:
> 
> How do you use zlib to create a zip archive using the pkzip
> format?  I looked through the documentation, it seems no where
> indicates that it supports the zip archive format?

You don't. zlib just handles compression. You could check 
out ftp://ftp.interet.com/pub/pylib.html.

- Gordon



From thantos at chancel.org  Sat Jan 29 04:58:30 2000
From: thantos at chancel.org (Alexander Williams)
Date: Sat, 29 Jan 2000 09:58:30 GMT
Subject: statements and blocks
References: <86t0ov$4en$1@news.psc.edu>  <86tt56$d4a$1@nnrp1.deja.com>
Message-ID: <slrn895ee6.ha3.thantos@brimstone.mecha>

On Sat, 29 Jan 2000 05:17:29 GMT, rdudfield at my-deja.com
<rdudfield at my-deja.com> wrote:

>I find indentation is better in most cases except this.  This isn't too
>much trouble, as I usually find the problem pretty quickly.  However
>occasionally I waste a bit too much time on it than I'd like to :)

Two things:

If your function is more than a page long, odds are you need to break
it up into multiple functions.  Python is a highly expressive
language; use that to encapsulate and combine smaller functionality
into larger structures.  In the long run, you'll be thankful.

Using a wider default indentation per level can help make things more
visually apparent as to indentation.  2 charas is easy to miss, but
its harder to miss being off by 4 or 8 chara.

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From gerrit.holl at pobox.com  Fri Jan 14 07:29:31 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 14 Jan 2000 13:29:31 +0100
Subject: Desperate: Linux Sound Module
In-Reply-To: <TFSHIQKY@scigen.co.uk>; from dbinks@scigen.co.uk on Fri, Jan 14, 2000 at 09:10:23AM +0000
References: <TFSHIQKY@scigen.co.uk>
Message-ID: <20000114132931.A1533@stopcontact.palga.uucp>

Dominic Binks wrote on 947837423:
> > Hello,
> > > I'm really desperate: I can't find any module for playing sounds on > a Linux system! I already searched everywhere: google, python.org, > parnassus.  I just want to play .au or .wav, but even edna can't > *play* *play* *PLAY*
> > an mp3!
> > > I don't have solaris, I don't use Windows. Can't I play sounds?
> >
> > regards,
> > Gerrit.
> 
> There are two things that spring to mind here.
> 
> 1. Does your linux have a sound driver installed?  This used to be one of the most troublesome parts of building the Linux kernel since there seemed to be no useful documentation on what was required.

Yes.

> 2. If you do have a correctly configured sound driver for your kernel, then simply open the .au and copy it into the /dev/audio device.  You can do this as:
> 
> cat file.au > /dev/audio

No.

I have es1373, and that's not supported. You should use 'play', but
unfortunately, 'play' isn't everywhere.

> If you want to play .wav's you'll probably need to convert them to .au.  In this case you will almost certainly need sox.

Is sox guaranteed to be everywhere?

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my mail!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From pinard at iro.umontreal.ca  Fri Jan 28 10:46:05 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 28 Jan 2000 10:46:05 -0500
Subject: lib reference in Unix man format
In-Reply-To: William Park's message of "Fri, 28 Jan 2000 09:09:50 -0500"
References: <200001281339.IAA12329@trm.bmo.com> <20000128090950.A2767@better.net>
Message-ID: <oqsnzirygy.fsf@titan.progiciels-bpi.ca>

William Park <parkw at better.net> writes:

> HTML is essentially killed INFO, in my opinion.

There is still one thing from Info which is sorely lacking in HTML, which
is the lack of capability to textually search in a whole document.

And if you jump one step above Info, which is Texinfo (Info is automatically
derived from Texinfo), you get something able to generate nice printed
manuals.  Printed dumps from HTML browsers look rather horrible by comparison
(unless you get seduced by images, but I'm thinking text, here).

There is a lot of machinery around HTML 4.0, trying to repair this aspect,
but yet, they are not getting close, still.

I have a big deal of Texinfo documentation in both Info and HTML format,
here, and Info is much more useful that HTML browsers.  In my opinion, Info
was hurt by die-hard `man' addicts on one side, and more by the popularity
of HTML than its real virtues.  Info needs you to know the first letter
of the words Next, Previous, Up and Menu, and to use the SPACE key to
page forward.  Nothing worth whipping a cat, really.

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





From kern at caltech.edu  Sat Jan 29 18:11:41 2000
From: kern at caltech.edu (Robert Kern)
Date: 29 Jan 2000 23:11:41 GMT
Subject: concurrency in Python
References: <000001bf6a98$9142a360$78a0143f@tim>
Message-ID: <86vs3d$95c@gap.cco.caltech.edu>

In article <000001bf6a98$9142a360$78a0143f at tim>,
	"Tim Peters" <tim_one at email.msn.com> writes:
> [posted & mailed; added Python's Thread-SIG]
> 
> [Greg Wilson]
>> Hi.  I'm interested in hearing from anyone who's ever built a
>> concurrency toolkit for Python that used a model other than
>> threads + locks --- futures, tuple spaces, active expressions,
>> concurrent aggregates, actors, or anything else.

[snip]
 
> Of course, Java's recent offshoot of Linda's tuple spaces should be
> available thru JPython.

FWIW (since I really don't know squat about what you guys are talking
about), PyBrenda (http://www.snurgle.org/~pybrenda/) might also be of
use.  From the README:

"""\
+ What is PyBrenda?

PyBrenda is a system for easily doing parallel computation using the
Python programming language.  It is somewhat similar to the C-Linda
language; it uses the notion of "tuple space" to simplify the
interactions among processes.
"""

> Let us know if you turn up anything else!  Most thread activity in the
> Python world has been on the commercial side, where threads+locks is "it"
> for the usual lowest common denominator reasons.
> 
> mix-'em-in-with-floating-point-and-it's-an-unbeatable-
>     enabler-of-late-night-debugging<wink>-ly y'rs  - tim

Robert Kern
kern at caltech.edu


From georgethomas at pspl.co.in  Mon Jan 17 23:06:23 2000
From: georgethomas at pspl.co.in (George Thom@s)
Date: Tue, 18 Jan 2000 09:36:23 +0530
Subject: Python and MS SQL (Microsoft SQL server)
Message-ID: <004801bf6169$62c1ed80$cf02a8c0@intranet.pspl.co.in>

Hi Dave,
The odbc module that comes as part of the win32 extensions bundled with
Python 1.5.2 for the various flavours (!) of Windows should do just fine.
Additional functionality (with some tweaking ... ymmv ;-) can be obtained
with marc Lemburg's mxODBC module.
Regards,
George
-------------------------- Dave wrote -----------------------------
Is there a python module available to access MS SQL (Microsoft SQL server)
My client really wants to use this over mySQL. (Which I know there is a
module
for)

Dave


--------------------------------------------------------------------------
George Thomas
Resident,Planet Earth,Third Rock from the Sun
georgethomas at pspl.co.in


Flat No. FI-5, Fifth Floor,
Himagirinath Co-op. Hsg. Society,
486A Elphinstone Road,
Kirkee, Pune - 411003,
Maharashtra State,
India.
--------------------------------------------------------------------------




From coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com  Wed Jan 19 15:35:44 2000
From: coursesm at sbdhcp-4022.statenisland-ny.est.tcg.com (Stephen Coursen)
Date: 19 Jan 2000 20:35:44 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net> <388521FB.BCC45087@callware.com> <8653jg$esi@news.or.intel.com> <002401bf62b8$c30da8a0$f29b12c2@secret.pythonware.com> <8656ne$h1j@news.or.intel.com>
Message-ID: <slrn88c7qh.ja9.coursesm@sbdhcp-4022.statenisland-ny.est.tcg.com>

On Wed, 19 Jan 2000 12:24:55 -0800, tye4 <tye4 at yahoo.com> wrote:
>
>Fredrik Lundh <fredrik at pythonware.com> wrote in message
>news:002401bf62b8$c30da8a0$f29b12c2 at secret.pythonware.com...
>AC "tye4" <tye4 at yahoo.com> wrote:
>
>> As time goes on, the code using 'end' will increase substantially while
>the
>> legacy code without 'end' will reduce.
>
>so you mean that if you make 'end' optional, everyone
>will add extra end statements to their code?
>
>yeah, right...
>
>Oh yes, they will. Failure to do so will result in barrage of 'warning'
>messages from the compiler that they are using a 'deprecated' language
>construct. (BTW, that's how Java implements gradual change in the language
>and its library.)
>
Of course, this assumes that people think that this is a good idea.  No one,
other you, has expressed the need for an "end" block during this thread.
Perhaps there's a reason for this...

>-tye4
>
>


From R.Brodie at rl.ac.uk  Mon Jan 24 08:33:56 2000
From: R.Brodie at rl.ac.uk (Richard Brodie)
Date: Mon, 24 Jan 2000 13:33:56 -0000
Subject: If you can only buy *one* Python book...
References: <N%rg4.4360$3b6.20735@ozemail.com.au> <20000116182502.A1106@better.net> <t6tg4.305$iL4.4241@news.rdc1.tx.home.com> <20000117204759.B1291@stopcontact.palga.uucp>
Message-ID: <86hkc6$mh4@newton.cc.rl.ac.uk>

"Gerrit Holl" <gerrit.holl at pobox.com> wrote in message
news:20000117204759.B1291 at stopcontact.palga.uucp...

> > The 'Python Essential Reference' is a good addition to the online
> > documentation as well.
>
> Disagreed. I think the online reference is far the best and most
> complete there is.

That really depends on how much you like online documentation,
and/or online development. I often find the presence of a computer
a distraction when coding, and consumes to much real estate on
the desk. It's pretty much personal taste - if your first reaction on
seeing HTML docs  is to look for the PDF/PS version to print then
PER should be your second stop after 'Learning Python'.





From tismer at tismer.com  Mon Jan 31 10:59:15 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 31 Jan 2000 16:59:15 +0100
Subject: Riskless deletion of nested structures
Message-ID: <3895B153.3D2E763D@tismer.com>

Howdy,

Please review!

While implementing Stackless Python, a new problem arose:
Nested structures like frame chains and tracebacks can now
easily grow somuch that they cause a stack overflow on deallocation.

To protect lists, tuples, frames, dicts and tracebacks against
this, I wrote a stackless deallocator.
At the moment, everything is done in trashcan.c .
This gives a slight performance loss of 5% for pystone, most probably
due to the double indirection and non-local code reference.
It is yet a hack, since I'm grabbing the tp->dealloc pointers of these
types and replace them by safe versions. This just in order to
try out things quickly. Later I will change this and incorporate
the stack checks into the affected modules, after I got some
feedback on this.

This patch applies to Stackless and standard Python as well:
Deallocation of deeply nested structures will never again cause
a stack overflow.

Installation for the intermediate version:
Insert a line

	_Py_trashcan_install();

at the end of Py_Initialize in pythonrun.c

Please try it and check my code wether there is a better solution.

cheers - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home
-------------- next part --------------
A non-text attachment was scrubbed...
Name: trashcan.c
Type: application/x-unknown-content-type-cfile
Size: 2210 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000131/4c063f63/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: trashcan.h
Type: application/x-unknown-content-type-hfile
Size: 246 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20000131/4c063f63/attachment-0001.bin>

From stalnaker at acm.org  Thu Jan 27 23:01:58 2000
From: stalnaker at acm.org (Max M. Stalnaker)
Date: Fri, 28 Jan 2000 04:01:58 GMT
Subject: newbie question on modules
References: <38872a70_4@goliath.newsfeeds.com>
Message-ID: <Ww8k4.487$4f5.132763@tw12.nn.bcandid.com>

Would it be adequate for module b to inherit module a?
Then in module b:

class b(a):

It seems possible that this issue is flowing not from python newness, but
from OO newness.  Taking a shot at rethinking the structure to account for
OO possibilities might be useful.  I had been going through that process on
a program that started out modular and is ending up OO and the program is a
structured a bit different than when it started.

"Janos Blazi" <jblazi at netsurf.de> wrote in message
news:38872a70_4 at goliath.newsfeeds.com...
> I have a module a in which a function m_add is defined. Now I want (as the
> last statement of module a) load and execute a module b. Now it seems that
b
> does not know the function m_add. How can I change that?
>
> I'd like to access everything defined in a in the module b that is the
> (variable) second part of a.
>
> What can I do?
>
> Janos Blazi
>
>
>
>
>   -----------== Posted via Newsfeeds.Com, Uncensored Usenet News
==----------
>    http://www.newsfeeds.com       The Largest Usenet Servers in the World!
> ------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers
==-----




From phd at phd.russ.ru  Wed Jan 12 08:26:54 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 12 Jan 2000 13:26:54 +0000 (GMT)
Subject: checking for file existance
In-Reply-To: <JHh8OAp=9qpii6CdulrM3o2KSTd9@4ax.com>
Message-ID: <Pine.LNX.4.21.0001121324440.10184-100000@fep132.fep.ru>

On Wed, 12 Jan 2000, Anders M Eriksson wrote:
> How do I check if a file exist on a webserver? I have tried
> urllib.urlopen() but it never give any exception IOError.

   There is no such thing as "file-by-WWW". URL gives you a reference to an
object, and if a WWW server can't resolve the reference, it is not
mandatory to return the error 404 - WWW server may decide to return another
object (a page with words and picture saying you have got an error).

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From helge.hess at mdlink.de  Sun Jan  9 16:57:41 2000
From: helge.hess at mdlink.de (Helge Hess)
Date: Sun, 09 Jan 2000 22:57:41 +0100
Subject: no return values for __init__ ??
References: <3874CFE5.765A40EA@mdlink.de> <852jbk$9tt$1@nntp9.atl.mindspring.net> <3875099A.2CF9333C@mdlink.de> <855idr$hb3$1@nntp1.atl.mindspring.net> <3876B79F.30439283@mdlink.de> <3876dcee.547930242@news.isomedia.com>
Message-ID: <38790455.1BB6FC47@mdlink.de>

Eugene Goodrich wrote:
> I suppose if it bothers you that much, you could just stop using
> __init__. :)

I thought about that and indeed this is some kind of solution. The
problem is that this is very 'unpythonic' therefore making it more
difficult to understand:

  a = MyString().initWithData("skjdflkajdf", encoding='utf8')

works fine but looks obviously less nice in Python syntax than:

  a = MyString(data="sadfjsadf", encoding='utf8')

Note that a major topic in my complaint about '__init__' is that is
different from other methods.

> Seriously, when I was first learning Python, I didn't know about any
> of the "magic" functions, so I lived without __init__.  As far as I
> knew, all functions really were alike.  I side with Aahz on keeping it
> simple enough to sucker, uh, induct, no, uh, attract new users to
> Python.

I don't think it complicates the environment, but this is obviously a
matter of taste.

Greetings
  Helge
-- 
SKYRIX-OS  Web Application Environment - http://www.skyrix.com
SKYRIX-IAS Groupware Application Suite - http://www.skyrix.com


From aa8vb at yahoo.com  Tue Jan 11 11:49:24 2000
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 11 Jan 2000 11:49:24 -0500
Subject: [ANN] Multimethod.py -- multimethods for Python
In-Reply-To: <20000111113656.A42270@vislab.epa.gov>
References: <000801bf5b43$69184d40$532d153f@tim> <387B19F2.AC17FA8D@home.com> <20000111113656.A42270@vislab.epa.gov>
Message-ID: <20000111114923.B42270@vislab.epa.gov>

Randall Hopper:
 |I had the same question yesterday and did a little surfing.  Basically,
 |instead of switching off just the type of 'self' to determine which method
 |to call, you switch off the type of other arguments as well.
 |
 |If you think best in implementation as I do:
 |
 |   def f( a,b ):
 |     if   type(a) == Type1 and type(b) == Type1: f_t1_t1(a,b)
 |     elif type(a) == Type1 and type(b) == Type2: f_t1_t2(a,b)
 |     elif type(a) == Type1 and type(b) == Type3: f_t1_t3(a,b)

To be closer to the truth, I really should have said:

       if   issubclass( a, Type1 ) and issubclass( b, Type1 ): f_t1_t1(a,b)
       elif issubclass( a, Type1 ) and issubclass( b, Type2 ): f_t1_t2(a,b)
       elif issubclass( a, Type1 ) and issubclass( b, Type3 ): f_t1_t3(a,b)

This alludes to one way multimethods can get into method resolution
troubles.  If none of the multimethods' parameter signatures match the
argument types exactly, but multiple multimethods accepting compatible base
class objects as parameters exist, ...

-- 
Randall Hopper
aa8vb at yahoo.com



From gregm at iname.com  Thu Jan  6 08:52:30 2000
From: gregm at iname.com (Greg McFarlane)
Date: Fri, 7 Jan 2000 00:52:30 +1100
Subject: how to include a file -- not import??
In-Reply-To: <m3n1qj5or5.fsf@havenrock.com>; from Matt Gushee on 06 Jan 2000 at 08:11:10AM
References: <8515bs$2vn$1@nnrp1.deja.com> <2aVc4.277$mj7.5541@typhoon.nyroc.rr.com> <38744512.377982379@news.isomedia.com> <m3n1qj5or5.fsf@havenrock.com>
Message-ID: <20000107005230.23566@nms.otc.telstra.com.au>

On 6 Jan, Matt Gushee wrote:
> bitbucket at isomedia.com (Eugene Goodrich) writes:
> 
> > On Wed, 5 Jan 2000 23:52:46 -0500, "Darrell" <news at dorb.com> wrote:
> > 
> > >from filename import *
> > >
> > >Doesn't create another namespace. Avoid this in most cases.
> > >
> > 
> > What are the implications of your statement for Tkinter?  Is that one
> > of the exceptions?
> 
> Some packages, including Tkinter, are designed to be 'safe' to use
> with 'from filename import *'. All that really means is that they are
> careful not to export any unnecessary names, and that the names they
> do export are unlikely to conflict with other commonly-used names.
> 
> E.g., when you do
> 
> from Tkinter import *
> 
> you have class names like Frame and Canvas in the global namespace --
> which are unlikely to be found except in another GUI module.
> 
> However, I've seen code that does
> 
> import Tkinter
> 
> and qualifies all the names. Guess it depends what you're doing.

Tkinter is not "careful" enough, IMHO.  It exports, inter alia, the
following names:

    At E EW N NE NO NS NW ON S SE SW Tk W Wm X Y
    ALL END OFF SEL TOP YES
    BOTH CHAR FLAT Grid LEFT Menu Misc NONE NSEW Pack TRUE Text WORD

I'm not sure how you can say that "X" and "Y" are not "commonly-used
names".

-- 
Greg McFarlane     INMS Telstra Australia     gregm at iname.com



From sei at sunse.jinr.ru  Wed Jan 19 15:50:00 2000
From: sei at sunse.jinr.ru (Evguenii Smogailov)
Date: Wed, 19 Jan 2000 23:50:00 +0300 (MSK)
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <006901bf62bf$5faf1ad0$f29b12c2@secret.pythonware.com>
Message-ID: <Pine.SOL.3.91.1000119234852.8351A-100000@sunse>

On Wed, 19 Jan 2000, Fredrik Lundh wrote:
> > Generally for improving readability,
> > (and Perl compatibility).
> 
> oh, come on.  since when is "Perl compatibility" a
> guiding principle for Python's design?

Well, i mean all that 1 million Perl programmers that would like to learn 
Python someday:)

But the MAIN reason is READABILITY

evguenii




From news at dorb.com  Wed Jan 19 21:04:15 2000
From: news at dorb.com (Darrell)
Date: Wed, 19 Jan 2000 21:04:15 -0500
Subject: Best way to work in debug code?
References: <85umq4$tlr$1@nnrp1.deja.com>
Message-ID: <R%th4.907$VD4.25359@typhoon.nyroc.rr.com>

This debugging solution may just be over kill, so be kind. But working with
6+ developers all in one application, debugging can be tricky. We have
plugin developers also that want more trace info than a typical user. Like
turning on or off debugging for specific areas with out editing the whole
system.

I use debug levels in each module or subsystem.
They are set from the command line with something like
-d moduleName level
or
-d subSystemName level

The debug module that parses command line debug switches, adds an alias of
it's self in the sys.modules list. This debug module must be imported early
in the main so the debug alias gets setup before it's needed.

When a module is imported it does the following:
# Some quick last min changes so this is untested code.
#### module to debug
import sys
    # The use of __debugMain__  instead of the module name is just a level
of
    # indirection. Like a symbolic link.
sys.modules['__debugMain__'].setDebug(__name__)

    # debugLevel was set into this module by the setDebug call
if debugLevel >=1:
    # Do some debug stuff here

#### This is in __debugMain__ which was imported as something like debug.py

def setDebug(name):
    """
    Called from modules as they are imported to setup their debugLevel.
    name:: Name of the module to work with
    It's debugLevel is determined and written back to it.
    """
    global debugDic
    if debugDic.has_key(name):
            # Set the value of debug in the module.
        sys.modules[name].debugLevel= debugDic[name]
    else:
        sys.modules[name].debugLevel= 0

--
--Darrell





From phd at phd.russ.ru  Wed Jan 12 08:24:07 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Wed, 12 Jan 2000 13:24:07 +0000 (GMT)
Subject: Python and Oracle
In-Reply-To: <dB_e4.1177$hJ1.22100@news.rdc1.tx.home.com>
Message-ID: <Pine.LNX.4.21.0001121323240.10184-100000@fep132.fep.ru>

On Wed, 12 Jan 2000, Paul E. Bible wrote:
> Has anyone had success installing the Python Oracle Call Interface (OCI)
> library written by Digital Creations (DCOracle)?  I'm currently having

   Many people did that, including me.

> compile problems with the C modules, which is a library reference.  I just
> curious as to whether this library is worth the effort to resolve the
> current issues that I'm having with it's installation.
> 
> My platform is:
> Python           1.5.2
> Oracle           7.3.4
> OS                Unixware 7.0.1

   My is Linux and Oracle 8.0.5. No problems.

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From hei at adtranzsig.de  Tue Jan 18 05:44:20 2000
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 18 Jan 2000 11:44:20 +0100
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
Message-ID: <948195584.987404@news.adtranzsig.de>

tye4 schrieb in Nachricht <860ftk$bas at news.or.intel.com>...
>Indentation is a confusing and lame way to separate what is inside a block
>and what's outside it.

I'm irritated by the subject.
The indentation complaint re-occurs every week,
but this is the first time it comes packaged with the
subject line "Python rocks!"

Get used to it, and don't use tabs.

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




From fraley at usfca.edu  Tue Jan 18 11:18:14 2000
From: fraley at usfca.edu (Michael Fraley)
Date: 18 Jan 2000 16:18:14 GMT
Subject: Good Grief! Why does this work?
Message-ID: <slrn8893ll.67.fraley@cumquat.usfca.edu>

I was trying to run the cgi samples from "Internet Programming with Python"
and kept getting an attribute exception on os.environ. So by trial and error
I changed all references in month.cgi and genCGI.py from os.environ to
cgi.os.environ, and it works with the latest Python version now! (The book
refers to version 1.4). Can anyone tell me why it works? In other words,
what changed about the underlying cgi class, that referring to os.environ
from within genCGI.py no longer works?

sample snippet from genCGI
--------------------------

class cgiBase:
   """Superclass for developing CGI processing objects."""

   # some default parameters
   Content_length_limit = None # set to integer to limit content length
   Form_data_Timeout = 3       # set to None for no timeout

   traceback = 1 # set to zero to disable debug traceback dump.

   def __init__(self, infile=myStdin(sys.stdin), outfile=sys.stdout,
                      environ=cgi.os.environ):
					  #!!
					  #!! NOTE:
					  #!! This is where it bombed on environ=os.environ
					  #!!
       self.infile = infile
       self.outfile = outfile
       self.environ = environ



From da at ski.org  Mon Jan 31 17:00:02 2000
From: da at ski.org (David Ascher)
Date: Mon, 31 Jan 2000 14:00:02 -0800
Subject: DEADLINE EXTENDED: July O'Reilly Python Conference
Message-ID: <04ca01bf6c36$861b19a0$c355cfc0@ski.org>

If you're just back from the 8th International Python Conference and wish
that you'd had a chance to present your work at that (superb!) conference
while the blizzard was shutting down the federal US government, how about
presenting your project/ideas/results in sunny Monterey, California in July
instead?  If you couldn't make it to IPC8 but have something you're proud
of, now's your chance to show off to Python experts and refugees from other
language communities as well.

We're still accepting applications for paper presentations (case studies,
discussion panels, etc. -- 30 minutes), tutorials (3 hours), or other types
of presentations.  Please see the call for participation at
http://conferences.oreilly.com/oscon2000/call.html or see text version
below.

=================================================
IMPORTANT: The deadline has been extended to at least 2/15/00.
=================================================

Any questions regarding the Python track can be sent to mailto:da at ski.org or
to the Conference chair, Joseph McIntyre (see below).

Looking forward to reading your submissions,

David Ascher & the rest of the program committee

PS: presenting a paper makes attending the conference a lot cheaper!

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

O'Reilly Open Source Software Convention
July 17-20, 2000
Monterey, California

Call for Participation


O'Reilly & Associates is pleased to announce the 2nd annual Open Source
Software Convention. This is the largest cross-technology gathering of the
open source community. It features 13 concurrent tracks focused on Linux,
Perl, Apache, Python, Sendmail, Tcl/Tk, and Open Source Business Strategies.
Last year's event attracted over 1700 attendees. The Perl Conference 4.0 is
a 'conference-within-a-conference' at this event.

The Open Source Convention is the central gathering place for the open
source community to exchange ideas, techniques, and to advance the open
source software. The Convention is a four-day event designed for
programmers, developers and technical staff involved in or interested in
open source technology and its applications. The Convention will be held at
the Monterey Conference Center in Monterey California, July 17-20, 2000.

The Convention includes two days of intensely focused tutorials aimed mainly
at intermediate and advanced users of open source technology. These
tutorials are designed to provide concrete knowledge that leads directly to
better application of open source tools. Two days of multi-tracked
conference sessions focus on the cutting edge of open source technology and
feature talks, demonstrations, and panel debates.

Technically sophisticated individuals who are actively working with open
source technology make all tutorial and conference presentations.
Presentations by marketing staff or with a marketing focus will not be
accepted.

Important Dates

     Proposals due: February 1, 2000
     Notification to conference and tutorial presenters: March 1, 2000
     Tutorial presentations due: May 1, 2000
     Conference presentations due: June 1, 2000

Participation Opportunities

Individuals and companies interested in making technical presentations at
the Convention are invited to submit proposals to the Convention organizers
following the guidelines below. Proposals will be considered in three
classes: tutorial programs, conference presentations, and refereed papers.
The program committees for the appropriate Convention track will evaluate
all proposals.


Topics

A wide variety of topics will be considered for the following tracks:
     Linux
     Apache
     Python
     Open Source Business Strategies
     Tcl/Tk
     Other open source technologies

This year the Convention organizers are focusing on the open source
technologies enabling secure, high-volume e-commerce. Technical talks in any
of the above tracks that focus on e-commerce will be particularly welcomed.

Presenters interested in Perl are directed to review the separate call for
participation for the Perl Conference 4.0.


Tutorial Presentations

The first two days of the conference are devoted to tutorial presentations.

Tutorial proposals must include:
     Tutorial Title
     Tutorials length (3 or 6 hours)
     Target audience including any prerequisites for taking advantage of the
material
     What attendees will learn--the knowledge or skill take-away attendees
will have at the completion of the tutorial
     Tutorial outline--a short bullet list or description of the course
itself and the material it covers
     Speaker name
     Speaker biography
     Complete speaker contact information; if applicable, include
administrative support contact information


Conference Presentations

Conference presentations and participation will beconsidered for these types
of talks:
     Case studies showing an application of open source technology, the
problems that were solved and the lessons learned
     Focused technology discussions covering a specific open source
technology and showing a variety of implementation or application options
     New product/technology demonstrations--live demonstrations of released
or near released open source software
     Panel participation--speakers to participate in one of the various
panel discussions that are part of the conference

Conference proposals must include:
     Type of talk--technology, new product, case study, or panel
     Title of talk or demonstration
     Requested talk length: 30, 45, 60, or 90 minutes
     Abstract of talk, maximum 250 words
     Speaker name
     Speaker biography
     Complete speaker contact information; if applicable, include
administrative support contact information

Submitting Your Proposal

Please send all proposals to:
oscon00proposals at oreilly.com
(Note: use this address to submit proposals to the Open Source Convention
only; see the instructions for the call for proposals for the Perl
Conference 4.0 for submission to the Perl tracks of the Convention).

Submit all proposals via electronic mail in plain text with no attachments.
Submit one proposal per email.

The subject line of your email must follow this format:
Track: Your Last Name: Proposal Type (tutorial, presentation): proposal
title
For example: Python: Smith: Tutorial: Advanced Python Programming. You will
receive an automatic confirmation of receipt for each proposal.

Hard copy confirmation may be sent via mail or fax to the address below.
Address all questions regarding proposals and proposal status to the
Convention Director.

     Joseph McIntyre
     Director of Conferences
     O'Reilly & Associates
     101 Morris St., Sebastopol, CA 95472
     joseph at oreilly.com
     Phone: (707) 829-0515 x372
     Fax: (707) 823-9746





From tim_one at email.msn.com  Thu Jan 20 01:49:15 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 20 Jan 2000 01:49:15 -0500
Subject: Python Rocks!
In-Reply-To: <slrn88d6g5.13f.wtanksle@hawking.armored.net>
Message-ID: <000d01bf6312$77edd660$31a2143f@tim>

To tye4's bold rhetorical ploy:

>> There is not single programming language that forces
>> such a bizarre indentation rule.

William Tanksley makes a futile rejoinder:

> Yes, there is -- Python.

Futile because, alas, it now appears open to question whether Python really
is a "programming language" <wink>.

AFAIK, the first programming language to use indentation to denote block
structure was Landin's ISWIM (If You See What I Mean).  That was about 30(!)
years ago.  It was very influential at the time, although ISWIM was a
functional language and so outside "the mainstream".  Its influence persists
in modern functional languages, which *still* usually use indentation to
denote block structure.  For example, look up Haskell.  That also has an
optional notion of explicit block closers (in fact, the all-whitespace form
is defined as syntactic sugar for the explicit form), but it's *so* optional
<wink> that I don't recall ever seeing a Haskell program use explicit block
markers (except in the language reference manual).

Python got its treatment of whitespace from ABC, definitely not a functional
language.  I don't know whether ABC's designers were influenced by the
indentation tradition in the functional world, or thought of that on their
own.

In a third line of development <wink>, a friend designed a turtle graphics
language to help his son learn how to program (LogoMation; see

http://www.magicsquare.com/LM2/

).  He was not a functional language fan, nor had he used Python; his
favorite language appeared to be Prolog.  Simply by watching what his son
got confused by, he apparently hit upon the idea of indentation == block
structure on his own, and that's what LogoMation does.  Its "industrial
strength" successor L3 (http://www.magicsquare.com/L3/ -- impressive web
technology) also uses this scheme.

So, programming language or not, Python isn't the first or the only, by
decades or by a long shot.

although-it-may-be-the-last-ly y'rs  - tim





From calishar at my-deja.com  Fri Jan 21 16:43:03 2000
From: calishar at my-deja.com (calishar at my-deja.com)
Date: Fri, 21 Jan 2000 21:43:03 GMT
Subject: Installer and COM client
Message-ID: <86ajt3$lsk$1@nnrp1.deja.com>

Hello again folks,

  Well, I seem to keep running into odd little things, this one is a
real puzzler though, and I'm hoping someone can help me out.

  I have created a script which opens an Excel workbook, reads
information off one of the sheets, and exports it to a different file
(and filetype). As long as it is a script, it's fine...but (you knew it
was coming) my client doesn't want to have to install python on all
their systems, so I am trying to compile it.

  Now it is crashing as soon as it has to convert on of the numbers
into a float. I have added in a line to print out the variable, and its
type, under python I get the following

673 <type 'string'>

when I run the version generated by installer (using simple) I get

<PyUnicode: '673'> <type 'PyUnicode'>

Why is it suddenly turning things into PyUnicode? How can I pull things
out of PyUnicode? Should I stop what I am doing and look at totally
different things to do?

Calishar

P.s.  As always, thanks in advance for any help on this


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


From Lance.Ellinghaus at avnet.com  Wed Jan 19 17:51:47 2000
From: Lance.Ellinghaus at avnet.com (Lance Ellinghaus)
Date: Wed, 19 Jan 2000 14:51:47 -0800
Subject: Mapping Column Names to results using DBI
Message-ID: <8825686B.007D977F.00@avnet.com>

I had the same need and this is what I did. I did my query using the cursor
(cur) and then did the following code.
This code would then return an array of dictionaries. Each dictionary would be
one returned record and each of the
column names would be elements in the dictionary.
I am using the DCOracle module, but it should be pretty similar.

    dsc = cur.description
    l = len(dsc)
    lst = cur.fetchall()
    rtn = []
    for data in lst:
     dict = {}
     for i2 in range(l):
         dict[string.lower(dsc[i2][0])] = data[i2]
     rtn.append(dict)
    return rtn

-------From: Timothy Grant  on 01/19/2000 4:37:22 PM-------

From: Timothy Grant <tjg at avalongroup.net>
To: Python People <python-list at python.org>
cc:
Subject: Mapping Column Names to results using DBI






Hi again,

Right now I have an app that works ok, but I forsee a problem in the
future.

As you know, the results from a database query are returned in a tuple.
I can access those results by position just fine, is it possible to map
the column names onto the results, so I can access columns by name
rather than by number?

While indexing by number works, it is incredibly fragile and will start
failing the minute the underlying DB changes.

Once again, it appears to me from the DBI spec that there should be a
way to do this, I just can't figure it out from the spec.

I'm using mxODBC if that helps.

Thanks.
--
Stand Fast,
    tjg.

Chief Technology Officer              tjg at exceptionalminds.com
Red Hat Certified Engineer            www.exceptionalminds.com
Avalon Technology Group, Inc.                   (503) 246-3630
>>>>>>>>>>>>EXCEPTIONAL MINDS, INNOVATIVE PRODUCTS<<<<<<<<<<<<

--
http://www.python.org/mailman/listinfo/python-list





From mikael at isy.liu.se  Tue Jan 18 04:51:35 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Tue, 18 Jan 2000 10:51:35 +0100 (MET)
Subject: Python Rocks!
In-Reply-To: <860ftk$bas@news.or.intel.com>
Message-ID: <XFMail.000118105135.mikael@isy.liu.se>

[tye4 going through the "indentation sucks" and "I miss end tags"
 stuff that we've heard before]

I've said this before: This is one of the things that got me hooked on 
Python. I was already used to indent the Python way before I knew 
about Python, and I used to forget all these start and end tags (like {} 
or whatever) in the languages I used then, probably because they were 
forced on me by the languages, and because they didn't help me read the
code. Today, with Python, I tend to forget the ":" in ifs, fors, whiles, 
and so forth, for the same reason. However, I do not write to this list
and ask why the ?"#@ there is a ":" that does not make sense to me. I'm 
sure that there's a good reason, and if there were no good reason for it 
to be there from the beginning, there still is one now: Backwards 
compatibility.

I agree with the others who answered your posting: 

Try it, you'll like it.

And if you do not like it after you've tried it for a while, there is 
probably some other language that makes you feel better. Every language 
has its own flavour, and the indentation idea is a fundamental part of 
the Python flavour. 

I'm-glad-not-all-people-are-the-same'ly y'rs

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    18-Jan-00
Time:    10:17:21

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From thomas at bibsyst.no  Mon Jan 10 06:41:35 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Mon, 10 Jan 2000 12:41:35 +0100
Subject: Filelisting in Tkinter
Message-ID: <3879C56F.FF577344@bibsyst.no>

Hi,

Is there a widget suitable for filelisting in Tkinter?

I got a filename, size, modification date etc. that I want to put into a
grid-like list, with no visible lines bewteen "cells". Similar to any
decent filemanager.

Any tips?

Thomas


From bradh at mediaone.net  Thu Jan 27 23:20:36 2000
From: bradh at mediaone.net (Brad Howes)
Date: Fri, 28 Jan 2000 04:20:36 GMT
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
References: <1392272911@nds-3.azg.nl>
Message-ID: <m3d7qmzv0j.fsf@h0050047c4794.ne.mediaone.net>

"P.J.W.S. Vrijlandt" <P.J.W.S.VRIJLANDT at INT.azg.nl> writes:

> According to the manual,  hasattr is implemented by calling getattr 
> and seeing if an exception occurs, so hasattr will not be faster than 
> try...except...

A quick look at the source reveals that the builtin hasattr indeed uses the
object method getattr. However it clears any exception that getattr may post
for the Python interpreter loop to handle before returning, so there's no
exception processing being done. I bet hasattr is faster.

Brad

-- 
    "Were people this stupid before TV?" -- _White Noise_ by Don DeLillo


From Jim.Tittsler at tokyopc.org  Mon Jan 31 21:17:31 2000
From: Jim.Tittsler at tokyopc.org (Jim Tittsler)
Date: 1 Feb 2000 02:17:31 GMT
Subject: Zope 2 on Starship, when?!
References: <389198D1.8C0A0BCB@darwin.in-berlin.de>
Message-ID: <slrn89cgih.rrf.Jim.Tittsler@twin.dskk.co.jp>

On Fri, 28 Jan 2000 14:25:37 +0100, Dinu C. Gherman wrote:
>At IPC8 I've asked several people when we'll finally have some
>working version of Zope 2 running on the starship with a tiny
>bit of information telling people what to do in order to use 
>it?

The best place to ask is probably the Starship zope-users mailing list.
http://starship.python.net/mailman/listinfo/zope-users

Andrew Kuchling had 2.1.2 running on port 8080 and was encouraging people to
switch from the Zope1 installation by the end of January.


From andrew.holt at uk.sun.com  Thu Jan 20 04:24:49 2000
From: andrew.holt at uk.sun.com (Andrew Holt)
Date: Thu, 20 Jan 2000 09:24:49 +0000
Subject: Compiling Python on Solaris 7
Message-ID: <3886D461.A83008FF@uk.sun.com>

Hi all,

I am having a problem building pythne for Solaris 7 & NetBSD.  

I have built "vanilla" python on both these platforms.

My problems begin when I try to add the tkinter stuff.


After adjusting the Setup file and then running make 
I get the following message:



cd Modules;  make OPT="-g -O2" VERSION="1.5" \
                prefix="/usr/local" exec_prefix="/usr/local" \
                LIBRARY=../libpython1.5.a link
gcc   python.o \
          ../libpython1.5.a -L/usr/local/lib -ltk8.2 -ltcl8.2 -L/usr/openwin/lib
-lX11  -lsocket -lnsl -ldl -lm  -o python 
Undefined                       first referenced
 symbol                             in file
Py_Main                             python.o
ld: fatal: Symbol referencing errors. No output written to python
*** Error code 1
make: Fatal error: Command failed for target `link'
Current working directory /opt/home/andrewh/src/python/Python-1.5.2/Modules
*** Error code 1
make: Fatal error: Command failed for target `python'

I have just done the same on the NetBSD box, so I can at least start to learn
the language.  However Solaris is going to be my target so I need to get this
going.  

Any help will be much apreciated.

Rgds Andrew


From jeremiah at widomaker.com  Thu Jan 27 23:08:09 2000
From: jeremiah at widomaker.com (Jeremiah Rogers)
Date: Fri, 28 Jan 2000 04:08:09 GMT
Subject: how do i get user input?
Message-ID: <389117E3.748B44ED@widomaker.com>

I hate to ask this as it's almost certainly in the documentation, but
what function(and module) do i use to get user input? I've looked and
looked and i can't find it. Thanks a bunch for any help.


From rickhigh at aol.com  Fri Jan 21 05:23:40 2000
From: rickhigh at aol.com (RICKHIGH)
Date: Fri, 21 Jan 00 10:23:40 GMT
Subject: JPython to be supported by NetBeans!
Message-ID: <mt2.0-27365-948450220@news.informatik.uni-bonn.de>

NetBeans (Forte CE) to support compiling JPython code in IDE. 

NetBeans said the above at the Java Live dicussion regarding "Forte for Java
Community Edition." at JavaTM Live during the Fort? for Java Community Edition
January 11, 2000.

The speakers Christine Bodo (Christine Bodo), Ian Formanek (Ian Formanek).

http://developer.java.sun.com/developer/community/chat/JavaLive/2000/jl011
2-print.html

Partial Transcript:

Rick Hightower (Question): 
 "A few weeks back you had a poll on your NetBeans website that spoke about
adding support for JPython. 

   What kind of support do you plan to add for JPython? 
   (Do you plan) scripting for NetBeans IDE? (GOOD) 
   (Or) An extension that compiles and runs JPython? (BETTER) 
   (JPython is a free language that compiles and runs in the JavaTM Virtual
Machine1(JVM)"

Ian Formanek (Answer): 

"In response to Rick: With this poll we were evaluating what scripting
languages people use most. The addition of scripting to the IDE is planned as
an extension, which would address *both* of the usages you mentioned. Actually,
once you do one, the second is for free. "
(End Transcript)

Is that Cool Beans?

Check out the full transcript:
http://developer.java.sun.com/developer/community/chat/JavaLive/2000/jl011
2-print.html

Additional 

Rick Hightower: 

"Excellent, I am glad to hear that you are adding JPython support for Fort? CE.
I know I will use it. I think JPython is great! Having an IDE that supports it
and allows compiling and integrating is great (idea). Do have plans to support
any other languages that compile or run in the JVM? Rhino (JavaScript)?, Bistro
(Smalltalk)?, Skij (Scheme)?, etc." 
"Also there are versions of JSP that work with Python (JPython) and JavaScript.
Will your JSP support other JSP languages?"

Ian Formanek: 
"Regarding Rick's question about other languages: there are currently no plans
to provide such support in the basic version, but one of the nice effects of
the Open APIs is that any 3rd party, being it a company or a Linux opensource
team or a single developer, can very easily write new add-on modules to the
IDE. And the stuff you mentioned is very good example of things that may be
supported by other parties."

End Transcript

BTW For those who do not know Netbeans is a Java IDE maker that was bought out
by Sun. Sun is offering a free IDE via NetBeans called the Forte Community
Edition (CE) IDE (or something like that). They had a poll a few weeks ago
about scripting languages and JPython won by a landslide. To find out more
about netbeans and Forte CE go to http://www.netbeans.com/.

Does anyone know when NetBeans will add this support?

--Rick Hightower

-- 
----------- 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 cole at ccdc.cam.ac.uk  Mon Jan 24 11:42:43 2000
From: cole at ccdc.cam.ac.uk (Jason Cole)
Date: Mon, 24 Jan 2000 16:42:43 +0000
Subject: tkinter problems
Message-ID: <388C8103.26B810AE@ccdc.cam.ac.uk>

Hi All,
	I'm sure you all think this is yet another example of someone not being 
able to figure out how to edit autoexec.bat on Windows 95, but we a
bizarre problem
to do with good old "init.tcl" and tkinter (I found > 100 messages about
this when
searching comp.lang.python, so I'm scared I may not be seeing wood for
trees %)

	We have an installation script that attempts to install Tcl/Tk, and
it makes some edits to autoexec.bat: If it encounters the environment
variable TCL_LIBRARY
in the current autoexec.bat, it doesn't overwrite it, but appends the
new search path
into the environment variable, i.e. say it contained (as ours does ...
;-)

SET TCL_LIBRARY=C:\PROGRA~1\OTHER_APPLICATION\TCL\LIB\TCL8.0

then the install code would then add another line as follows after the
first one

SET
TCL_LIBRARY=%TCL_LIBRARY%;C:\PROGRA~1\CCDC\CONQUEST_NEW\TCL\LIB\TCL8.0

Now, it seems that when we do this, we get a message as follows

Traceback (innermost last):
  File "lib\csds.py", line 560, in ?
    a = New()
  File "lib\csds.py", line 31, in __init__
    Tk.__init__(self, parent)
  File "C:\PROGRAM FILES\CCDC\CONQUEST\PYTHON\lib\lib-tk\Tkinter.py",
line 886,
in __init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
   
{C:\PROGRA~1\OTHER_APPLICATION\TCL\LIB\TCL8.0;C:\PROGRA~1\CCDC\CONQUEST_NEW\TCL\LIB\
TCL8.0} {} ./lib/tcl8.0 ./lib/tcl8.0 ./library ./library
./tcl8.0/library ./tcl8
.0/library

It would seem to us that tkinter does not break up the path in the
conventional way, but
treats the semi-colon concatenated list as a single entity .... is this
the case? If so,
why, and what is the alternative to using this method. (we'd really like
to sandbox our
tcl/tk on Windows95 from other junk that may or may not be present in
the system, like
you can do for unix systems, but we're not sure how to do this in a
*sophisticated* way)

Jason Cole & Ian Bruno


From Pollarda at Lextek.com  Mon Jan  3 20:54:51 2000
From: Pollarda at Lextek.com (Art Pollard)
Date: Mon, 03 Jan 2000 18:54:51 -0700
Subject: Quick Python v.s. Learning Python
Message-ID: <387152EB.10DD0E17@Lextek.com>

I just purchased Programming Python.  As I am 
totally new to Python, I learned from this book
that I want to learn more.  Unfortunely, I don't 
think this is the best book for a Python newbie.
(Though from what I've seen, it makes a good 
language reference.)

So right now, I am looking for another learning
book.  The two which popout are:

Quick Python (In my local bookstore)
Learning Python (Not in stock but obtainable)

Does anyone have any preferences / recommendations???

I have extensive C/C++ experience but would like a good
book to help me break through the wall and reach that
Ahhh Haaaa experience when the language becomes clear.
(I still remember the Ahhhh Haaa from many years ago 
when I figured out C's pointers... What a day. :-)

-Art


From fredrik at pythonware.com  Fri Jan 14 03:40:04 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 14 Jan 2000 09:40:04 +0100
Subject: Parser Generator for Python
References: <387E8B81.23C91488@attglobal.net>
Message-ID: <009f01bf5e6a$f93b7170$f29b12c2@secret.pythonware.com>

Wolfgang Roesner wrote:
> I haven't found a good reference so far for a parser generator (a la
> ANTLR) that produces Python code. Is there a robust generator available
> anywhere?

http://starship.python.net/crew/amk/python/string.html
(look under "parsing")

Yapps and Spark appear to be among the more popular:
http://theory.stanford.edu/~amitp/Yapps/
http://www.csr.uvic.ca/~aycock/python/

</F>




From nickb at earth.ox.ac.uk  Fri Jan 21 09:53:03 2000
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Fri, 21 Jan 2000 14:53:03 +0000
Subject: Morse Test (was: Whitespace delimiters suck)
References: <1263658532-14676484@hypernet.com>
Message-ID: <388872CF.232CD98C@earth.ox.ac.uk>

50 years ago in my early days of programming at Bletchley Park....

we ...uhm... had a ..er ...Morse Test...but I forget the details.

Anyway - indentation was the only thing that worked!

enigma-tically-yours

Nick/Oxford




From linli at computer.org  Thu Jan 27 12:45:50 2000
From: linli at computer.org (lin li)
Date: 27 Jan 2000 17:45:50 GMT
Subject: Any graph editors?
References: <tq9oga7cx8m.fsf@Twister.MI.Uni-Koeln.DE>
Message-ID: <86q08e$27e$1@pollux.ip-plus.net>

Thank you very much, Alexander. I had a look at the editor and it is an
very good 
starting point for me. Now I just have to add differently shaped nodes,
connection
restrictions, bending edges, hand layouts, etc. Simple enough :-).
Thanks again. 
Lin

Message by Alexander Schliep <schliep at Octopussy.MI.Uni-Koeln.DE> on: 1/27/00 5:09:27 PM
>lin li <linli at computer.org> writes:
>> scratch. So here is the question: is there any implementation of
>> some graph editors in Python, JPython (or maybe Java) that I can
>> copy and modify?
>
>Dear Lin,
>
>Gato <http://www.zpr.uni-koeln.de/~gato> contains Gred -- the GRaph EDitor
>-- which is written in Python/Tkinter. The whole thing is released under
>the LGPL.
>
>Yours,
>Alexander
>
>-- 
>Alexander Schliep                    schliep at zpr.uni-koeln.de
>ZPR/ZAIK                             Tel: +49-221-470-6040 (w) 
>University of Cologne                FAX: +49-221-470-5160
>Weyertal 80                          http://www.zpr.uni-koeln.de/~schliep
>50931 Cologne, Germany               
>


From Gary.Pennington at uk.sun.com  Fri Jan 21 14:27:58 2000
From: Gary.Pennington at uk.sun.com (Gary Pennington)
Date: Fri, 21 Jan 2000 19:27:58 +0000
Subject: POST processing/mime problem
Message-ID: <3888B33E.A77559C@uk.sun.com>

Hi,

I've written some CGI code in Python to serve up a file in response to a
POST request from a browser.

I'm trying to do something like this :-

            zipFile=open("da200.zip","rb")
            dataObj=zipFile.read()
            zipFile.close()
            print "Content-Type: application/zip\n"
            print dataObj

The browser picks up the response and correctly identifies that it is an
application/zip data stream. However, I then have two problems.

1. The file is corrupted at the client. It is larger and thus I guess
that the browser has done a CRLF transform on the data or something like
that. How do I stop this happening?

2. The script is called download.cgi and the dialog popped up by
Netscape prompts you to save the downloaded file as download.cgi. How do
I control this behaviour? Is there a header which you can send to
specify the suggested file name?

I know these aren't really Python questions, but all the code is Python.
Any suggestions gratefully received.

Yours,

Gary



From gmcm at hypernet.com  Wed Jan 19 00:00:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 19 Jan 2000 00:00:32 -0500
Subject: More DBI question
In-Reply-To: <3885087E.91A8A205@exceptionalminds.com>
Message-ID: <1263864373-2295484@hypernet.com>

Timothy Grant wrote:
 
> Thanks Gordon, I appreciate the input. I'm am now going through my code
> and closing all cursors after I get the result table.
> 
> As I stop to think about it though, If I were to need to make a
> different query mid-fetch(), I would then need to open a separate cursor
> and use it. 

And hope that the driver supports multiple open cursors (a few 
widely used ODBC drivers don't).

> I'm significantly new enough to this type of DB programming
> that I just don't know what is good style and what is not.

Whatever you do, don't even THINK about isolation levels 
<wink>.

- Gordon



From Jean-Daniel.Aussel at bull.net  Thu Jan 20 11:58:08 2000
From: Jean-Daniel.Aussel at bull.net (Jean-Daniel Aussel)
Date: Thu, 20 Jan 2000 17:58:08 +0100
Subject: Changing the passed arguments value after returning from a module
Message-ID: <867ebt$d5i2@tosca.frcl.bull.fr>

Hi there,

I'm trying to write a C module that changes the values of a passed argument,
which I am sure is a classic problem. The Python script would look as:

    import mymodule
    i=0
    mymodule.test( i )
    print i

where i would hold a meaningful value on return. Problem is I can't figure
out how to do it. This is my latest try, which still does not work:

    static PyObject* sct_tt( PyObject* zself, PyObject* args)
    {
        PyObject* o;
        int i;

        if (!PyArg_ParseTuple(args, "O", o ))
            return NULL;
        PyTuple_SetItem(o, 0, PyInt_FromLong(101L));
        if (PyErr_Occurred()) {
            printf("kaboom!\n");
        }
        result=0;
        return Py_BuildValue("i", result );
    }

Any hints or help would be welcome. Regards,

/jda




From warlock at eskimo.com  Fri Jan 14 00:23:54 2000
From: warlock at eskimo.com (Jim Richardson)
Date: Thu, 13 Jan 2000 21:23:54 -0800
Subject: converting tuple to list
References: <000701bf5dcd$f2b6d350$3acbd9c2@peridot.optichrome.com>  <004a01bf5dd0$a77d9830$f29b12c2@secret.pythonware.com>
Message-ID: <slrn87tcna.vj.warlock@raven.myth>

On Thu, 13 Jan 2000 15:14:59 +0100, 
 Fredrik Lundh, in the persona of <fredrik at pythonware.com>,
 brought forth the following words...:

>Adrian Eyre <a.eyre at optichrome.com> wrote:
>> Just out of interest: what is the canonical way of converting a list
>> of tuples to a dictionary? i.e. the opposite of dict.items()
>> 
>> [("key1","value1"),("key2","value2")] -> {"key1":"value1","key2":"value2"}
>
>I doubt you can do much better than:
>
>mydict = {}
>for k, v in mylist:
>    mydict[k] = v
>
></F>
>

thankyuothankyou this solves a niggling problem I was having _so_
much more elegantly than I had thought of myself, new to python,
and loving every byte...

-- 
Jim Richardson
	Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
	Linux, because life's too short for a buggy OS.



From wtanksle at hawking.armored.net  Sun Jan 23 21:02:40 2000
From: wtanksle at hawking.armored.net (William Tanksley)
Date: 24 Jan 2000 02:02:40 GMT
Subject: What about REBOL (www.rebol.com)
References: <000901bf646e$699de980$caa0143f@tim>
Message-ID: <slrn88ndbj.p0q.wtanksle@hawking.armored.net>

On Fri, 21 Jan 2000 19:19:55 -0500, Tim Peters wrote:
>[William Tanksley]
>> Heck, not even discovering Joy
>> (http://www.latrobe.edu.au/www/philosophy/phimvt/j00syn.html),
>> surely the best language ever designed, has not sullied my
>> happiness with Python.

>[Michel Pelletier]
>> Weird, it looks like the bastard child of a paren-less LISP
>> and an inverted FORTH.

>Yes, Billy *does* leave that impression on people <wink>. 

I'm keeping this one.  Thanks, Tim -- made my day.  :-)

Although I'm a bit confused by the specific charicterization chosen --
Joy's more like a cross between Forth and an inverted, paren-less Lisp.

>As you say, though, Joy looks pretty:
>> Cool.
>for-austerely-elegant-defns-of-coolness-ly y'rs  - tim

I actually wouldn't care to write anything significant in the language
itself -- but reading the docs has changed my understanding of programming
languages.

I'm thinking of writing a Forth based on Joy, and then a Joy based on that
Forth derivative.  (How will you tell the difference?  Simple: the former
won't have GC or types, but will instead offer direct access to its VM.)
My thought is that while Joy reveals much of the math which is behind
Forth's usefulness, Forth has gained a lot of experience in being a real
programming language, experience which Joy lacks.

I also have contemplated making the second language Rebol-like (via
careful choice of quoting semantics, types, parsing strategies, and
libraries).

Hmm, I think I'll make it whitespace sensitive.  Yummy -- RPN _and_
whitespace.

-- 
-William "Billy" Tanksley, in hoc signo hack


From gerrit.holl at pobox.com  Thu Jan  6 08:31:30 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Thu, 6 Jan 2000 14:31:30 +0100
Subject: Who hosts CGI/Python?
In-Reply-To: <3873EE03.1CE5F8CC@usit.net>; from echuck@usit.net on Wed, Jan 05, 2000 at 08:21:07PM -0500
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <20000106143130.A4003@stopcontact.palga.uucp>

Chuck Esterbrook wrote on 947100067:
> Seems like many web host providers, such as MindSpring, offer CGI with
> Perl, but not other languages, notably Python.
> 
> Does anyone know who would host a CGI Python site?

I think vex.net's doing that... I don't think parnassus runs on perl ;-)
For opensource projets, www.nl.linux.org will host your project. It runs
Python 1.5.2. with HTMLgen, more on demand. It's ONLY FOR FREE PROJECTS.

regards,
Gerrit.

-- 
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From thor at localhost.localdomain  Wed Jan  5 06:37:37 2000
From: thor at localhost.localdomain (Manuel Gutierrez Algaba)
Date: 5 Jan 2000 11:37:37 GMT
Subject: Python ring propaganda
Message-ID: <slrn876erg.18q.thor@localhost.localdomain>

Python ring propaganda
-------------------------

This is a piece of information about the existing (La)TeX ring in:
http://www.ctv.es/USERS/irmina/pythonring.html

What is it?
>From a functional point of view, it's a list of linked web sites
that share a common interest or information, in this case TeX
related stuff.

How does it work?
The HTML code included in each node of the ring calls a CGI that
performs the movement thru the ring.

How can I join?
In the above web site , fill in your data and that's all.

Why I should ? 
If you've done a web site, I think it's worth you spend 15 minutes
promoting it, after your web site has been included in it you'll get
more hits than ever.
Moreover, your information may be interesting for other people, or
you can be interested in other people's work, a big Webring provides
the community a lot of unusual but somehow interesting data.

How can I get more hits? 
Because the Webring is a democratic way of organization, that is,
all its members tend to receive the same ammount of visits, the
big ones as much as the small ones.

Well, I don't have a Web page.., so?
So  you must (  :) ) start doing one right now. Seriously, we
( at least me,MGA) are interested in people's work, perhaps you've
been collecting data or even your libraries or methods of work are
clearly defined, the ammount of time required for a fast compilation
of that info is not too much. And I remark that you can learn only
if all ( or many) want to show or teach what they know.

Of course we admit commercial Web sites also, *any* Web site about
python.

Thanks for your time and patience reading this.
MGA





From gklix at hassler.de  Tue Jan 25 13:03:20 2000
From: gklix at hassler.de (Gerald Klix)
Date: Tue, 25 Jan 2000 19:03:20 +0100
Subject: New install. Getting ImputError
References: <388AAE5A.B20A7BCE@slip.net> <OgJxCSaZ$GA.224@net025s>
Message-ID: <388DE568.126268B2@hassler.de>

Vinnieboy wrote:

> Hi, maybe I can help,
>
> I'm a newbee too, but..
> I think vtest must be a function (or?????) He can't find vtest because there
> isn't a function called like that. So create a function called vtest, and in
> that function your typing:
> print 'this is a test'
> In the code from your program you'll have to write..
>
> vtest ()
>
> Greetz from Vinnieboy (the Netherlands),
>
> -I know, my English isn't very well.
>
> Thomas D'Urso <tjdurso at slip.net> schreef in berichtnieuws
> 388AAE5A.B20A7BCE at slip.net...
> > Just installed / started with Python and reading Learning Python. When I
> > try to execute this code:
> >
> >    import test
> >    from test import vtest    # <-- error on this line
> >    print vtest
> >
> > file "test" contents:
> >
> >    vtest = 'this is a test'
> >
> > I get debugger error for 2nd line:    ImputError: cannot input name
> > vtest
> > The global variable says ImputError is:     Input can't fine module, or
> > can't find name in module.
> >
> > It give no other errors as far as opening the file. Working on a Mac G4
> > with OS 9. Python 1.5.2.
> > Doing the "import test.autotest" reports no errors.
> >
> > TIA Thomas
> >
> >
> >
> >

If the file name is realy "test", python will not find it, you have to rename
it to "test.py" to make things work.





From neelk at brick.cswv.com  Thu Jan  6 19:37:29 2000
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 7 Jan 2000 00:37:29 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net> <14452.45436.641890.197509@beluga.mojam.com>  <005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com>
Message-ID: <slrn87ad9u.hp9.neelk@brick.cswv.com>

Darrell <darrell at dorb.com> wrote:
> From: "Skip Montanaro"
> >
> > I find nothing wrong with it.  Some people object to using try/except for
> > anything but handling errors however.
> 
> Don't know if this if fair. But exceptions are slow when raised.

Random question: why?

When I tried timing a dummy loop with 100,000 exceptions versus 100K
function calls, I found that the exceptions took roughly 3 times
longer. (On CPython 1.5.2)

This doesn't make a whole lot of sense to me. The exceptions shouldn't
have to create new stack frames, it just has to deallocate an old one
and climb back up the stack. The function call should create a new
stack frame, deallocate it, and then climb back up the stack.

A quick examination of the source doesn't seem to disabuse me of this
notion (except some stack frames are saved), but that would seem to
affect function calls and exceptions more or less equally. Is it
because the function calling code is inlined into eval_code_loop() and
do_raise() isn't?

The code:

  def exception_test():
    r = range(100000)
    t1 = time.time()
    for i in r:
      try:
  	raise "Error"
      except:
  	pass
    t2 = time.time()
    return t2 - t1

Median of 5 runs: 1.7
  
  def call_test():
    r = range(100000)
    f = lambda: 3
    t1 = time.time()
    for i in r:
      f()
    t2 = time.time()
    return t2 - t1

Median of 5 runs: 0.45


Neel


From jbauer at rubic.com  Tue Jan  4 17:32:56 2000
From: jbauer at rubic.com (Jeff Bauer)
Date: Tue, 04 Jan 2000 16:32:56 -0600
Subject: Python & Apache
References: <A4uc4.411$hJ1.3104@news.rdc1.tx.home.com> <387272A1.F036408C@callware.com>
Message-ID: <38727518.9895C19F@rubic.com>

Ivan Van Laningham wrote:
> The biggest gotcha is that the Python scripts *must* start with
> 
>         #!c:\Python\python.exe
> 
> (or wherever Python actually lives). You can't use
> 
>         #!/usr/local/bin/python
> 
> because Apache won't find it, even if the path exists 
> on Apache's drive--it insists on the driveletter and 
> the backslashes.

Hi Ivan.

My notes document otherwise.  I could be wrong, since
I don't use the latest Win32/Apache, but I just copy
python.exe into C:\usr\local\bin and I can use
portable (for some loose definition of "portability")
python cgi scripts between Unix and NT.

http://starship.python.net/crew/jbauer/apachenotes/

Jeff Bauer
Rubicon Research



From gmcm at hypernet.com  Thu Jan 20 09:40:23 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 20 Jan 2000 09:40:23 -0500
Subject: ( 2.31.New operators: 'eq', 'ne', 'last', '..' ) ?
In-Reply-To: <slrn88e4pl.67b.grant@grante.comtrol.com>
Message-ID: <1263743270-9579629@hypernet.com>

Grant Edwards wrote:

> As long as we're at, I think we should try to make Python code
> not only more readable but also more polite. Rather than yet
> another imperitive language, the world needs more polite
> request languages!
> 
> I'm proposing the mandatory use of the keywords 'please',
> 'excuse_me', 'I'm_sorry' and 'thank_you'.  

I have only one quibble. "Excuse me" is, in fact, imperative. 
This is something that English hides, but becomes obvious in 
German.

Oh, and instead of "except please", it would be more polite to 
say "thank you very much but please". And don't forget to 
throw an exception unless sys.argv[1:5] == ['how', 'are', 'you', 
'today?'].

- Gordon



From markus.oberhumer at jk.uni-linz.ac.at  Wed Jan 26 08:41:20 2000
From: markus.oberhumer at jk.uni-linz.ac.at (Markus F.X.J. Oberhumer)
Date: Wed, 26 Jan 2000 14:41:20 +0100
Subject: [ANNOUNCE] PySol 3.30 - a solitaire game collection
Message-ID: <20000126144120.A25631@laetitia.oberhumer.com>

-----BEGIN PGP SIGNED MESSAGE-----


                    PySol - a Solitaire Game Collection
                              Version 3.30

                          http://pysol.tsx.org

           Copyright (C) 1998, 1999, 2000 Markus F.X.J. Oberhumer
                   <markus.oberhumer at jk.uni-linz.ac.at>


What is PySol ?
===============

PySol is an exciting collection of 160 solitaire card games.

Among the supported games are classics like Aces Up, Baker's Game,
Canfield, FreeCell, Forty Thieves, Golf, Klondike, Monte Carlo,
Osmosis, Pyramid, Scorpion, Spider, Yukon, and many more...

PySol may be freely distributed under the terms of the GNU GPL.

PySol aims to be Commercial Quality Freeware.


Why yet another solitaire game ?
================================

Here are some highlights of PySol:

  - freely available
  - distributed under the GNU GPL with full source code
  - currently supports 160 (!) distinct solitaire variants
  - based upon an extensible solitaire engine
  - very nice look and feel including multiple cardsets
    and background table tiles
  - sound samples and background music
  - unlimited undo & redo
  - load & save games
  - player statistics and log files
  - hint system and demo games
  - support for user written plug-ins - add your own solitaire variants
  - integrated HTML help browser
  - lots of documentation
  - fully portable across Unix/X11, Windows 95/98/2000/NT and MacOS
  - written in 100% pure Python
  - just run it - no need to compile anything


Nice. And what's new ?
======================

  * Implemented 7 new games.

  * Converted the pysol-sound-server into a Python extension module
    which should give much improved results.


Cool. Where can I get it ?
==========================

Point your browser to http://pysol.tsx.org
The PySol Gallery is awaiting your visit as well.


What do I need to start playing ?
=================================

PySol requires Python 1.5.2 and Tcl/Tk 8.0.5. Both packages are
freely available for Unix, Windows 95/98/NT and Macintosh.

BTW, there is no need to compile anything since the whole program is just
a Python script. Just run it, and that's all.

The Windows version ships as a completely self-contained setup file.


License terms
=============

PySol is Copyright (C) 1998, 1999, 2000 Markus Franz Xaver Johannes Oberhumer

PySol is distributed under the terms of the GNU General Public License (GPL).
See the file COPYING.


Have fun,
Markus

http://pysol.tsx.org



-----BEGIN PGP SIGNATURE-----
Version: 2.6.3ia
Charset: noconv

iQCVAwUBOI75E210fyLu8beJAQHiswQAk1tCJEJV9QXtcFUaTapqe8hQGDxXL3K9
nGVuk13ldH4Hdo/rhf+6y37ZysyW+G6J9PIBPp8FgSplCCNqzEpYaiGUf1tqlitg
DUhNjRQFCFu/B4TqE6vmcZkmCIPTWjXZBnTTK3JNo9Dfbm0aPzlJr3ns2LTXhfh/
7CVQyyp8wtk=
=S33/
-----END PGP SIGNATURE-----



From janssen at parc.xerox.com  Wed Jan 19 15:54:53 2000
From: janssen at parc.xerox.com (Bill Janssen)
Date: Wed, 19 Jan 2000 12:54:53 PST
Subject: Python Rocks! 
In-Reply-To: Your message of "Wed, 19 Jan 2000 11:58:47 PST."
             <002401bf62b8$c30da8a0$f29b12c2@secret.pythonware.com> 
Message-ID: <00Jan19.125324pst."3619"@watson.parc.xerox.com>

> why bother with Python when you can use corbascript instead?

Because Python is a widely used and mature languages with oodles of
libraries and experience programmers?

Bill



From chega at my-deja.com  Sun Jan  9 21:27:47 2000
From: chega at my-deja.com (Vadim Chugunov)
Date: Mon, 10 Jan 2000 02:27:47 GMT
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <011501bf58a4$7b0db540$0100a8c0@rochester.rr.com> <3877513B.73B762F4@maxtal.com.au> <004901bf5a08$11a35be0$0100a8c0@rochester.rr.com> <387890F6.2D1136F8@maxtal.com.au> <85b1kp$ts8$1@nnrp1.deja.com>
Message-ID: <85bg2l$7vd$1@nnrp1.deja.com>


> states= {
>     1: compile("x=x+1","<string>","single"),
>     2: compile("x=x+2","<string>","single")
> }

Dammit. This doesn't work for compiled code because of a bug in Python.
Plain strings work fine tho':
states= {
    1: "x=x+1",
    2: "x=x+2"
}


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


From rob at hooft.net  Mon Jan 31 09:58:47 2000
From: rob at hooft.net (Rob W. W. Hooft)
Date: Mon, 31 Jan 2000 15:58:47 +0100 (CET)
Subject: IOError: (0,"Error") when closing "os.popen"ed file?
In-Reply-To: <000701bf6bf1$672bb030$3acbd9c2@peridot.optichrome.com>
References: <m3n1pmmlse.fsf@devel.no.rwd.dinet>
	<000701bf6bf1$672bb030$3acbd9c2@peridot.optichrome.com>
Message-ID: <14485.41767.332703.614501@temoleh.chem.uu.nl>

>>>>> "AE" == Adrian Eyre <a.eyre at optichrome.com> writes:

 >> After hacking for a while, I managed to work around the problem by
 >> doing:
 >> 
 >> time.sleep(0.1) f.close()
 >> 
 >> But this is highly unsatisfactory..... Is there anyone that can
 >> give me a better suggestion?

 AE> I expect the OS hasn't finished with the file before it receives
 AE> the close().

 AE> Try replacing "time.sleep(0.1)" with "f.flush()". IIRC flush()
 AE> blocks untils the file operation has completed.

Actually I have seen the same error message from a f.flush() before,
so I already had coded:

   try: 
      f.flush()
   except IOError:
      pass

But maybe I should repeat the flush() until it succeeds, and only
then close() the file. Since I wrote my first message I have already
found occasions where even the sleep(0.1) doesn't solve the problem.
Bad. I really hope I can properly fix this.

Rob

-- 
=====   rob at hooft.net          http://www.xs4all.nl/~hooft/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========



From gmcm at hypernet.com  Sat Jan 15 21:22:32 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 15 Jan 2000 21:22:32 -0500
Subject: How to use Pythonwin's ability to access windows dll's???
In-Reply-To: <85r768$isb$1@nnrp1.deja.com>
Message-ID: <1264133143-1099206@hypernet.com>

Sam Schulenburg writes:
> In article <1264144125-438654 at hypernet.com>,
>   gmcm at hypernet.com wrote:
> > Sam Schulenburg writes:
> >
> > > I can access windows dll's using Sam Rushing's calldll.pyd and
> windll.py. My
> > > question is are thes features built into Pythonwin?
> >
> > Nope. Everything in the Win32 extensions is wrapped (much
> > of it via SWIG).  Sam's stuff is regarded as dangerous,
> > because it's too easy to crash Python.

> Seriously what is the purpose of the LoadLibrary function if not to
> access Dll's>

You can load & use resources from a dll (for dialogs, views 
etc). Not much use, otherwise.


- Gordon



From cjc26 at nospam.cornell.edu  Sun Jan 30 01:25:31 2000
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sun, 30 Jan 2000 06:25:31 GMT
Subject: lib reference in Unix man format
References: <200001281339.IAA12329@trm.bmo.com>
Message-ID: <slrn897mc0.31l.cjc26@bridget.mindriot.net>

Pada Fri, 28 Jan 2000 08:39:20 -0500 (EST), Tao Lu bilang:
| 
| If each built-in function has its own man page, I have to know the function
| name before I can find its man page.

Yeah, I realized this just after hitting `send' ;)


| This will be like the manual of Unix
| system calls. If I don't remember the name of message send operation, i.e.,
| msgsnd, I have to use 'man -s 2 intro' to figure out.
| 
| What I like 'perlfunc' most is that I can find similar function as well -- I
| may also learn some thing new every time when I browse 'perlfunc'.  ;-)

Good point..
I did a quick conversion of the web page describing the builtin functions
to man page format.  It's available at
<http://www.people.cornell.edu/pages/cjc26/pyfunc.1>.  Check it out and
tell me what you think..


-- 
cliff crawford  -><-  http://www.people.cornell.edu/pages/cjc26/
                     "All I need is a small town
                      with good Internet connectivity" - Nadine


From darrell at dorb.com  Mon Jan 24 20:22:09 2000
From: darrell at dorb.com (Darrell)
Date: Mon, 24 Jan 2000 20:22:09 -0500
Subject: Multidimensional Dictionaries: Possible?
References: <001101bf66cd$84128bc0$a42d153f@tim>
Message-ID: <007e01bf66d2$996e2100$0100a8c0@rochester.rr.com>

[Tim Peters]
> Note too that you can use dict[key1, key2, ...] to "fake" a
multidimensional
> dict.  Sometimes that's more appropriate.  The equivalent in Perl
> interpolates the multiple subscripts into a string; Python's dicts are
more
> general (can be indexed by any immutable object, not just strings), and
> Python constructs a *tuple* of keys under the covers.  That's faster and
> can't suffer accidental "superkey" collisions.

Cool!

>>> d={}
>>> d[1,2]=12
>>> d[1,2]
12
>>> d[1,2,3,4,5,6]='hugeDimension'
>>> d[1,2,3,4,5,6]
'hugeDimension'
>>> d.keys()
[(1, 2), (1, 2, 3, 4, 5, 6)]
>>>

Now I just need a reason to do this :-/

The rumor was that you're fiscal manifestation might be sited at the
conference.
Can you assign a probability to this ?

--Darrell




From pf at artcom-gmbh.de  Thu Jan 27 02:40:07 2000
From: pf at artcom-gmbh.de (Peter Funk)
Date: Thu, 27 Jan 2000 08:40:07 +0100 (MET)
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
Message-ID: <m12DjXD-000CxiC@artcom0.artcom-gmbh.de>

Look at the following coding alternatives often seen in generic python code:

class A:
    def load(self): 
        # ... do something e.g. load a file
        try:
            self.update()
        except AttributeError:
            pass

class B:
    def load(self):
        # ... do something e.g. load a file
        if hasattr(self, 'update'):
            self.update()

Which coding idiom is better?  The second form is one line shorter than
the first, but it has the disadvantage, that the method (or attribute) must
be named twice.  Both idioms are used in the standard libary.

Regards, Peter
-- 
Peter Funk, Oldenburger Str.86, 27777 Ganderkesee, Tel: 04222 9502 70, Fax: -60



From boud at rempt.xs4all.nl  Mon Jan 31 11:52:11 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 31 Jan 2000 16:52:11 GMT
Subject: Python MySQL.connect() question
References: <3893CD31.B17E8B84@regnoc.com> <870tjf$514$1@news1.xs4all.nl> <3894737C.A676BC16@regnoc.com> <381BAA74.863A8048@cs.rmit.edu.au>
Message-ID: <874ejr$35o$1@news1.xs4all.nl>

Hafeez Bana <hafhat at cs.rmit.edu.au> wrote:


> Try using "127.0.0.1" instead of localhost.

Well, that shouldn't make a difference - especially since localhost
does work on my box. I wonder what the results would be if Jim used
another MySQL binding, like MySQLdb?

-- 

Boudewijn Rempt  | http://www.valdyas.org


From ivanlan at callware.com  Tue Jan 18 21:31:23 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 18 Jan 2000 19:31:23 -0700
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <slrn888p7j.h3r.scarblac-spamtrap@flits104-37.flits.rug.nl> <8631qm$26p@news.or.intel.com> <86349g$37r$1@nntp8.atl.mindspring.net>
Message-ID: <388521FB.BCC45087@callware.com>

Hi All--

Aahz Maruch wrote:
> 
> In article <8631qm$26p at news.or.intel.com>, tye4 <tye4 at yahoo.com> wrote:
> >
> >How hard can it be to add 'end' indicators to the language.. If these end
> >indicators are added, the language would still be backward compatible. I
> >can't believe nobody has fixed this for so long.
> 
> Are you suggesting the addition of 'end' indicators IN ADDITION to the
> indentation?
>

Yes, he is.  tye4, wake up and smell the tabs.  Most people who use
Python like the indentation the way it is, and feel no need for the
unnecessary clutter your hack would produce.  Python is not Perl and we
don't want it to be.

Years ago, Susan Sontag discovered Science Fiction.  Being a big name
with a large reputation, she thought it incumbent upon her to call
attention to the small, overlooked ghetto of SF.  That would have been
fine, but when she wrote articles about SF, she failed to do even the
minimal research needed to find out what fans of SF called it.  So she
made up a name, and referred to it as "Sci-Fi" in her essays.

The SF community has hated her ever since; she had a big enough name and
reputation to make it stick, despite vehement objections from the
practitioners, and today we have such blights as the "Sci-Fi Channel"
because of her ignorance, arrogance and insensitivity.

<and-susan-sontag-you're-not>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From parkw at better.net  Fri Jan 28 09:09:50 2000
From: parkw at better.net (William Park)
Date: Fri, 28 Jan 2000 09:09:50 -0500
Subject: lib reference in Unix man format
In-Reply-To: <200001281339.IAA12329@trm.bmo.com>
References: <200001281339.IAA12329@trm.bmo.com>
Message-ID: <20000128090950.A2767@better.net>

The HTML docs works for me.  Try using 'lynx' (text browser) instead of
Netscape.  The problem with manpages is that it's difficult to write
(hence, one does it manually nowdays), and difficult to search/jump to
another docs.  HTML is essentially killed INFO, in my opinion.

William



From mikael at isy.liu.se  Mon Jan 24 07:34:58 2000
From: mikael at isy.liu.se (Mikael Olofsson)
Date: Mon, 24 Jan 2000 13:34:58 +0100 (MET)
Subject: Living without static member functions
In-Reply-To: <86hbmt$g8s@newton.cc.rl.ac.uk>
Message-ID: <XFMail.000124133458.mikael@isy.liu.se>

On 24-Jan-00 Richard Brodie wrote:
 >  class Milkman:
 >  
 >      allMilkmen = {}
 >  
 >      def __init__(self, name):
 >          ...
 >          allMilkmen[name] = self
 >  
 >      def find(self, name):
 >          return(allMilkmen[name])
 >  
 >  It's handy for debugging or garbage collecting like activities. Also
 >  for distributed applications, where it's harder to keep object references
 >  around.
 >  
 >  Firstly, I'm curious if this is well-known, useful or boneheaded design
 >  on my part.
 >  
 >  Secondly, it doesn't work in Python, of course, because I need a
 >  Milkman to find all the others (i.e. Milkman.find['Phil'] doesn't work).
 >  I can do something similar by taking allMilkmen and find to module
 >  scope; that seems kind of ugly, and worries me that I'm abusing the
 >  language. Can I do better?
 >  
 >  Thirdly, is there some neat way that I can find all members of a class
 >  by inquiring of the Python internals?

FWIW, if you move allMilkmen outside of the class, you at least get something
that works:

allMilkmen = {}
class Milkman:
    def __init__(self, name):
        allMilkmen[name] = self
    def find(self, name):
        return(allMilkmen[name])

But then again, that may not be what you want. If you do

a=Milkman('Mikael')
b=Milkman('David')
a=Milkman('Mattias')

then allMilkmen would be 

{'Mikael': <__main__.Milkman instance at e0f68>, 
'David': <__main__.Milkman instance at e0de8>, 
'Mattias': <__main__.Milkman instance at e0aa0>}

Hence, allMilkmen still keeps a reference to the first instance (Mikael), 
which has been replace by the third (Mattias). Plus, you have the risk
of accidentally doing something with allMilkmen without touching any
instance of Milkman, since it is not obvious that allMilkmen is actually
closely related to Milkman. Perhaps you could make it more obvious by
declaring it as global inside the class:

class Milkman:
    global allMilkmen
    allMilkmen = {}
    def __init__(self, name):
        allMilkmen[name] = self
    def find(self, name):
        return(allMilkmen[name])

Then again, I might just be spreading crap.

/Mikael

-----------------------------------------------------------------------
E-Mail:  Mikael Olofsson <mikael at isy.liu.se>
WWW:     http://www.dtr.isy.liu.se/dtr/staff/mikael
Phone:   +46 - (0)13 - 28 1343
Telefax: +46 - (0)13 - 28 1339
Date:    24-Jan-00
Time:    13:14:27

This message was sent by XF-Mail.
-----------------------------------------------------------------------


From hnowak at cuci.nl  Tue Jan 25 16:51:54 2000
From: hnowak at cuci.nl (Hans Nowak)
Date: Tue, 25 Jan 2000 22:51:54 +0100
Subject: Dictionary from a file
Message-ID: <200001252152.WAA18611@dionysus.fw.cuci.nl>

On 25 Jan 00, at 10:55, Adrian Eyre wrote:

> > # writeabledict.py
> > """A writeable dictionary."""
> > [snip]
> 
> Isn't this exactly what the 'shelve' module provides?

I've never really used shelve, but AFAIK, the file produced by shelve does
not contain 'readable' code... it looks like it's pickled? Anyway, the
WriteableDict class writes Python-readable code; in fact it exports an
module with a dictionary in it. It's far from perfect; it will not
successfully store class instances, to name something; but in general
cases it works well. (At least for me. :-)

I have one program that uses the WriteableDict principle a lot... an
auxiliary program parses text files and puts the results (Python lists) in
modules. Generating these modules only happens once (well, theoretically
:) ; they are imported by the main program, which is much faster than redo
the parsing every time the program is started, especially when .pyc files
are created.

I found that importing .pyc files is pretty fast. Didn't benchmark it
against shelve/pickle/marshall/etc, though.

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
You call me a masterless man. You are wrong. I am my own master.



From shang.remove_edu at st.jyu.fi.edu  Thu Jan 20 16:05:03 2000
From: shang.remove_edu at st.jyu.fi.edu (Sami Hangaslammi)
Date: Thu, 20 Jan 2000 23:05:03 +0200
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com>
Message-ID: <867t9d$aoi$1@mordred.cc.jyu.fi>

> I intend to modify the language's syntax to add
> 'end' delimiters for the blocks.

Wouldn't it be better (and easier) to write a script that converts
between .py and .pye (python with end delimeters) scripts. This would
allow you to both use existing Python libraries and distribute modules
you have made to others.

--
Sami Hangaslammi
shang (at) st (dot) jyu (dot) fi




From fredrik at pythonware.com  Wed Jan 12 10:06:10 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 12 Jan 2000 16:06:10 +0100
Subject: Perl speed vs. Python speed
References: <85hjfa$mlp$1@nnrp1.deja.com> <pd0f4.26480$W2.358425@iad-read.news.verio.net>
Message-ID: <00f801bf5d0e$ae601140$f29b12c2@secret.pythonware.com>

Tom Culliton" <culliton at clark.net> wrote:
> The "new" Python re module needs to be optimized and possibly pushed
> into C code, the older regex module is about twice as fast.

if everything goes according to plan, 're' (and possibly also
'regex') will be replaced by a new unicode-aware engine in
1.6.  I've attached some (somewhat outdated) benchmarks.

note that 'regex' is faster than 're' only if you apply simple
regular expressions many times.  if you can reorganize the
code to use a single regular expression on a larger string,
're' beats the hell out of 'regex'.  ...and if you make things
complicated enough, 'regex' stops working... (but as usual,
some people prefer to get the wrong answer quickly ;-)

</F>

          0     5    50   250  1000  5000 25000
----- ----- ----- ----- ----- ----- ----- -----
search for Python|Perl in Perl ->
sre8  0.007 0.008 0.010 0.010 0.020 0.073 0.349
sre16 0.007 0.007 0.008 0.010 0.020 0.075 0.353
re    0.097 0.097 0.101 0.103 0.118 0.175 0.480
regex 0.007 0.007 0.009 0.020 0.059 0.271 1.320

search for (Python|Perl) in Perl ->
sre8  0.007 0.007 0.007 0.010 0.020 0.074 0.344
sre16 0.007 0.007 0.008 0.010 0.020 0.074 0.347
re    0.110 0.104 0.111 0.115 0.125 0.184 0.559
regex 0.006 0.006 0.009 0.019 0.057 0.285 1.432

search for Python in Python ->
sre8  0.007 0.007 0.007 0.011 0.021 0.072 0.387
sre16 0.007 0.007 0.008 0.010 0.022 0.082 0.365
re    0.107 0.097 0.105 0.102 0.118 0.175 0.511
regex 0.009 0.008 0.010 0.018 0.036 0.139 0.708

search for .*Python in Python ->
sre8  0.008 0.007 0.008 0.011 0.021 0.079 0.379
sre16 0.008 0.008 0.008 0.011 0.022 0.075 0.402
re    0.102 0.108 0.119 0.183 0.400 1.545 7.284
regex 0.013 0.019 0.072 0.318 1.231 8.035 45.366

search for .*Python.* in Python ->
sre8  0.008 0.008 0.008 0.011 0.021 0.080 0.383
sre16 0.008 0.008 0.008 0.011 0.021 0.079 0.395
re    0.103 0.108 0.119 0.184 0.418 1.685 8.378
regex 0.013 0.020 0.073 0.326 1.264 9.961 46.511

search for .*(Python) in Python ->
sre8  0.007 0.008 0.008 0.011 0.021 0.077 0.378
sre16 0.007 0.008 0.008 0.011 0.021 0.077 0.444
re    0.108 0.107 0.134 0.240 0.637 2.765 13.395
regex 0.026 0.112 3.820 87.322 ...

search for .*P.*y.*t.*h.*o.*n.* in Python ->
sre8  0.010 0.010 0.014 0.031 0.093 0.419 2.212
sre16 0.010 0.011 0.014 0.030 0.093 0.419 2.292
re    0.112 0.121 0.195 0.521 1.747 8.298 40.877
regex 0.026 0.048 0.248 1.148 4.550 24.720 ...

(searching for the given patterns in strings padded
with blanks on both sides.  sre8 is the new python
1.6 engine on 8-bit strings, sre16 is the same engine
using unicode)

reg?u?lar
1: belonging to a religious order




From pinard at iro.umontreal.ca  Wed Jan 19 14:08:13 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 19 Jan 2000 14:08:13 -0500
Subject: LISP Rocks! (was: Re: Python Rocks! :-)
In-Reply-To: Timothy Grant's message of "Tue, 18 Jan 2000 21:44:26 -0800"
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <oqpuuystm6.fsf@titan.progiciels-bpi.ca> <38854F3A.4FBA74F9@exceptionalminds.com>
Message-ID: <oqg0vtyj4y.fsf_-_@titan.progiciels-bpi.ca>

Timothy Grant <tjg at avalongroup.net> ?crit:

> > My humble feeling is that you need your editor fixed, and that you report
> > the problem you have as being one of Python language design.

> A good editor makes such a huge difference that I can barely work in
> one that doesn't have a "python" mode and syntax highlighting.

I learned LISP on batch systems, using punched cards.  You cannot easily
shift parentheses right or left on punched cards (yes, I know the trick
of duplicating cards while pressing with the thumb on one of the card,
so it sticks still despite the rollers).  We did not have much context to
check vertical alignment of parentheses.  We were counting and balancing
parentheses all the time, really.  Program listings were double-line spaced,
with every other line holding counters below each parenthesis.  Many people
said that LISP stands for `Lots of Insipid and Stupid Parentheses'. :-)

Now that editors automatically take care of proper indentation, and quickly
show me how parentheses match, parenthesis counting wholly disappeared.
It is real, total comfort.  LISP is easier now than ever.  Yet, it never
occurred to us that it could have been ill-designed, because uneasy to punch.

P.S. - SNOBOL was particularly uneasy to punch.  Multi-punches all the time.
Do some remember? ;-)

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





From samschul at pacbell.net  Thu Jan 27 20:59:58 2000
From: samschul at pacbell.net (Sam Schulenburg)
Date: Fri, 28 Jan 2000 01:59:58 GMT
Subject: Python startup seems slow... (NT)
References: <000201bf67bf$b27527a0$592d153f@tim> <MPG.12f911f129ce00ff989681@or.news.verio.net> <nDKj4.747$UH3.67876@news4.usenetserver.com> <38900E46.909421F8@quantisci.co.uk> <2eYj4.1931$UH3.230869@news4.usenetserver.com> <86q5vv$mci$1@nnrp1.deja.com> <Qq2k4.61$RU4.2585@news4.usenetserver.com>
Message-ID: <86qt6t$7ta$1@nnrp1.deja.com>

In article <Qq2k4.61$RU4.2585 at news4.usenetserver.com>,
  "Joel Lucsy" <jjlucsy at concentric.net> wrote:
> It's new in MSVC6.0. If you use MSVC5, you won't be able to find it.
>
> - Joel Lucsy (jjlucsy at concentric.net)
>
>

I see where this technique will speed up the initial loading
of Python,but you will loose time on all subsequent library calls. Is it
not more efficient to have all the library functions available
immediately.


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


From tiddlerdeja at my-deja.com  Wed Jan 19 14:13:51 2000
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Wed, 19 Jan 2000 19:13:51 GMT
Subject: Python -> EXE?
References: <38831BE5.76B189A8@mchp.siemens.de> <002801bf60fa$80b4b440$f29b12c2@secret.pythonware.com>
Message-ID: <8652d3$ih2$1@nnrp1.deja.com>

I tried to use the Standalone.py from this distribution, but I got an
error.

 found OLEAUT32.dll
 found python15.dll
 added filter <TypeFilter {'d': 1}>
Creating createFruitCategories.exe
  support\Run.exe -> createFruitCategories.exe
  C:\TEMP\~535-1 -> createFruitCategories.exe
Gathering components of dist_createFruitCategories
 added filter <TypeFilter {'d': 1}>
Creating dist_createFruitCategories
Traceback (innermost last):
  File "Builder.py", line 502, in ?
    main(opts, args)
  File "Builder.py", line 484, in main
    target.build()
  File "Builder.py", line 87, in build
    self.assemble()
  File "Builder.py", line 234, in assemble
    shutil.copy2(path, self.name)
  File "d:\Program Files\Python\Lib\shutil.py", line 63, in copy2
    copyfile(src, dst)
  File "d:\Program Files\Python\Lib\shutil.py", line 17, in copyfile
    fsrc = open(src, 'rb')
IOError: [Errno 2] No such file or directory: 'c:\\winnt\\system32
\\pywintypes.dll'



I have pywintypes15.dll. Could someone tell me what I'm doing wrong?

Thanks.


In article <002801bf60fa$80b4b440$f29b12c2 at secret.pythonware.com>,
  "Fredrik Lundh" <fredrik at pythonware.com> wrote:
> Christian Schaller <Christian.Schaller at mchp.siemens.de> wrote:
> >   Is there a tool similar to Perl2Exe that transforms a Python
program
> > to a MS .EXE programc that can be executed without installing
Python =
> on
> > a machine?
>
>     http://www.python.org/doc/FAQ.html#4.28
>     "How can I create a stand-alone binary from a Python script?"
>
> </F>
>
> <!-- (the eff-bot guide to) the standard python library:
> http://www.pythonware.com/people/fredrik/librarybook.htm
> &mdash; see it live on the python conference! -->
>
>


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


From mrecks at my-deja.com  Sat Jan 29 17:07:12 2000
From: mrecks at my-deja.com (mrecks at my-deja.com)
Date: Sat, 29 Jan 2000 22:07:12 GMT
Subject: strange CGI problem
References: <85ga10$p8u$1@nnrp1.deja.com> <14459.47851.670815.455938@beluga.mojam.com> <85hb86$jjq$1@web1.cup.hp.com>
Message-ID: <86voad$k42$1@nnrp1.deja.com>

We found that this was caused by our giving our cgi programs name with
unrecognised or non-standard extension, in our case '.cgi' We found two
ways to overcome this problem.

1) Give your cgi program a recognised extension i.e. instead
of 'program_name.cgi' use 'program_name.html' Don't forget to also
change the cgi program name in the page that calls it.

2) Alternatively use the Netscape 'applications' settings to handle
files with '.cgi' extensions by Netscape internally.

Hope this helps

X



In article <85hb86$jjq$1 at web1.cup.hp.com>,
  "Arad Rostampour" <arad at fc.hp.com> wrote:
> Do you have a:
>
> print "Content-type: text/html\n"
>
> as your first output?
>
> Enjoy,
> Arad
> --
> Arad Rostampour- arad at fc.hp.com
> NT Platform and I/O Team
> Software and System Development Lab
> Hewlett-Packard
>
> Skip Montanaro wrote in message
> <14459.47851.670815.455938 at beluga.mojam.com>...
> >
> >    >> I'm creating a little web-site; for some reason when I view
the page
> >    >> in Internet Explorer, it's fine, but Netscape shows me the
raw html
> >    >> as if it's a text file.  What would cause this?
> >
> >Sounds like IE is looking at the contents and saying, "hmmm... looks
like
> >HTML", and displaying it.  Make sure your CGI script is spitting out
an
> >appropriate Content-type header.
> >
> >
>
>


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


From dweber1 at austin.rr.com  Mon Jan 24 22:19:59 2000
From: dweber1 at austin.rr.com (Daniel Weber)
Date: Tue, 25 Jan 2000 03:19:59 GMT
Subject: Tab notebook control available for Tkinter?
Message-ID: <388D24E8.EB1318B8@austin.rr.com>

I'm trying to write a new front-end to a product (written in OS/2) to
make it cross-platform.  However, I need a notebook tab control, which
I haven't been able to find in the archives or any reference in
deja.com.

Other than moving to wxWindows (which isn't available on OS/2 that I can
see), are there any Tkinter based solutions?

TIA -- daniel



From evan at tokenexchange.com  Fri Jan  7 11:25:19 2000
From: evan at tokenexchange.com (Evan Simpson)
Date: Fri, 7 Jan 2000 10:25:19 -0600
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net>
Message-ID: <s7c54hj9oj8110@corp.supernews.com>

Emile van Sebille <emile at fenx.com> wrote in message
news:016a01bf57d3$cecb2900$01ffffc0 at worldnet.att.net...
> Doesn't this do that now? or is there some objection to this usage?

Three objections, none of them decisive:

1. Robustness: A typo in either the 'raise' or 'except' string will cause
your program to either die or jump to an unexpected location, possibly long
after you made the typo if the case is rare.  A typo in your argument or
parameter list will cause a runtime error; make sure you don't forget to
make them both lists or both tuples.  A typo in 'break' or 'continue' is
caught at compile time.

2. Clarity: When reading this code, how can I tell what is exceptional and
what is normal flow control?  I can't be sure that any given 'raise' is
meant to be handled locally without checking for a matching 'except'.

3. Efficiency/elegance:  Why invoke the exception machinery when a simple
JUMP will do? Why load up your runtime with these strings (which ought to be
nice and long and descriptive) when they're only needed as matching labels?

Cheers,

Evan @ 4-am




From tye4 at yahoo.com  Thu Jan 20 17:44:15 2000
From: tye4 at yahoo.com (tye4)
Date: Thu, 20 Jan 2000 14:44:15 -0800
Subject: Python Rocks! - get rid of colons
References: <000701bf6302$5eb88a60$31a2143f@tim> <388765BA.3FC8@marshall.edu> <867qt1$a49@news.or.intel.com> <388774C7.1784B662@callware.com>
Message-ID: <86834t$fs0@news.or.intel.com>

Ivan Van Laningham <ivanlan at callware.com> wrote in message
news:388774C7.1784B662 at callware.com...
> Hello all--
>
> If you had bothered to use Python, you would already know that it is not
> a compiler but an interpreter.
>
If you have bothered to use Python, you should know that it is neither a
compiler nor an interpreter - but something in between: a byte-code
compiler. That's why it's fast and not sluggish like many BASIC
interpreters.

-tye4






From eddyw at lsl.co.uk  Thu Jan 20 09:39:31 2000
From: eddyw at lsl.co.uk (Edward Welbourne)
Date: Thu, 20 Jan 2000 14:39:31 +0000
Subject: In defence of the two-namespace rule
Message-ID: <E12BIkF-0003Zz-00@lsls4p>

Markus Stenberg and Bernhard Herzog consider the two-namespace rule to
be a deficiency in python, and think it would be `simpler' to use
`lexical scoping'.  The Tim has given an excellent and erudite summary
of what that could be asked to mean, since Evan asked and its advocates
didn't answer.  Personally, I consider the two-namespace rule to be a
major *strength* of python: it is a major ingredient in pythonicness.

Define a function, foo, which returns a function it has defined, as bar,
during its suite; have bar refer to some of the local variables of foo.
With lexical scoping, the interpreter is obliged to keep the namespace
foo used (on that invocation) for as long as anyone retains a reference
to what it returned.

I contend that the namespace in which a function executes its suite
*should* be discardable at the time the function returns.  Lexical
scoping would mean that we can't discard foo's namespace when foo
finishes execution.  The implications of this for garbage collection are
singularly unpleasant - give them some thought.

The only way out is for the values bar accesses from foo to be
transcribed, before foo finishes executing, into some namespace whose
lifespan is at least as long as that of bar (and, ideally, coeval with
it).  Since bar carries around a namespace, for its defaults, with
exactly the right life-span, that's the right place to which to
transcribe such values (promoting them to globals being, I trust, to
no-one's liking).  Now, some folk might prefer to have that done
implicitly: personally, I'm all in favour of each suite of code needing
to be explicit about its dependencies on others.  Like Evan,

> I rather like the idea of having to explicitly bring names into an
> inner scope, and with semantics identical to parameter defaults.

So, yes, this is a matter of taste: the two-namespace rule is an emetic
to some, to whom I would commend Algol as a language to use instead of
python; but to pythoneers, the two-namespace rule is harmonious,
beautiful and kind to the garbage-collector.  Treat the humblest of thy
servants kindly, lest they rise up and smite thee with just zeal.

> when-guido-sez-"too-radical"-he-means-"not-radical-
>     enough"-ly y'rs  - tim
Ach, so back-pedalling would be a wrong strategy ?
I'll bear that in mind - thanks for the advice ;^>

	Eddy.



From thantos at chancel.org  Thu Jan 27 04:27:03 2000
From: thantos at chancel.org (Alexander Williams)
Date: Thu, 27 Jan 2000 09:27:03 GMT
Subject: Which idiom is better? try: ... except Attr..Err or hasattr?
References: <m12DjXD-000CxiC@artcom0.artcom-gmbh.de>
Message-ID: <slrn8903r7.al2.thantos@brimstone.mecha>

On Thu, 27 Jan 2000 08:40:07 +0100 (MET), Peter Funk
<pf at artcom-gmbh.de> wrote:

>Which coding idiom is better?  The second form is one line shorter than

Personally, I prefer the former idiom with try/except; some object
because its slightly slower than the other, but I find it clearer.

Perhaps most important from an efficiency PoV is that the try/except
structure really implies you expect the except clause to /really/ be
the uncommon case, a true exception.  If it happens roughly 50% of the
time, the if hasattr() construct may be more what you want in terms of
intent.  If its the general case, reverse the sense of your try.  :)

-- 
Alexander Williams (thantos at gw.total-web.net)           | In the End,
  "Join the secret struggle for the soul of the world." | Oblivion
  Nobilis, a new Kind of RPG                            | Always
  http://www.chancel.org                                | Wins


From thomas at bibsyst.no  Sun Jan 16 23:52:20 2000
From: thomas at bibsyst.no (Thomas Weholt)
Date: Mon, 17 Jan 2000 05:52:20 +0100
Subject: Removing tags from a html-file
Message-ID: <3882A004.A5F150F@bibsyst.no>

Hi,

I want to remove all tags from a html-file, so that only the plain-text
remains. How, if so, can this be done using htmllib.py?

If not, is there any other way to do this ??

Thomas



From richer at eastmail.com  Wed Jan  5 15:33:00 2000
From: richer at eastmail.com (richer at eastmail.com)
Date: Wed, 5 Jan 2000 15:33:00
Subject: (no subject)
Message-ID: <744.104101.470336@mail.mia.spjiioj.com>

 Hello:

 My name is Robert Todd

 The Golden Era Of Internet Marketing Has Arrived!

 Will You Be Part Of It?

 How Can You?

 Having the potential to reach millions with relatively little
 investment, the Internet provides opportunities for mass
 marketing that have never been seen before!  You can now
 compete on the same level as the Big Boys.

 Your presence on the World Wide Web can be as great as
 the biggest of the big corporations --

 IF YOU HAVE THE TOOLS  - and -
 IF YOU KNOW HOW TO USE THEM!

  You now have before you the Opportunity to get in on the
Ground Floor of a BRAND NEW Network Marketing Opportunity? Have
  you ever wished you could have started out in the beginning 
with
 Nu-Skin, Amway or any of the other networking Giants?

  Our plan is to bring massive signups into our program.

 The support and tools we provide are unequaled . These tools
 have been used by networkers to build enormous downlines in 
short periods of time. (One Networker built a downline of 80,000 
in 12 months while working a full time job, using our tools).

 And I promise you will not find a more qualified nor will you
find a more supportive upline anywhere in the world.

 What we are offering is a business with the following

   * FREE State of the art webpage
   * FREE On hands Training.
   * FREE Tremendous Upline Support!
   * FREE Step-By-Step Business Building Plan!
   * FREE Opportunity Business Conferences Daily!
   * FREE take advantage of our unlimited lead generating system.

   Also Everyone Who Joins Us can have
 * Merchant account - 99% approval
* Web Hosting
 * Flat Rate Long Distance Telephone Service for only $39.99 per 
month.
 * You can add unlimited calling card for additional $59.99 
Monthly.
 This service is 24 hours X 7 days a week only in the lower 48 
states



 When you decide to take advantage of this
   Opportunity, you will:
  * HAVE the Potential to Earn an unlimited Income!
  * HAVE a Golden Opportunity to Secure Your Financial Future!
  * AND finally earn the Money most Networkers only Dream about
  * Be Mentored by networking Leaders from Nu-Skin, Amway, Excel, 
Etc.
  * Unlimited access to all our leaders.
  * $5,500 Bonus commission on 6 sales! No limit on $5,500 
bonuses.

    Be a  part of the fastest growing market in the world!

  About the Internet:

 The Internet has grown from 2 million users in 1993
 to over 70 million users today
 Experts project that there will be 500 million users by the year 
2000
 The number of new users on the "Net" doubles every 8 months
 Every cable, long distance company and "Baby Bell" will insure 
that all
 Americans have easy access to the Internet including TV set 
access
through "Web TV" providers

 We Provide a complete "Marketplace on the Internet" for home and
business  consumers

-This is a precious commodity. Busy consumers are looking for 
faster,
more  hassle free ways to do their shopping-

 According to The Kiplinger Letter, May 9, 1997, the number of 
firms
 currently using the Internet to buy or sell is growing 10% per 
month...
 doubling every seven months!

 Make sure you DO NOT miss this new and exciting opportunity that 
is
taking the Internet  by storm!

  WE ALREADY HAVE MLM LEADERS JOINING OUR COMPANY DAILY. SECURE
  YOUR POSITION NOW FOR A MOST REASONABLE JOINING FEE!

If interested please fill in info below and I'll get back  to you 
ASAP.

Send to kahuna-info2 at online-venture.com

 Name:

 Email address:

 Telephone:

 Best time to call:

  Sincerely yours,

 Robert Todd

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Robert Todd  ICQ#1769450
 PO Box 3077 Oneco, FL 34205 USA

http://www.angelfire.com/biz5/cyberkahuna/index.htm

 Toll Free VoiceMail 877-817-6020
 FAX  310-495-0415



 
 
 
 
 



From alin at tumbleweed.com  Wed Jan  5 01:09:15 2000
From: alin at tumbleweed.com (alin at tumbleweed.com)
Date: Wed, 05 Jan 2000 06:09:15 GMT
Subject: zlib and zip archives
References: <QBc74.12$O3.154@uchinews>
Message-ID: <84un63$8ku$1@nnrp1.deja.com>

Hi,

It sounds stupid but I really don't have a clue:

How do you use zlib to create a zip archive using the pkzip format?  I
looked through the documentation, it seems no where indicates that it
supports the zip archive format?

Your help is greatly appreciated!

Thanks,
Anne

In article <QBc74.12$O3.154 at uchinews>,
  "Nick Collier" <nick at src.uchicago.edu> wrote:
> Hi,
>
> I've been using zlib to create a zip archive using the pkzip format,
and
> I've only been half succesful. The file headers and the central
directory
> records and end of central directories records seem to get written
> correctly. Winzip and infozip's unzip both report the correct files
and
> details when listing the files in the archive. However, when I try to
> decompress the data and I get an error with the message "invalid
compressed
> data to inflate". I'm using zlib.compress with a level of 8 to
compress the
> data.
>
> Any suggestions as to what I'm doing wrong?
>
> thanks,
>
> Nick
>
>


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


From ivanlan at callware.com  Wed Jan  5 21:41:31 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 05 Jan 2000 19:41:31 -0700
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com>
Message-ID: <387400DB.3C683629@callware.com>

Hi All--

tmick at mail.com wrote:
> 

[snip]

> I have not read _Python Essential Reference_. All I can say about it is
> that I have heard that Guido's contribution to it is minimal and that
> his name is prbably there more for name recognition. That says nothing
> about the quality of the book though.
> 

Guido is not listed as the co-author.  It says clearly "Foreword by
Guido van Rossum, Creator of Python" right there on the cover.

I think it's a good book, but not one you'd want to learn Python from. 
Right now, your best bet is _Learning Python_.

<learn-python-and-rule>-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From benwittner at mediaone.net  Sun Jan 16 00:05:39 2000
From: benwittner at mediaone.net (Ben Wittner)
Date: Sun, 16 Jan 2000 05:05:39 GMT
Subject: conflict between my package and os standard module
Message-ID: <Dkcg4.19933$zU5.216301@wbnws01.ne.mediaone.net>

I'm looking for help.

I have a bunch of extensions to standard modules that are pretty
specific to the needs of my project.

I chose to create a package for my project, let's call it my_proj.

In there, I chose to create modules corresponding to standard
modules (e.g., my_proj.os, my_proj.re) to hold the project-specific
extensions of the standard modules.

I put the directory containing my_proj on my PYTHONPATH,
but I did NOT put my_proj on my PYTHONPATH.

In my_proj/os.py I put the following line:

import os

Since my_proj is NOT on my PYTHONPATH, I thought that the line above
would cause the standard module os to be imported.
Instead, it seems that my_proj/os.py is imported.
(I seems that because
    (a) I get an attribute error for "path"
         when I use os.path.isdir() in my_proj/os.py and
    (b) when I print dir(os) I see my_proj.os stuff instead of standard
          module os stuff).

I tried taking '' off sys.path, before importing os, but that did not help.

Of course, I could rename my extension modules to things like
os_x.py and re_x.py, but it seems like that should be unecessary
since I'm already indicating that it's part of my project related package
by prepending with "my_proj.".

Any ideas before I rename all my extension packages and change all
references to them?

Thanks.

-Ben

benwittner at yahoo.com





From dheise at debitel.net  Sun Jan 23 22:33:44 2000
From: dheise at debitel.net (Dirk Heise)
Date: 23 Jan 2000 22:33:44 MEST
Subject: Python seems to be growing fast.
Message-ID: <01bf65e9$6e2d2160$c9dc343e@dreadzone>

I mean, the Windows download exe does -
from 2.3 MByte in 1.5.1 to 5MByte in 1.5.2.
Having a slow connection here at home, 
this annoys me a little. Can a 0.0.1
version difference really explain such a 
large difference?

Dirk


From calishar at *NOSPAM*home.com  Wed Jan 26 20:56:40 2000
From: calishar at *NOSPAM*home.com (Calishar)
Date: Thu, 27 Jan 2000 01:56:40 GMT
Subject: Creating Objects
Message-ID: <7g8v8s01mjtouok31d7elm09bvqc184tu7@4ax.com>

Hi folks,

  I'm working on solving a problem that my boss put to me the other
day, and I have run into a solution that I don't know how to implement
in Python (or any other language for that matter)

Basic description of problem (particulars not included)

Work out a system that can import information from (any number) of
systems, and write the infomration out (meaningfully) to any system in
the set.

I.e.
Assume a set a set of systems [A-G] the system needs to be able to
import from any of the seven systems, and write out it back out to any
of the seven. Ther may or may not be multiple instances of each system
type.


What I have in mind is defining a set of base classes and defining the
actual implementation at run-time

i.e.

# module my_classes.py
# imported as first part of main app

class system_a:
    def __init__(self):
	.....

class system_b:
    def __init__(self):
	.....
#(and so on)

and then during the main program what I would like to do is
...
    def do_translate(insystem, outsystem):
	class translateclass(system_dictionary[insystem],

system_dictionary[outsystem]):
    translateobject=translateclass(insystem, outsystem)
    translateobject.setsystems(input=insystem, output=outsystem)
    translateobject.translate()
....


Unfortunately I don't think run-time class definitions like this are
supported (understandably), but the more I think about it, the more I
am convinced that this is how to do what my boss wants.

  I would much rather do 8 class definitions (in this example) than
the 64 I would need to do otherwise.

  Am I pretty much looking at something that is un-doable, or am I
just looking at it wrong? (I think the latter because there seems to
be an idea floating just outside my brain range)

Calishar

P.s.  This would make my third project done fully within in Python (at
least for the RAD stage) Not bad considering my fist one started Jan
4th (yes, it was a Y2K patchup) 


From skip at mojam.com  Wed Jan  5 14:40:05 2000
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 5 Jan 2000 13:40:05 -0600 (CST)
Subject: TypeError: slice index must be int ?
In-Reply-To: <20000105111525.C1172@lexgen.com>
References: <20000105111525.C1172@lexgen.com>
Message-ID: <14451.40469.112192.31909@beluga.mojam.com>

    Hao> Is it possible to use long int (say 2000000000L) to slice a string
    Hao> with gigabytes size?

Apparently not:

    >>> import string
    >>> string.letters[5L:10L]
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
    TypeError: 'slice index must be int'

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 a.eyre at optichrome.com  Tue Jan 25 06:05:18 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 25 Jan 2000 11:05:18 -0000
Subject: Integer literal overflow error with -2147483648 but not with 0x80000000 ?
In-Reply-To: <CC59868A0A76D311B1F50090277B321C1F30A8@VRTMAIL>
Message-ID: <000201bf6724$10ab5250$3acbd9c2@peridot.optichrome.com>

> >>> 0x80000000
> -2147483648
> >>> -2147483648
> Traceback (innermost last):
> OverflowError: integer literal too large

I'm guessing this is because Python parses it as:

-(2147483648)

and can't evaluate 2147483648 before making it negative.

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From gmcm at hypernet.com  Wed Jan 19 23:30:21 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 19 Jan 2000 23:30:21 -0500
Subject: Mapping Column Names to results using DBI
In-Reply-To: <388680BE.EB94E957@exceptionalminds.com>
Message-ID: <1263779872-7378085@hypernet.com>

Timothy Grant wrote:

> Aahz Maruch wrote:
>  
> > Try "SELECT x, y, z" instead of "SELECT *".
> 
> I knew it couldn't be that simple...
> 
> I must be missing something

DBs return the columns in the order requested. So you will 
always get tuples of (x,y,z).

"Select *" is like "from module import *" - something you 
should really only do in interactive mode.

I'll agree that this creates a distasteful coupling between the 
SQL code and the Python, but performance makes it more 
palatable.



- Gordon



From P.J.W.S.VRIJLANDT at INT.azg.nl  Thu Jan 20 06:52:41 2000
From: P.J.W.S.VRIJLANDT at INT.azg.nl (P.J.W.S. Vrijlandt)
Date: Thu, 20 Jan 2000 12:52:41 +0100
Subject: extended slice
Message-ID: <139104006C4@nds-3.azg.nl>

Hi all,

In the reference manual under slice I find hints of an extende slice 
syntax. It goes like a[2:3:step], a[2:4, 6:8] or a[... , 2:4].

Indeed 'ellipsis' is a builtin and seems to denote ...

However, I can find no way of using extended slicing. Do we have this 
feature?

Patrick



From embed at geocities.com  Thu Jan 27 11:41:43 2000
From: embed at geocities.com (Warren Postma)
Date: Thu, 27 Jan 2000 11:41:43 -0500
Subject: tabs do WHAT?
References: <000301bf6706$99fca7c0$632d153f@tim> <86joac$l2s$1@mach.vub.ac.be> <86k30t$i5r$1@nntp2.atl.mindspring.net> <landauer-4BDC4F.15462726012000@news.apple.com> <86pcii$600$1@mach.vub.ac.be>
Message-ID: <eD_j4.7221$S64.211493@nnrp1.uunet.ca>

"Thomas Hamelryck" <thamelry at vub.ac.be> wrote in message
news:86pcii$600$1 at mach.vub.ac.be...
> So python is missing a lot
> of potential momentum beacuse of the indentation stuff.

Exactly. This is why C++ flew and why other [insert favourite language that
never achieved widespread use] never did live up to the initial ballyhoo.
Programmers have a low tolerance for linguistic weirdness.  Of course that
theory totally fails to explain Perl, which is the biggest exception to the
rule I'm making here.   Heck, C++ is weird too. I'm just used to the
weirdness, and it doesn't bother me anymore.

Perhaps "to each his own" is all that can be said.  I seem to have kicked
off a very nice round of debate on this whole topic, though, so I'm quite
pleased with myself! ;-)

Warren





From magnus at cs.brandeis.edu  Thu Jan 27 13:09:44 2000
From: magnus at cs.brandeis.edu (Magnus Bjornsson)
Date: Thu, 27 Jan 2000 13:09:44 -0500
Subject: Python script monitoring another process
Message-ID: <389089E8.32BB7710@cs.brandeis.edu>

Hi all,

I'm new to Python so this is probably a typical newbie question...
Anyway, I'm trying to build a Python script which does the following:
 1. starts a process (that's easy)
 2. checks the output from the process (from its stdout)
 3. when a certain token is found in the process' output the script 
    sends a new token to the process' stdin

So, how can I redirect the process' stdout so I can manipulate it in the
Python script?  And how can I redirect my scripts stdout to the process'
stdin?
Using sys.stdin doesn't seem to do the trick since it doesn't affect the
environment for the running process...

Any help would be appreciated,
- magnus e.


From alex at somewhere.round.here  Wed Jan 12 19:46:12 2000
From: alex at somewhere.round.here (Alex)
Date: 12 Jan 2000 19:46:12 -0500
Subject: a small problem
References: <20000112015923.A8450@xs4all.nl> <000701bf5cb4$c72320a0$da2d153f@tim> <20000112145107.B8450@xs4all.nl> <20000112165932.A2758@stopcontact.palga.uucp>
Message-ID: <etdln5ug5nv.fsf@m2-032-4.mit.edu>

> But suppose I have:
> t = [('foo', 'bar'), ('foo_two', 'bar_two'), ('foo_three', 'bar_three')]
> first_ones = map(lambda s: s[0], t)
> last_ones = map(lambda s: s[1], t)
> 
> How would I do that without lambda's, if this code is in a method?
> Should I nest two functions?

from operator import getitem
first_ones=map(getitem,t,len(t)*[0])

:)

Alex.


From alex at magenta.com  Wed Jan  5 18:16:52 2000
From: alex at magenta.com (Alex Martelli)
Date: Thu, 6 Jan 2000 00:16:52 +0100
Subject: lazy (?) evaluation, request for advice
References: <20000105170007.706CE1CDCE@dinsdale.python.org>
Message-ID: <014701bf57d2$f9f69a20$c02b2bc1@martelli>

I'm planning to prototype in Python a bunch of bridge-related
functionality, including selective dealing.  In selective dealing,
the program must repeatedly generate random deals, and
apply on each generated deal a user-supplied function, which
can accept or reject the deal depending on whether it matches
the characteristics the user is trying to select for -- this is very
useful for practice, analysis, and statistical studies, and
several custom programs (as well as a set of extensions to
Tcl) exist for the purpose... but I'd much rather let the user
write the deal-selection function in Python, than in either Tcl,
or any custom, ad-hoc "little language".

The number of features of a bridge deal that could be
selected for is huge, and some of them are pretty costly
to compute (for the huge number of deals that are often
needed when one is selecting for rare combinations
and/or wants a lot of selected hands, in order to collect
statistics about them).  So, I would like to compute only
those features that the user actually needs on a given run.

Further, since the typical user will not be a programmer,
I would like to make the user's life as easy as possible.
I.e., despite the important mantra about explicit being
better than implicit, I think the user would much rather
be able to write, say:
    (spades>=6 or hearts>=6) and (3.0<losers<4.0)
rather than have to mention "current_hand.spades" or
other such explicit notation.


If I didn't need lazy evaluation, I could easily provide
the latter nice-for-the-user syntax, e.g.:

    def eval_hand(a_hand):
        d={}
        d['spades']= # compute spade length of a_hand
        # ditto for 'hearts','diamonds','clubs'
        d['losers']= # compute a_hand's losers
        # ditto for umpteen other criteria
        return d

    for a_hand in some_huge_sequence:
        if(eval(user_expression,eval_hand(a_hand))):
            # add a_hand to those being selected


Or, if I were willing to make the user write "x.spades",
etc, instead of just "spades", I could do lazy evaluation
by using as 'x' an object with a suitable __getattr__.


I've thought about doing the eval repeatedly, within
a try/except block to catch the NameError's, inserting
the missing names in the dictionary one by one, but
I'm suspicious about the performance characteristics
of such an approach.  Maybe it would be all right if,
for hands after the first one, I pre-inserted values for
the names that are already in the dictionary from
previous iterations -- that would swiftly reduce the
number of NameError's to nil on most iterations,
although it would mean I would often have to compute
more features than necessary (e.g., for the above
example expression, I would be doing the costly
computation of losers for every hand, a pity since
it's only actually needed for the rare hands that have
at least 6 spades or hearts).


Of course, if I could use a UserDict rather than a
real dictionary in the eval(), I'd be in clover, but I
can't -- that gives a TypeError.


So... once more, this newbie is looking for advice
and counsel from you kindly experts!  What do you
think is the best approach for this situation...?
Thanks in advance, of course!


Alex





From breiter at usf.Uni-Osnabrueck.DE  Tue Jan 11 08:00:56 2000
From: breiter at usf.Uni-Osnabrueck.DE (Bernhard Reiter)
Date: 11 Jan 2000 13:00:56 GMT
Subject: test
References: <ltiu14qvf5.fsf@test.test.com>
Message-ID: <85f9i8$fg4$2@newsserver.rrzn.uni-hannover.de>

In article <ltiu14qvf5.fsf at test.test.com>,
	Test test <test at test.test> writes:
> test

Don't do that.
Test newspost are for test newsgroups not for real ones.
	Bernhard

-- 
Free Software Projects and Consulting 		         (intevation.net)  
Association for a Free Informational Infrastructure            (ffii.org)


From phd at phd.russ.ru  Fri Jan 21 10:15:19 2000
From: phd at phd.russ.ru (Oleg Broytmann)
Date: Fri, 21 Jan 2000 15:15:19 +0000 (GMT)
Subject: HTMLparser
In-Reply-To: <869lbo$r3$1@snipp.uninett.no>
Message-ID: <Pine.LNX.4.21.0001211514510.10102-100000@fep132.fep.ru>

On Fri, 21 Jan 2000, Asle Pedersen wrote:

> I'm a beginner Python user and experimenting with the HTMLparser. I want to
> convert all relative urls to absolute urls without thouching the rest of the
> file containment.
> 
> this is what I have so far. but somehow it is throwing away all tags but the
> anchor tags. (which is not what I want)

   You should add unknown_starttag and unknown_endtag...

> class minparser(htmllib.HTMLParser):
> 
>     def __init__(self, formatter, verbose=0):
>  htmllib.HTMLParser.__init__(self, formatter, verbose)
> 
>     def anchor_bgn(self, href, name, type):
>         self.anchor = urlparse.urljoin("baseurl",href)
>         if self.anchor:
>      self.save_bgn()
> 
>     def anchor_end(self):
>         if self.anchor:
>      text = self.save_end()
>     #need to do something here
>      self.handle_data("%s <%s>"%(text,self.anchor))
>      self.anchor = None

Oleg.
---- 
    Oleg Broytmann      Foundation for Effective Policies      phd at phd.russ.ru
           Programmers don't die, they just GOSUB without RETURN.




From paul.robinson at quantisci.co.uk  Wed Jan  5 11:35:36 2000
From: paul.robinson at quantisci.co.uk (Paul Robinson)
Date: Wed, 05 Jan 2000 16:35:36 +0000
Subject: Python and ADSI (more ADSI really)
References: <s5vh2ahnrg382@corp.supernews.com> <%PU74.7445$NV6.31305@news-server.bigpond.net.au>
Message-ID: <387372D8.309210B0@quantisci.co.uk>

On the subject of ADSI I have just returned to a problem I was having
several months ago...

I was using ADSI, specifically the WinNT provider. I was using Python as
it happens, but I have also tried a VBScipt version with exactly the
same results.
Any creation of an ADSI object was taking around 15 seconds. I could
find no mention of this in the SDK documentation or MSDN.
I started my searching for an answer again and found this useful nugget
at:
http://x33.deja.com/getdoc.xp?AN=516935348

Thus transforming my code from completely un-usable to really quite
useful....

E.g.
the code below (t2.cgi) produces the following output:

D:\Temp\cgi>python t2.cgi
Fast way: 0.039999961853
Slow way: 15.0319999456

So as you can see that is a significant speedup!!!
I hope this is useful to any ADSI users (and potential ADSI users)....

I thought I'd share this to alleviate (at least one of) the potential
problems that anyone could possibly have in the future.

	Paul.


#######################################################################
from time import time
import win32com.client, pywintypes

NS = win32com.client.Dispatch('ADSNameSpaces')

def bindUserWinNt(computerName, bindUser, userName=None,
                  passwd=None):
    if not userName:
        try:com = NS.getobject('','WinNT://' + computerName +
                               '/'+ bindUser +',user')
        except pywintypes.com_error, why:
            com = None
            print why[1], computerName, bindUser
    else:
        dso = NS.getobject('','WinNT:')
        try:com = dso.OpenDSObject('WinNT://' + computerName + '/'+
                                   bindUser + ',user', userName,
                                   passwd, 1)
        except pywintypes.com_error, why:
            com= None
            print why[1], computerName, userName
    return com

def bindUserWinNt2(computerName, bindUser, domain,
                   userName=None, passwd=None):
    if not userName:
        try:com = NS.getobject('','WinNT://' + domain+'/'+
                               computerName + '/'+ bindUser +',user')
        except pywintypes.com_error, why:
            com = None
            print why[1], computerName, bindUser
    else:
        dso = NS.getobject('','WinNT:')
        try:com = dso.OpenDSObject('WinNT://' + domain + '/' +
                                   computerName + '/' + bindUser +
                                   ',user', userName, passwd, 1)
        except pywintypes.com_error, why:
            com= None
            print why[1], computerName, userName
    return com

if __name__ == '__main__':
    t = time()
    u = bindUserWinNt2('pc-pdr2', 'test', 'isdg')
    print 'Fast way: '+`time()-t`
    t = time()
    u = bindUserWinNt('pc-pdr2', 'test')
    print 'Slow way: '+`time()-t`
#######################################################################



From gerrit.holl at pobox.com  Sun Jan 23 03:20:40 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 23 Jan 2000 09:20:40 +0100
Subject: Python compiler?
In-Reply-To: <86duuq$8a1@OM9.omantel.net.om>; from khimj@gto.net.om on Sat, Jan 22, 2000 at 06:02:13AM +0400
References: <86duuq$8a1@OM9.omantel.net.om>
Message-ID: <20000123092040.A2010@stopcontact.palga.uucp>

J Dubal wrote on 948517333:
> Hello good people!

Hello,

> I am new to python and I understand that although it is a free-to-use
> language, it can be used to build commercial, propritary applications.
> Is this correct?

Absolutly.

> Does it have a compiler for Linux and/or windows that can produce native
> executables?

Yes, it has one. It's included in the Python distribution, Tools/freeze.
It's called freeze, and with it, you won't need any Python interpreter.

> If not, do we have to distribute python application sources?

Well, not really, there's another way. Python produces bytecode,
which is also non-readable text. However, you'll need an
interpreter for it, and there are ways to decompile it.

> Are there modules or libs available to bulid data entry/query applications
> for character terminals?

I don't know which platform you're using, but on Unix, there is a
module called "curses". A slightly larger but less crossplatform (won't
work on all unices) is downloadable; search for it on www.vex.net/parnassus.

kind regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From grant at nowhere.  Mon Jan 17 13:42:39 2000
From: grant at nowhere. (Grant Edwards)
Date: Mon, 17 Jan 2000 18:42:39 GMT
Subject: Sucking e-mail from Exchange Server vis HTTP?
References: <slrn886kml.284.grant@grante.comtrol.com> <38835712.E787B04B@roguewave.com>
Message-ID: <slrn886oku.2ac.grant@grante.comtrol.com>

In article <38835712.E787B04B at roguewave.com>, bjorn wrote:

>> I just found the http server page on Exchange server that
>> allows access to e-mails via a web browser.  What I'd really
>> like is to write a Python program that just sucks the mail out
>> of the inbox and delivers it locally.

My other option would be to write a MAPI client that runs under
Win32 and then gives the mail to the SMTP server on my Linux
machine.  There's a MAPI module for Python, isn't there?
Perhaps that would be easier -- but I've never programmed under
Win32 before...

>It might be easier to use the exchange server as either an IMAP
>or POP server instead (both should be on by default).

The admin shut off both POP and IMAP to try prevent people from
using mail clients other than Outlook.  He's convinced
management that people who don't use Outlook were increasing
his workload.

IMAP used to be enabled to allow people to check e-mail from
home using Outlook Express, but once he figured out how to
enable the web interface, he told people to use that from home
and then he shut off IMAP (I was using fetchmail and mutt
at the time).

He shut off my dial-in access a month or two ago (I don't know
why, presumably because I like Unix better than his beloved
MS-Windows). The guy's a real prize. If he ever figures out how
to prevent Unix users from mounting SMB shared filesystems, I'm
sure he'll do that too...

-- 
Grant Edwards                   grante             Yow!  Hello. I know
                                  at               the divorce rate among
                               visi.com            unmarried Catholic Alaskan
                                                   females!!


From erno at iki.fi  Wed Jan 19 19:56:50 2000
From: erno at iki.fi (Erno Kuusela)
Date: 20 Jan 2000 02:56:50 +0200
Subject: no multithreading?
References: <864ih0$5is$1@nnrp1.deja.com>
Message-ID: <87k8l5tval.fsf@fun77.kivikko.hoas.net>

>>>>> "moonseeker" == moonseeker  <moonseeker at my-deja.com> writes:

    moonseeker> Hi!  I have understood that the python interpreter
    moonseeker> switches to another thread if a fixed amount of python
    moonseeker> commands is completed. Is this true?

after a fixed number of bytecode instructions have been executed, yes.
you can set it with sys.setcheckinterval().

    moonseeker> That means if a task hangs on a command (i.e. file
    moonseeker> open), all tasks are locked? What happens if I use
    moonseeker> sleep() in a task?

no, afaik i/o operations and sleep() and such release the interpreter
lock, enabling other threads to excute.

   -- erno


From fredrik at pythonware.com  Thu Jan 20 03:23:37 2000
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 20 Jan 2000 09:23:37 +0100
Subject: Python Rocks!
References: <000401bf623e$16e712c0$9a2d153f@tim> <slrn88cnpm.7ts.neelk@brick.cswv.com>
Message-ID: <00cf01bf631f$a7366880$f29b12c2@secret.pythonware.com>

Neel Krishnaswami <neelk at brick.cswv.com> wrote:
> The only time indentation has been a problem for me is when using
> eval/exec, and encoding programs as strings is basically the wrong
> thing to do anyway. This fact has occasionally tempted me to write a
> PythonGen module along the same lines as Robin Friedrich's HTMLgen. It
> would make writing things like code walkers and program transformers
> much more elegant. It seems like an idea obvious enough that someone
> else must have beaten me to the punch, though.

http://www.deja.com/getdoc.xp?AN=361628215

</F>




From tjreedy at udel.edu  Sun Jan 16 22:03:26 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Sun, 16 Jan 2000 22:03:26 -0500
Subject: [Types-sig] Suggestions for python 2
References: <m129zpL-00006JC@vortigen.demon.co.uk>  <200001170221.VAA10441@eric.cnri.reston.va.us>
Message-ID: <85u0qh$g5g$1@news.udel.edu>

Initial responses after trying to read the web-page proposal:

The tunneling idea of allowing name=expression pairs after *,** in function
definitions to indicate read-only compile-time-evaluated local constants
seems like a good one in that it would eliminate the problem with the
current pre-* default param hack (modifiability) without disabling current
correct code.

Much of the rest I did not get too well (but I feel better after reading
Guido's post).  One of the benefits of the current Python object system is
that it is fairly clear and simple.  This means that non-experts without a
CS degree can understand and use it.  It also means that such people
(including myself) can read and understand most code written by others.  The
straightjack which understandably frustrates some CS-level experts has some
benefit for the rest of us.

Some unification of similar concepts might be good, but not at the cost of
what seems like substantial complexification.

Terry J. Reedy





From scarblac-spamtrap at pino.selwerd.cx  Thu Jan 20 03:39:45 2000
From: scarblac-spamtrap at pino.selwerd.cx (Remco Gerlich)
Date: 20 Jan 2000 08:39:45 GMT
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <vndn1q340ez.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <slrn88diae.jje.scarblac-spamtrap@flits104-37.flits.rug.nl>

Justin Sheehy wrote in comp.lang.python:
> Nick Belshaw <nickb at earth.ox.ac.uk> writes:
> > > Get used to it, and don't use tabs.
> 
> > ! - whats wrong with tabs?
> 
> They don't look the same in every editor, or with each individual's
> tab-width settings.

But that's the *good* thing about tabs. You can make code look the way
you want even if you don't agree with the author on indentation style.

Tabs are fine, but mixing tabs and spaces may confuse you.


-- 
Remco Gerlich,  scarblac at pino.selwerd.cx

   This is no way to be
     Man ought to be free      -- Ted Bundy
       That man should be me


From seanh at unforgettable.com  Wed Jan 26 22:27:04 2000
From: seanh at unforgettable.com (Sean Hummel <seanh@unforgettable.com>)
Date: Thu, 27 Jan 2000 03:27:04 GMT
Subject: Installing an idle function
Message-ID: <Pine.LNX.4.20.0001270054020.12019-100000@saddam.pseudonetworks.com>

Does Python support a method for installing an Idle function from the code
of a "C"  module.


------------------------------------------------
I run halfway to my death, then turn around
and run back.
------------------------------------------------



From ht at cogsci.ed.ac.uk  Thu Jan 13 03:57:58 2000
From: ht at cogsci.ed.ac.uk (Henry S. Thompson)
Date: 13 Jan 2000 08:57:58 +0000
Subject: Perl speed vs. Python speed
In-Reply-To: "Fredrik Lundh"'s message of "Wed, 12 Jan 2000 16:06:10 +0100"
References: <85hjfa$mlp$1@nnrp1.deja.com> <pd0f4.26480$W2.358425@iad-read.news.verio.net> <00f801bf5d0e$ae601140$f29b12c2@secret.pythonware.com>
Message-ID: <f5baemaic15.fsf@cogsci.ed.ac.uk>

"Fredrik Lundh" <fredrik at pythonware.com> writes:

> if everything goes according to plan, 're' (and possibly also
> 'regex') will be replaced by a new unicode-aware engine in
> 1.6.  I've attached some (somewhat outdated) benchmarks.

Thanks for the details, I'm moderately desperate to get integrated
unicode support, when are we hoping to see the first 1.6 beta hit the
streets?

ht
-- 
  Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
     2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
	    Fax: (44) 131 650-4587, e-mail: ht at cogsci.ed.ac.uk
		     URL: http://www.ltg.ed.ac.uk/~ht/



From amused at webamused.com  Mon Jan 31 03:01:45 2000
From: amused at webamused.com (Joshua Macy)
Date: Mon, 31 Jan 2000 08:01:45 GMT
Subject: Can someone fix my code?
References: <8733rg$ral$1@nnrp1.deja.com>
Message-ID: <38953F58.E8A6FE65@webamused.com>

It would be easier to help you if you included the actual error that
you're getting, but it appears to me that part of your problem is that
you're forgetting which variables hold what (one of the reasons that
it's generally not a good idea to have too many variables named
temp_this and temp_that).  Also, commenting your intent can help.
 The specific problem seems to be that you attempt to use your output
file as input.  body is a file opened for writing, so attempting to loop
over it as in:
	for line in body:
		body.write(line)
generates an "TypeError: loop over non-sequence" exception.  That makes
sense: body is a file, not a sequence.  If what you're trying to do is
combine two existing files into a new file, you're going to need another
file.  Something like:
    body_file = open(body_path, "r")  # open the text of the web page
body FOR READING
    body = body_file.readlines()      # now body is a sequence, so for
line in body: should be okay
    out_path = raw_input("Please type the path of the output file: ")
    out_file = open(out_path, "w")  # open the output file
Then change body.write(line) to out_file.write(line) in the rest of the
file.

BTW, if you don't mind some suggestions on coding style, it's generally
regarded as poor practice to mix the interface (in this case the
questions about the file names) up with the code that's really doing the
work.  If, for instance, you moved the questions up into the ssi_or_exit
function, and then passed the file names in to the ssi function, it
would be a lot easier down the road to reuse the ssi function, e.g. to
provide a batch mode that gets the file names from the command line, or
from another file.

I hope this helps.

  Joshua



bovinesoft at my-deja.com wrote:
> 
> I am a complete Python newbie and am trying to learn it as my first real
> programming language (I have dabbled in C, Java, and Assembly, but never
> really used them).  I am trying to write a program to get some practice,
> but am getting errors when I try to run it.  If someone could take a
> look at my code and tell me what is wrong with it, I would be greatly
> appreciative.
> 
> import sys  # to use the exit() function
> 
> def ssi_or_exit():
>    sorx = raw_input("Would you like to generate HTML <g> or exit <x>? ")
>    if sorx == "g":  # if they want to generate a web page
>       ssi()
>    elif sorx == "x":  # if they want to exit
>       sys.exit("Thank you for using Bovine No Server SSI!")
>    else:
>       ssi_or_exit() # try it again
> 
> def ssi():
>    body_path = raw_input("Please type the path of the body section: ")
>    body = open(body_path, "w")  # open the text of the web page body
>    temp_path = raw_input("Please type the path of the HTML template: ")
>    temp = open(temp_path, "r")  # open the HTML template
>    temp2 = temp.readlines()
>    where_temp = raw_input("Put the template before <b> or after <a> the
> body? ")
>    if where_temp == "b":  #around here is where my trouble starts
>       for line in temp2:
>          body.write(line)
>       for line in body:
>          body.write(line)
>    else:
>       for line in body:
>          body.write(line)
>       for line in temp2:
>          body.write(line)
>    body.close()
>    temp.close()
>    ssi_or_exit()
> 
> ssi_or_exit()
> 
> Sent via Deja.com http://www.deja.com/
> Before you buy.


From nickliz at t-online.de  Sat Jan  1 14:08:41 2000
From: nickliz at t-online.de (Eide)
Date: Sat, 1 Jan 2000 20:08:41 +0100
Subject: need help getting tk to work with python
Message-ID: <84lje7$hrj$1@news00.btx.dtag.de>

I too have been having this problem.  Tk once worked, but does no longer.
I run W98 with python at C:\program files\python
and tcl at C:\program files\tcl

I tried the autoexec.bat fix as suggested at python.org to no avail.  Here
is what I get when trying the Tkinter._test()
--------------------------------------------------
Traceback (innermost last):
    File "<stdin>", line1, in ?
    File "C:\PROGRA~1\PYTHON\LIB\LIB-TK\Tkinter.py", line 1947, in _test
        root = Tk()
    File "C:\PROGRA~1\PYTHON\LIB\LIB-TK\Tkinter.py", line 886, in __init__
        self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories:
    {} ./lib/tcl8.0 C:/tcl8.0/library {C:/Program Files/library}

This probably means that Tcl wasn't installed properly.
--------------------------------------------------

My autoexec.bat reads:

PATH
C:\WINDOWS;C:\WINDOWS\COMMAND;C:\PROGRA~1PYTHON;C:\PROGRA~1\PYTHON\LIB;C:\PR
OGRA~\TCL\BIN

and I have tried every possible combination of "..\tcl\lib\" or
"..\tcl\lib\tcl8.0" etc and again, to no avail.  I also uninstalled it all,
cleaned my registry and reinstalled it.  I have no clue at this point.
Please advise.

Nick





From bitbucket at isomedia.com  Thu Jan  6 22:35:57 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Fri, 07 Jan 2000 03:35:57 GMT
Subject: python with ASP - pytest.asp (1/1)
References: <000001bf58c9$8c26ff60$39f4fea9@desktop>
Message-ID: <38755f1d.450186874@mail.isomedia.com>

begin 644 pytest.asp
M/"5`($Q!3D=504=%(#T at 4'ET:&]N("4^#0H\)0T*:68@;&5N("A297%U97-T
M+D9O<FTH(G1I=&QE(BDI. at T*"7-3='(@/2`G/&(^($=O="!4:71L93PO8CXG
M#0IE;'-E. at T*"7-3='(@/2`G)R<\1D]232!!0U1)3TX](BXO<'ET97-T+F%S
M<"(@;65T:&]D/2)03U-4(CX-"@D)/&(^5&ET;&4Z)FYB<W`F;F)S<"9N8G-P
M)FYB<W`F;F)S<"9N8G-P)FYB<W`F;F)S<#PO8CX\0E(^#0H)"3QI;G!U="!T
M>7!E/2)T97AT(B!V86QU93TB(B!S:7IE/38P(&YA;64](G1I=&QE(CX\4#X-
M"@D)/&EN<'5T('1Y<&4]<W5B;6ET('9A;'5E/2)3=6)M:70B/@T*"0D\+V9O
H<FT^)R<G#0HE/@T*/&AT;6P^#0H\)3T@<U-T<B`E/@T*/"]H=&UL/@``
`
end


From loewis at informatik.hu-berlin.de  Fri Jan 28 15:35:34 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 28 Jan 2000 21:35:34 +0100
Subject: nesting class definitions?
References: <20000127155551.A2069@stopcontact.palga.uucp>
 <slrn891nqh.bt0.neelk@brick.cswv.com>
 <20000128134010.C1005@stopcontact.palga.uucp>
Message-ID: <p6qd7qmj5nt.fsf@informatik.hu-berlin.de>

Gerrit Holl <gerrit.holl at pobox.com> writes:

> Ah, but what I meant was:
> 
> class Jan:
>     class Piet:
>         ...
>     class Henk:
>         ...
> 
> Is there any point in using this feature?

I've seen it used in the Gnome libglade Python module, for
encapsulation reasons: The name started with __, so the only way to
create instances of this class was from inside methods of the outer
class (or by knowing how __ names are translated). I feel they should
have better used a module-scoped class, starting with an underscore
(so from libglade import * would not import it).

Another example is in the threading module's test() function, where
classes are nested inside functions; the same happens in copy._test.

profile.Profile has a nested class fake_code, again for hiding
purposes. Whether the use is justified is up to the reader.

Regards,
Martin



From tim_one at email.msn.com  Thu Jan 20 18:53:12 2000
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 20 Jan 2000 18:53:12 -0500
Subject: Python Rocks! - get rid of colons
In-Reply-To: <8680qp$e9t@news.or.intel.com>
Message-ID: <000001bf63a1$832e5740$6ea0143f@tim>

[posted & mailed]

[tye4]
> I intend to modify the language's syntax to add
> 'end' delimiters for the blocks.

[Sami Hangaslammi]
> Wouldn't it be better (and easier) to write a script
> that converts between .py and .pye (python with end
> delimeters) scripts. This would allow you to both
> use existing Python libraries and distribute modules
> you have made to others.

[tye4]
> Brilliant idea. Thanks. That's a lot less work than I
> had planned.

As I mentioned a couple days ago in this thread, 99% of such a script ships
with Python, in Tools/Scripts/pindent.py.  From the top of that file:

# This file contains a class and a main program that
# perform two related (though complimentary) formatting
# operations on Python programs.  When called as
# "pindent -c", it takes a valid Python program as input
# and outputs a version augmented with block-closing
# comments.  When called as "pindent -r" it assumes its
# input is a Python program with block-closing comments
# but with its indentation messed up, and outputs a
# properly indented version.

I haven't used this script myself (I prefer indentation, for reasons covered
earlier), but there's no good reason it *can't* be used for its stated
purpose <wink>.

block-closing-ly y'rs  - tim





From slinkp at angelfire.com  Fri Jan  7 13:55:52 2000
From: slinkp at angelfire.com (Paul Winkler)
Date: Fri, 07 Jan 2000 13:55:52 -0500
Subject: Who hosts CGI/Python?
References: <3873EE03.1CE5F8CC@usit.net>
Message-ID: <387636B8.F71D42B7@angelfire.com>

Chuck Esterbrook wrote:
> 
> Seems like many web host providers, such as MindSpring, offer CGI with
> Perl, but not other languages, notably Python.
> 
> Does anyone know who would host a CGI Python site?

www.jumpline.net seems to have pretty good service for the money...
I'm hosting two domains there. Their FAQ mentions only Perl, but I
just telnetted in and typed "python" and got the interactive prompt.
Tried a quick dumb CGI script too, it worked.

-- PW
................    paul winkler    ..................
slinkP arts:   music, sound, illustration, design, etc.
A member of ARMS    ----->    http://www.reacharms.com
or http://www.mp3.com/arms or http://www.amp3.net/arms
personal page   ---->    http://www.ulster.net/~abigoo


From boud at rempt.xs4all.nl  Mon Jan  3 01:41:03 2000
From: boud at rempt.xs4all.nl (Boudewijn Rempt)
Date: 3 Jan 2000 06:41:03 GMT
Subject: DB API
References: <999u6sog16j9fci9il2vrivvrtbea896t4@4ax.com> <84nfct$bal$1@news1.xs4all.nl> <umav6ssq7uubv4gpb192lp2fs1jcan6hc2@4ax.com> <84olef$5j1$1@news1.xs4all.nl> <84omaa$qln$1@nntp9.atl.mindspring.net>
Message-ID: <84pg9v$rda$1@news1.xs4all.nl>

Aahz Maruch <aahz at netcom.com> wrote:

> Mind expanding on this?  Are you asserting it's also a Bad Idea to slurp
> multi-megabyte BLObs to a client?

Well, what databases are good at, is picking one thing out of many.
If you use a database as a flat file, and then do the picking of one
out of many you've essentially written a distributed database. I've
never used blobs myself - I know they can be useful, even necessary,
but until now I've just stored an url in the database, and put the blobs
behind a http server. That way, the database doesn't have to work so
hard, and it becomes easy to distribute the load over two machines.

-- 

Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt


From shouldbe at message.com  Thu Jan 20 08:10:04 2000
From: shouldbe at message.com (Arinte)
Date: Thu, 20 Jan 2000 13:10:04 GMT
Subject: A modest Error (I mean Proposal)
Message-ID: <MODh4.1808$D4.62243@newsread1.prod.itd.earthlink.net>

I am getting this weird error???
Traceback (innermost last):
  File "Beer", line 17, in ?
  File "D:\POSSApps\anotherposs app\PossDevice.py", line 37, in IOCTL
    if type(command)==type(2):
TypeError: len() of unsized object

Here is the code for the function it craps out on...

 def IOCTL(self, command, dargs):
  command1 = long(0)
  if type(command)==type(2):
   poss.write("IOCTL was int")
   command1 = long(command)
   poss.write("IOCTL Set to Long");
  else:
   if type(command) == type("  "):
    poss.write("IOCTL was a string")
    command1 = command
   else:
    if type(command)!=type(2L):
     poss.write("IOCTL fail")
     return -1
  self.dump();
  poss.IOCtl(self.devname,command1,dargs) #make this pass the ioctl command
as an PossArg with a special name
  return 0

This is part of a middle class between my C++ program and the user written
python code.




From anders.eriksson at morateknikutveckling.se  Tue Jan 11 10:56:04 2000
From: anders.eriksson at morateknikutveckling.se (Anders M Eriksson)
Date: Tue, 11 Jan 2000 16:56:04 +0100
Subject: mxODBC
Message-ID: <rk17OF5=o=S0QxS2fUYwTLmxxrno@4ax.com>

hELLO!

I'm trying to use the mxODBC module to access an dBASE 5 database on a
Windows NT system. I have used this script:

import ODBC.Windows

db = ODBC.Windows.connect('myDatabase')		# MyDatabase is a
local DSN
<snip>
db.close()

When I run this script I get this error on the connect line

NotSupportedError: ('S1C00', 106, '[Microsoft][ODBC dBase
Driver]Optional feature not implemented ', 4226)

If I use the Win32 odbc then it works!

any idea what is wrong??

// Anders



From gmcm at hypernet.com  Fri Jan 21 08:49:06 2000
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 21 Jan 2000 08:49:06 -0500
Subject: Whitespace delimiters suck
In-Reply-To: <roy-11C630.23024320012000@netnews.nyu.edu>
Message-ID: <1263658532-14676484@hypernet.com>

Roy Smith wrote:

> When I was first learning programming (a quarter of a century ago!), 
> there was the idea of the "telephone test", i.e. your code should be 
> simple enough that if you read it to somebody on the phone they would 
> understand it.  Python not only fails the phone test,

Do the same thing that the parser (or tokenizer) does: say 
"INDENT" and "DEDENT". Or run it through pyindent and read 
that over the phone.

> it fails the 
> printed page test; I can print two programs on a piece of paper, which 
> differ only in tabs and spaces, and they will appear identical yet which 
> do different things.  Barf.

While theoretically possible, I have never seen whitespace 
confusion induced by different editor settings create a 
syntactically valid program.

And there's tabnanny and tabcleaner if this is a common 
problem for you.

- Gordon



From Gaetan_Corneau at baan.com  Wed Jan 26 14:16:54 2000
From: Gaetan_Corneau at baan.com (Gaetan Corneau)
Date: Wed, 26 Jan 2000 14:16:54 -0500
Subject: DARREN MULLIGAN SEEKING MODELS! 89582
Message-ID: <816010E2456BD111A48700805FBBE2EE01C60408@ex-quebec-u1.baan.com>

I'm really tempted to hack this guy's machine. 
Destroy everything. 
Or at least bust his mailbox.

The kind of spammer I hate the most.
Sorry, I know this is off-topic.
______________________________________________________
   Gaetan Corneau
   Software Developer 
   Software Engineering Process Group
   BaaN  Supply Chain Solutions  
   http://www.baan.com
   E-mail: Gaetan_Corneau at baan.com                  
   Tel: (418) 266-8252         
______________________________________________________
"Profanity is the one language all programmers know best"


-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS/B/MU d- s+:++ a+ C++ 
UL+ P--- W+ N- K- W++ t-- 
!5 X- R+ tv-- b++ DI++ 
G e++ h---- r+++ y++++
------END GEEK CODE BLOCK------


> -----Original Message-----
> From: 
> 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> [mailto:24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.
> pacbell.ne
> t]
> Sent: Sunday, January 23, 2000 3:45 PM
> To: python-list at python.org
> Subject: DARREN MULLIGAN SEEKING MODELS! 89582
> 
> 
> My name is Darren Mulligan and I am an aspiring photographer 
> from California.
> I will get right to the meat and gravy of this post, so as 
> not to waste your time!
> 
> +++++WHAT AM I SEEKING?+++++
> 
> I am looking for young models (prefer early teen 14-16 year 
> old female) for nude and semi-nude photography. This is a 
> great way for me to update my portfolio, and also a great way 
> for you to get a free portfolio, too! This could lead to 
> employment in the Adult XXX entertainment business. What a 
> great way to make money!
> 
> Please respond to one of the following e-mail addresses. I 
> run my very own mail server on an ADSL connection so you can 
> be assured to complete discretion and privacy. If one e-mail 
> address does not work, simply try the next! 
> 
> Sorry for the long list! My ADSL connection goes down from 
> time to time, so I posted all the e-mail addresses I can be 
> reached at. If one does not work, just move on to the next!
> 
> nanae at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> usenet at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> 24hoursupport.helpdesk at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> news.admin.net-abuse.email at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> spamlinkpage at adsl-63-197-143-124.dsl.scrm01.pacbell.net
> alt.make.money.fast at nntp.8m.com  nanas at nntp.8m.com   
>   hostmaster at FREESERVERS.COM
> 
> THIS IS NOT A SCAM! IT IS NOT CHILD PORN!!
> There is nothing wrong or illegal about the artistic 
> representation of a young body. If you have a problem with 
> this post, please e-mail me and I will not post to your 
> newsgroup again!
> 
> Sincerely,
> Darren Mulligan
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From paul.magwene at yale.edu  Wed Jan 19 16:54:17 2000
From: paul.magwene at yale.edu (Paul M)
Date: Wed, 19 Jan 2000 16:54:17 -0500
Subject: A modest proposal
Message-ID: <865bmq$dum$1@news.ycc.yale.edu>

Dear Pythoneers,

Inspired by recent postings to this newsgroup I've decided that
Python, and in particular the entire Python user base, has been going
about things all wrong.  I've decided I would like to be able to code
in python something like this:

def fun(a,b,c)
{
    if a eq b:
  if b ne c
                 print "Oooh, random changes are fun"
                   a = b = c
                 end
end
}

Important points to note:

1) I can use whatever god-awful indentation scheme I want
2) Colons after "def" and "if" are purely discretionary
3) The compiler can decide what I mean when I use "="
4) Readability is greatly improved

I can't believe that nobody has changed this before - can't you see
that MILLIONS of users are crying out for these features!

<python-schmython>-ly y'rs - Paul





From thomas at xs4all.net  Thu Jan  6 07:29:59 2000
From: thomas at xs4all.net (Thomas Wouters)
Date: Thu, 6 Jan 2000 13:29:59 +0100
Subject: Simple cartesian product
In-Reply-To: <Pine.LNX.4.10.10001061110050.6929-100000@gaia.gfy.ku.dk>; from tk@gaia.gfy.ku.dk on Thu, Jan 06, 2000 at 11:13:29AM +0100
References: <000801bf571c$0cf448a0$1a2d153f@tim> <iacembwr7qg.fsf@vier.idi.ntnu.no> <Pine.LNX.4.10.10001061110050.6929-100000@gaia.gfy.ku.dk>
Message-ID: <20000106132958.E7015@xs4all.nl>

On Thu, Jan 06, 2000 at 11:13:29AM +0100, Thomas Knudsen wrote:

> > Now - you're not going to get that "new millennium"-discussion
> > started, are you? We are still in the same millennium, century, and
> > decade as last year. The third millennium starts with 2001.01.01.

> You should, however, use proper ISO8601 syntax and write 2001-01-01
> This is actually quite important: We really *need* a consistent,
> global format for numeric date representation. And since we actually
> have it - *use it!*

So is this YYYY-MM-DD or YYYY-DD-MM ? This creates a lot of confusion
whenever I need to exchange dates with americans, or with people who think
_I'm_ american. (the nerve.) Or is ISO8601 a real standard in the
ANSI sense, and thus claim two different things in different sections of the
standard, and later claim it's *both* YYYY-MM-DD and YYYY-DD-MM ? :)

History---will-teach-us-nothing-ly y'rs.
-- 
Thomas Wouters <thomas at xs4all.net>

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



From robin at jessikat.demon.co.uk  Sun Jan 16 03:41:20 2000
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 16 Jan 2000 08:41:20 +0000
Subject: How to use Pythonwin's ability to access windows dll's???
References: <1264133143-1099206@hypernet.com>
 <007701bf5fd4$4b1d4c60$01646464@computer>
Message-ID: <AsxhMAAwQYg4Ewbp@jessikat.demon.co.uk>

In article <007701bf5fd4$4b1d4c60$01646464 at computer>, Neil Hodgson
<neilh at hare.net.au> writes
>> Seriously what is the purpose of the LoadLibrary function if not to
>> access Dll's>
>
>   One thing you can do is load DLLs that define window classes and then
>create windows of those classes. Then you send messages to those windows.
>This is how Scintilla is integrated into PythonWin.
>
>   I'd like to see calldll in the Python distribution rather than in
>PythonWin. Its only a very small wafer of code that provides access to all
>OS functionality. Most of my chain saw scars heal eventually anyway.
>
>   Neil
>
>
>

There are two more general libraries that could be used as the guts of
the calldll thing. There's libffi (from cygnus) and ffcall from Bruno
Haible, <haible at clisp.cons.org>. Both of these support multiple
platforms via a well defined api. Roger E Critchlow Jr recently provided
a Tcl interface for these, replacing an earlier calldll like thing which
I had hacked up a couple of years ago and which only worked on x86
(win32 & linux). The only serious disadvantage is that various C++
problems are likely to remain eg the name mangling is different for
different compilers (even on win32) and the this passing convention is
quite variable (often using a register).
-- 
Robin Becker


From johngrayson at home.com  Fri Jan 21 18:07:01 2000
From: johngrayson at home.com (John Grayson)
Date: Fri, 21 Jan 2000 23:07:01 GMT
Subject: Python and Tkinter Programming 
Message-ID: <3888E695.A929D7C2@home.com>

I know that this is a *shameless* plug, but anyway...

My book, Python and Tkinter Programming, is shipping 
from the printer today. Copies should be available at IPC8.

There are two example chapters, in PDF format, on
the Manning web site. If you'd like to see them
before deciding to buy or not, point your browser
at http://www.manning.com/grayson/

     Cheers!

       :-j ohn


From loewis at informatik.hu-berlin.de  Thu Jan 20 07:47:56 2000
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 20 Jan 2000 13:47:56 +0100
Subject: Using glade to build guis
References: <38845051.5A9A1A49@bibsyst.no>
Message-ID: <p6q3drsdi4j.fsf@informatik.hu-berlin.de>

Thomas Weholt <thomas at bibsyst.no> writes:

> I`ve tried to use Glade to build guis and run them with the glade.py
> module/script. What I don`t understand is how to make it react to
> keypressing or other events. Does a manual/tutorial/example exists on
> how to prototype a gui in glade then code events-handling and the
> functional parts of the gui later ??

Documentation is largely missing, it appears. To connect to signals,
you first have to declare a signal handler in glade. You can then
connect the signal handler with a specific function. To do so, you can
either use .connect(signalname, function), or .auto_connect(dictionary
of signal/handler pairs).

Hope this helps,
Martin


From shouldbe at message.com  Wed Jan 19 16:23:24 2000
From: shouldbe at message.com (Arinte)
Date: Wed, 19 Jan 2000 21:23:24 GMT
Subject: To decref or not to dercref that is my question???
Message-ID: <gXph4.77$N4.3733@newsread1.prod.itd.earthlink.net>

When using PyObject_GetItem, when I am done getting the value that I need
from it should I Py_XDECREF it?

How do I know when to use Py_XDECREF?




From kain at abalonesoft.com  Sun Jan 16 11:27:17 2000
From: kain at abalonesoft.com (Scott Barron)
Date: Sun, 16 Jan 2000 16:27:17 GMT
Subject: Stopping a thread from the calling process
Message-ID: <FoFsDH.8EL@boss.cs.ohiou.edu>

is there an easy and fairly portable way of doing this?


From linuxtech at redshift.com  Tue Jan 18 23:49:34 2000
From: linuxtech at redshift.com (David Correa)
Date: Tue, 18 Jan 2000 20:49:34 -0800
Subject: MySQLdb-0.1.1 install on linux
Message-ID: <3885425E.63888079@redshift.com>

Hi,

 I dl. MySQLdb-0.1.1 and im trying to install the module.
When i run "python build.py" i ends with this error:

{sniped stuff was here}

/lib/python1.5/config/Setup
make[2]: Leaving directory `/home2/MySQLdb-0.1.1'
make[1]: Leaving directory `/home2/MySQLdb-0.1.1'
gcc -fPIC -g -O2 -I/usr/include/python1.5 -I/usr/include/python1.5
-DHAVE_CONFIG_H  -I/usr/include/mysql -c ./_mysqlmodule.c
./_mysqlmodule.c: In function `_mysql_ConnectionObject_affected_rows':
./_mysqlmodule.c:478: warning: return makes pointer from integer without
a cast
./_mysqlmodule.c: In function `_mysql_ConnectionObject_insert_id':
./_mysqlmodule.c:817: warning: return makes pointer from integer without
a cast
./_mysqlmodule.c: In function `_mysql_ResultObject_num_rows':
./_mysqlmodule.c:928: warning: return makes pointer from integer without
a cast
./_mysqlmodule.c: In function `init_mysql':
./_mysqlmodule.c:1370: warning: assignment makes pointer from integer
without a
cast
gcc -shared  _mysqlmodule.o  -L/usr/lib/mysql -lmysqlclient -o
_mysqlmodule.so
Traceback (innermost last):
  File "build.py", line 14, in ?
    import MySQLdb
  File "MySQLdb.py", line 19, in ?
    from _mysql import *
ImportError: ./_mysqlmodule.so: undefined symbol:
PyLong_FromUnsignedLongLong  
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

any help would be appreciated,

dave


From juxjux at hotmail.com  Wed Jan 12 08:40:51 2000
From: juxjux at hotmail.com (jux)
Date: Wed, 12 Jan 2000 14:40:51 +0100
Subject: Newbie: Python accepted?
Message-ID: <85i46b$jrl$1@zingo.tninet.se>

Hi.

I've just made my first Python-script and signed up at
free.prohosting.com........ (I can't afford to pay for webspace)
Since they don't accept Python-scripts but only Perl-scripts I wonder if you
can help me.
Is there anyway to convert my script to a Perl-script (So I don't need to
learn Perl),
OR do you know any web-server where I can get some space for free and they
accept Python-scripts?


I hope you can help me! (A hopeful Newbie)

In case you want to look at the script (Maybe you could help me to make it a
Perl-script?) i put it here:



import string
import cgi

form = cgi.FieldStorage()

if form.has_key("email"):email = form[email].value

if form.has_key("firstname"):firstname = form[firstname].value

if form.has_key("lastname"):lastname = form[lastname].value

if form.has_key("text"):text = form['text'].value

filehandle = open ('contact/' + email+ '.txt', 'w')
filehandle.write(firstname)
filehandle.write(lastname)
filehandle.write(text)
filehandle.close()







From scherbi at bam.com  Mon Jan 10 07:20:43 2000
From: scherbi at bam.com (Bill Scherer)
Date: Mon, 10 Jan 2000 07:20:43 -0500
Subject: Zope and NT databases
References: <387956BD.CFDF17DD@bby.com.au>
Message-ID: <3879CE9B.BE77787C@bam.com>

You want ZODBCDA:

http://www.zope.org/Products/DA/ZODBCDA


Sarah Burke wrote:

> Hello...
> I want to connnect up to a NT Access 97 Database through Zope running on
> Unix. Is the mxODBC module in an External Method the way to go?
> Thanks in advance
>
> --
> http://www.python.org/mailman/listinfo/python-list

--
William K. Scherer
Sr. Member of Applications Staff
Bell Atlantic Mobile
914 365 7265


-------------- next part --------------
A non-text attachment was scrubbed...
Name: scherbi.vcf
Type: text/x-vcard
Size: 226 bytes
Desc: Card for Bill Scherer
URL: <http://mail.python.org/pipermail/python-list/attachments/20000110/52695dd3/attachment.vcf>

From gerrit.holl at pobox.com  Wed Jan 12 10:59:32 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Wed, 12 Jan 2000 16:59:32 +0100
Subject: a small problem
In-Reply-To: <20000112145107.B8450@xs4all.nl>; from thomas@xs4all.net on Wed, Jan 12, 2000 at 02:51:07PM +0100
References: <20000112015923.A8450@xs4all.nl> <000701bf5cb4$c72320a0$da2d153f@tim> <20000112145107.B8450@xs4all.nl>
Message-ID: <20000112165932.A2758@stopcontact.palga.uucp>

Thomas Wouters wrote on 947685067:
> On Tue, Jan 11, 2000 at 11:23:29PM -0500, Tim Peters wrote:
> 
> > Whereas people like Guido <wink> would not have considered writing anything
> > other than
> > 
> > >>> sum = 0
> > >>> for n, s in q:
> >         sum = sum + n
> 
> Aahh, yes. I dont know what I was thinking when I wrote it using map and
> reduce instead of a much more obvious for loop. My only excuse is that it
> was late yesterday night, and I'd just come back from an Amsterdam Perl
> Mongers meeting. Amusing evening, but it probably confused me into using the
> least obvious of all the ways to do it ;)

:-)

> > Seriously, I don't think I've ever seen a good use for reduce in Python --
> > although I've certainly used it <wink>.
> 
> People still care about speed.

People should hang around in comp.lang.c or even comp.lang.asm.i386 if
they care about speed.

> I know I sometimes do, even though I know
> intellectually that it isn't going to matter much in the end product. I
> still like to make it as speedy as possible, even if that entails using
> map/reduce or odd looking constructions like 'def spam(text, None=None)'. I
> suspect the use of map/reduce will drop when python is optimisable enough
> to make the for loop _feel_ faster than map(). Static typing and what not.

I never used reduce(), but I did use map(), filter() & lambda.

> On the other hand, oneliners will always feel or look faster to some people.

Sure:
l = filter(lambda x: x%3==0, range(100))
I _really_ read this one somewhere!

It's more complicated than range(0, 100, 3), but, er... ;-)

> I'm one of them, I must admit. Another thing is the geekiness and
> playfulness of oneliners. 'look, mom, only one line.' Maybe this is where
> list comprehensions come in ;)

Well, is most languages, *everything* is possible in one line. tr -d '\n'
does a lot in c...

> [ <snip> Guido saying lambda's suck (compared to the rest of python) ]

16:50:52:tty11:gerrit at stopcontact:~/cvs/ro/python/dist/src/Lib$ grep -w lambda *.py
Bastion.py:def Bastion(object, filter = lambda name: name[:1] != '_',
CGIHTTPServer.py:        nobody = 1 + max(map(lambda x: x[2], pwd.getpwall()))
asyncore.py:                    lambda x: string.join (x, '|'),
bdb.py:                 s = s + "<lambda>"
htmlentitydefs.py:    'Lambda': '&#923;',       # greek capital letter lambda, U+039B ISOgrk3
htmlentitydefs.py:    'lambda': '&#955;',       # greek small letter lambda, U+03BB ISOgrk3
imaplib.py:             l = map(lambda x,j=j:'%s: "%s"' % (x[0], x[1][0] and j(x[1], '" "') or ''), l)
keyword.py:        'lambda',
mailcap.py:        entries = filter(lambda e, key=key: e.has_key(key), entries)
poplib.py:              digest = string.join(map(lambda x:'%02x'%ord(x), digest), '')
profile.py:                     t = reduce(lambda x,y: x+y, t, 0)
random.py:      # ('lambda' is a Python reserved word)
re.py:                repl = lambda m, repl=repl, expand=pcre_expand: expand(m, repl)
types.py:LambdaType = type(lambda: None)         # Same as FunctionType
xdrlib.py:        (up.unpack_uint,   (), lambda x: x == 9),
xdrlib.py:        (up.unpack_bool,   (), lambda x: not x),
xdrlib.py:        (up.unpack_bool,   (), lambda x: x),
xdrlib.py:        (up.unpack_uhyper, (), lambda x: x == 45L),
xdrlib.py:        (up.unpack_float,  (), lambda x: 1.89 < x < 1.91),
xdrlib.py:        (up.unpack_double, (), lambda x: 1.89 < x < 1.91),
xdrlib.py:        (up.unpack_string, (), lambda x: x == 'hello world'),
xdrlib.py:        (up.unpack_list,   (up.unpack_uint,), lambda x: x == range(5)),
xdrlib.py:         lambda x: x == ['what', 'is', 'hapnin', 'doctor']),

Only 15 real lambda's, in fact. It looks like Guido has written a very large
part of the library...


But suppose I have:
t = [('foo', 'bar'), ('foo_two', 'bar_two'), ('foo_three', 'bar_three')]
first_ones = map(lambda s: s[0], t)
last_ones = map(lambda s: s[1], t)

How would I do that without lambda's, if this code is in a method? Should
I nest two functions?

> > I suppose finally posting that will rekindle a few old flamewars!  Brrrr.
> 
> Weeellll, I never bothered much with flamewars (though I did see a spot of
> red when Tom Christiansen made his appearance last year) but I was always
> sceptical when I heard someone say such things. Surely lambda's are good in
> some cases ! But lately I've been looking at a lot of things differently and
> considering wether the good it brings in some cases justifies the pain it
> brings in other cases (this goes for much more than just lambda, python,
> computers or corporeality, though.)
> 
> So I hope potential flamers are willing to take a step back and do what I
> originally did. Start with thinking, "Guido Is Right", then consider, "Why
> is Guido right ? What is he saying, exactly ?" and ending up in argument
> with yourself. If you do end up thinking lambda's are guido's gift to sane
> coding, at least you'll be prepared for the flamewar, having heard all
> arguments.

lambda's are good, because otherwise, it wouldn't be possible to write
obfuscated code in Python... Can someone please rewrite the one-liner
mandelbrot without lambda's? And after that, without reduce()? And without
filter()? And map()? And apply()? And __import__()? And any builtins?
In c, you can write anything with only getchar() and putchar() ;-)

> It's important, though, once started, to finish this process. Dont get stuck
> at 'Guido Is Right', even if it's a very comfortable situation and nobody is
> going to notice in the end. The bureau of AFT might pull a Waco on Guido,
> after all :P

Guido Is Right. Linus Is Right. Eric Raymond Is Right.

Let's hope there'll never be a flamewar between Guido, Linus, Eric,
Bruce, ...

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

Oh, help! My computer is infected! What should I do now?!

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my mail!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From mlh at swl.msd.ray.com  Mon Jan 31 09:32:31 2000
From: mlh at swl.msd.ray.com (Milton L. Hankins)
Date: Mon, 31 Jan 2000 09:32:31 -0500
Subject: [Thread-SIG] RE: concurrency in Python
In-Reply-To: <000001bf6a98$9142a360$78a0143f@tim>
Message-ID: <Pine.SGI.4.21.0001310927590.44373-100000@masox201.msd.ray.com>

Greg Wilson wrote:

> Hi.  I'm interested in hearing from anyone who's ever built a
> concurrency toolkit for Python that used a model other than
> threads + locks --- futures, tuple spaces, active expressions,
> concurrent aggregates, actors, or anything else.

PyBrenda is a tuple space implementation using 100% Python.  It only runs
on UNIX systems.  The URL:

http://www.snurgle.org/~pybrenda

Feel free to contact pybrenda at snurgle.org if you have any comments or
questions.

--
              Milton L. Hankins              \\  ><> Ephesians 5:2 ><>
      http://www.snurgle.org/~mhankins       //  <mlh at swl.msd.ray.com>
    These are my opinions, not Raytheon's.   \\      W. W. J. D. ?




From bitbucket at isomedia.com  Mon Jan 10 18:21:56 2000
From: bitbucket at isomedia.com (Eugene Goodrich)
Date: Mon, 10 Jan 2000 23:21:56 GMT
Subject: Filelisting in Tkinter
References: <3879C56F.FF577344@bibsyst.no>
Message-ID: <387a6921.780429207@news.isomedia.com>

How about using either a single Listbox widget with a fixed-width font
and formatting each row yourself, or using multiple Listbox widgets,
shoulder-to-shoulder, with a shared scrollbar?  The latter should
auto-width on their own, I would think.

     -Eugene

On Mon, 10 Jan 2000 12:41:35 +0100, Thomas Weholt <thomas at bibsyst.no>
wrote:

>Hi,
>
>Is there a widget suitable for filelisting in Tkinter?
>
>I got a filename, size, modification date etc. that I want to put into a
>grid-like list, with no visible lines bewteen "cells". Similar to any
>decent filemanager.
>
>Any tips?
>
>Thomas

import binascii; print binascii.a2b_base64 ('ZXVnZW5lQGlzb21lZGlhLmNvbQ==')


From sholden at bellatlantic.net  Wed Jan  5 09:43:48 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Wed, 05 Jan 2000 14:43:48 GMT
Subject: control structures (was "Re: Sins")
References: <Pine.LNX.4.10.9912261534560.8355-100000@palin.python>
Message-ID: <387358BB.E71EC4E7@bellatlantic.net>

Phil Jensen wrote:

> [...] 
> I forget if I posted this before, but what I'd like to
> see is the Zahn construct discussed in Knuth's article -
> Pythonically:
> 
>     loop until "element found" or "search failed":
>         ...
>         if blah1 == blah2:
>             "element found"
>         ...
>     when "element found":
>         ...
>     when "search failed":
>         ...

A common Pascal idiom, I seem to remember, could be Pythonically
expressed as something like:

    loopstate = "searching"
    while loopstate = "searching":
        ...
	if bah1 == blah2:
            loopstate = "found"
        elif <out of elements>:
            loopstate = "not found"
        else:
            <move on to next element>

Typically one would use enumerations rather that strings to
represent the loop states, to combine readability with efficiency.

> 
> Whatever the loop is, its exit or exits have some "meaning"
> in the program, and this control structure encourages an
> element of Literacy by forcing them to be named.
> 
One of the annoying features of the staement/expression dichotomy
is that one cannot return a value from a loop.  In Icon, for example,
everything is an expression, and so a loop returned the value of the
last expressions evaluated inside.  This made it easier to detect
different loop termination conditions.

> [...]
> 
> IMHO, "break 3" would be awful.
> 
I agree.  Again, the Icon "break" took an argument which was returned
as the value of the loop, and so it was possible to write

	break break break

to exit from three levels of looping.  However, given our propensity
to change code by adding additional iterative contexts, this would be
little better than "break 3" in terms of preserving semantics across
such modifications, and nowhere near as helpful as labelled breaks.

But, how to put the labels in?

regards
 Steve


From max at neomeo.com  Wed Jan 12 13:40:58 2000
From: max at neomeo.com (David Max)
Date: Wed, 12 Jan 2000 13:40:58 -0500
Subject: Embedding Python question
Message-ID: <387cca87$1_1@spamkiller.newsfeeds.com>

Hello everyone. I am developing an application that embeds Python, and I am
having some difficulty sorting out some of the Python/C API. I have noticed
that

PyRun_SimpleString("import os\nprint 'Current dir is %s' % os.getcwd()\n");

is not the same thing as:

PyRun_SimpleString("import os\n");
PyRun_SimpleString("print 'Current dir is %s' % os.getcwd()\n");

However, if I use PyRun_String(...) instead and pass an appropriate pointer
for the globals, then it works.

I think I understand why this is the case, but I am still sketchy on where
exactly those pointers to the globals and locals is supposed to come from.
This seems to work:

PyObject *main_module = PyImport_ImportModule ("__main__");
PyObject *main_dict = PyModule_GetDict(main_module);
PyRun_String("import os\n", Py_file_input, main_dict, NULL);
PyRun_String("print 'Current dir is %s' % os.getcwd()\n", Py_file_input,
main_dict, NULL);

This works, but I'm not sure that I understand what I'm doing. For example,
I am leaving the locals as NULL. What should go there? And how would I get a
handle to it? What does it mean to use Py_file_input as the start tag as
opposed to using Py_single_input or Py_eval_input?

Also, I have tried to replace the call to execute the string "import os"
with a call to one of the PyImport functions. Nothing I've hit on so far
seems to work, and I'm not sure that I can tell the difference between some
of the functions like PyImport_ImportModuleEx, PyImport_Import and
PyImport_AddModule. If anybody can help me figure out which call I should
use, I would be appreciative.

Thanks in advance,

- David




From jzouck at charm.net  Wed Jan 12 22:41:26 2000
From: jzouck at charm.net (John Zouck)
Date: Wed, 12 Jan 2000 22:41:26 -0500
Subject: can't find tk.tcl
References: <387BB7D8.F1888924@worldnet.att.net>
Message-ID: <387D4966.7BC156E6@charm.net>

I had a problem that sounds like this several weeks ago, and finally found that
there were some tcl/tk .dll files in /windows/system that , when deleted, made
things work right. Before I found this, I tried every other kind of clean-up and
reinstall of python to no avail. I had installed tcl/tk months (or years) before,
which probably caused these files to be there. I don't know why it worked, but it
did, and this is just a suggestion.

wes weston wrote:

> Guys,
>         At the command line prompt: "from Tkinter import *" then "w = Tk()" and
> a "usable" tk.tcl can't be found in the several places looked at. But,
> it's in one of them: C:\Progra~1\Tcl\lib\tk8.0.  The weirdest thing is
> I've been running programs that use Tkinter recently on this
> machine(WinNT Ws). It (Tkinter programs) works at work and on my
> portable.
>
>         I tried a fresh install in several different places. Help!
>
>                 wes

--
===================
John H. Zouck
jzouck at charm.net
===================




From ivanlan at callware.com  Sat Jan  8 20:36:55 2000
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Sat, 08 Jan 2000 18:36:55 -0700
Subject: Python plugin
References: <38771491.BA08010@iqsoft.hu>
Message-ID: <3877E637.D3C9CC11@callware.com>

Hi All--

Nemeth Miklos wrote:
> 

[snip]

> Thank you for all suggestions.
> 
> (1) I am completely fed up with Java after two years of permanent
> sucking (I suppose this may be the correct word).

Oh, trust me, it's the perfect word!  I like your phrasing, too, which,
while not usual, adds a distinctive character to a sentiment I share: 
"permanent sucking."  Yes, I like that.

> I am threatened by all languages or tools having the letter "J" anywhere
> in its name.

I finally got the chance to use Python at work for a very small project,
one that should have taken only a day.  A colleague of mine was working
on a web-based application which happens to also communicate with our
voice mail engine using tcp/ip sockets.  We're using the same packets to
log in and count messages, get info about them, and so on.  We wanted to
just build up the packets from scratch, rather than incur the overhead
of using a large, unwieldy and overly complex DLL to communicate with
the engine.  So, we proceeded.

I would have finished my project Friday, except that my friend was using
Java.  He was able to get logged in with the first packet and retrieve a
certain amount of information, but when he actually tried to retrieve
another sort of packet from the engine, it refused to talk to him. 
Eventually (like three hours or so) we discovered that Java was doing
all kinds of icky favors for him behind his back.  He had to build up
his packet using bytes *only*, which happen to be signed bytes.  The
form of casting required meant that to get a 32-bit unsigned integer
(for an opcode for the engine), we finally had to build up the 4 bytes
by assigning each byte separately, with the knowlege that both Java and
the engine use two's complement arithmetic.  For example, to get a
32-bit 0xFFFFFFFF, we had to assign -1 to each individual byte.  Other
opcodes were more complex.

I don't think this is very portable, myself.  Wasn't that a goal of
Java?

After arguing with Java all afternoon, my respect for Jim Huginin's
JPython has increased about a thousand-fold.



> Castanet was also suggested as an alternative - I tried it 1.5 yeras
> ago: never again anything from Marimba.
> 
> (2) No more COM or any Microsoft-only garbage. The only reason I am
> looking for plugin because (apart from simple HTML) this seemed to be
> supported by the browser giants. Everything else (Java applets - a
> nightmare!, COM, DHTML) are browser specific.
> 
> My conclusion after this discussion thread:
> (1) Grail was not mentioned as an alternative. I am a bit sad that this
> project was a failure.

Yah, me too.  But it would take an awful lot of funding to build a real
web browser in Python.  Maybe someday.

> (2) Why on earth do I want to use plugins!? If I want to create a
> sophisticated (ie. non-HTML) client, I do not need any browser, I may do
> it easily with PyGTK (I tried it it seems very fast and simple to use),
> PyQt, etc. communicating via HTTP, XML-RPC, CORBA (all supported by
> Python) with my server. If HTML is acceptable for my application I
> should use Zope with any browser.
> 

I like this distinction, too:  sophisticated != HTML ;-)

Welcome aboard the Python train, Nemeth.  Let's all hope it's not the
hell-bound one;-)

<dare-i-say-it?-oh-sure!-"keep-plugging-away">-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
Author:  Teach Yourself Python in 24 Hours
----------------------------------------------



From pinard at iro.umontreal.ca  Sun Jan  2 10:02:29 2000
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 02 Jan 2000 10:02:29 -0500
Subject: Python not a Very High-Level Language?
In-Reply-To: lin li's message of "2 Jan 2000 13:31:09 GMT"
References: <84lvpj$sn5$1@news1.xs4all.nl> <84njut$t74$1@pollux.ip-plus.net>
Message-ID: <oqvh5cldoa.fsf@titan.progiciels-bpi.ca>

lin li <linli at computer.org> writes:

> >I've tried a lot of folding editors over the years, and I've never
> >found one that was as pleasant to use as, say, Word's outline editor.

> Just an idea, how about folding on the indentation levels?  We just have
> to forget about all these other languages then.

The Python mode for Emacs could implement Python folding, ideally.  Or else,
we could adjust one of the few available folding editors in Emacs to fit the
job (I quite prefer Allout over pure Outline already, yet I did not really
study most recent improvements to Outline -- I vaguely remember these are
internal, maybe to use overlays more than text properties, so to have many
folded "views" of the same text in different windows, simultaneously).

But the first task to achieve is to list what we want.  If Word's outline
editor (which I do not know) has interesting features, let's analyse the
interest and identify the wanted features.

It is quite possible that we have a very good start already.  If Outline
may be easily installed as a minor mode over Python mode (I guess yes), it
should be a trivial task to get Emacs Outline to rely on indentation level,
just by having Python mode locally set the value of outline-regexp, or such.

We might consider what this combination offers as granted, and list missing
features from that point, so making that wish list essentially shorter.
I'm not necessarily suggesting that the implementation of the wish list
should extend the "good start" described above, another start might be
required, if we were finding that Outline mode is not flexible enough.

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





From a.eyre at optichrome.com  Mon Jan 31 11:11:12 2000
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Mon, 31 Jan 2000 16:11:12 -0000
Subject: Newbie: How to create a global or static variable?
In-Reply-To: <874b3n$mhd$1@nnrp1.deja.com>
Message-ID: <001001bf6c05$cac026d0$3acbd9c2@peridot.optichrome.com>

> 1 - How do I create a global variable in a script that is in scope for
> all the functions defined in the script?

myGlobal = "spam"

def changeMyGlobal(what):
	global myGlobal
	myGlobal = what

print myGlobal
changeMyGlobal("eggs, bacon and spam")
print myGlobal

> 2 - Can I create a static variable with an initial assignment in a
> function?

No. You'll have to use either class attributes, or globals (or some
other Python hackery.)

--------------------------------------------
Adrian Eyre <a.eyre at optichrome.com>
http://www.optichrome.com 




From tk at gaia.gfy.ku.dk  Thu Jan  6 05:13:29 2000
From: tk at gaia.gfy.ku.dk (Thomas Knudsen)
Date: Thu, 6 Jan 2000 11:13:29 +0100
Subject: Simple cartesian product
In-Reply-To: <iacembwr7qg.fsf@vier.idi.ntnu.no>
References: <000801bf571c$0cf448a0$1a2d153f@tim> <iacembwr7qg.fsf@vier.idi.ntnu.no>
Message-ID: <Pine.LNX.4.10.10001061110050.6929-100000@gaia.gfy.ku.dk>

On 5 Jan 2000, Magnus L. Hetland wrote:

> Now - you're not going to get that "new millennium"-discussion
> started, are you? We are still in the same millennium, century, and
> decade as last year. The third millennium starts with 2001.01.01.

You should, however, use proper ISO8601 syntax and write 2001-01-01

This is actually quite important: We really *need* a consistent,
global format for numeric date representation. And since we actually
have it - *use it!*


Thomas

-- 
Thomas Knudsen, Research Geophysicist, Ph.D. | http://www.gfy.ku.dk/~tk/
National Survey and Cadastre - Denmark       | thk at kms.dk   tk at gfy.ku.dk
Geodetic Department, Rentemestervej 8        | phone:    +45 35 87 52 64
DK-2400 Copenhagen NV, Denmark               | fax:      +45 35 87 50 52





From sposhua at my.pc  Wed Jan 19 06:36:44 2000
From: sposhua at my.pc (Sposhua)
Date: Wed, 19 Jan 2000 11:36:44 +0000
Subject: Python Rocks!
In-Reply-To: <vndn1q340ez.fsf@betelgeuse.ccs.neu.edu>
References: <860ftk$bas@news.or.intel.com> <948195584.987404@news.adtranzsig.de> <388466D4.CC797CFC@earth.ox.ac.uk> <vndn1q340ez.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <Pine.WNT.4.21.0001191135140.206-100000@wks_3.e-shopwork.com>

On 18 Jan 2000, Justin Sheehy wrote:

> > ! - whats wrong with tabs?
> 
> They don't look the same in every editor, or with each individual's
> tab-width settings.

!? If you use ONLY tabs and not mix it with spaces (who the heck wants to jit
the space bar 4 times when U can hit tab once?) surely all will be well, or have
I missed something...?



From nolan_d at bigfoot.com  Sun Jan  2 22:11:28 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Sun, 02 Jan 2000 21:11:28 -0600
Subject: Python language spec
In-Reply-To: <slrn8703jp.g72.neelk@brick.cswv.com>
References: <E124x1i-00053N-00@ethereal> <slrn8703jp.g72.neelk@brick.cswv.com>
Message-ID: <E124xu4-0005Bq-00@ethereal>

>>>>> "Neel" == Neel Krishnaswami <neelk at brick.cswv.com> writes:

    Neel> Now, a reasonable question to ask is: why are there so many
    Neel> different kinds of expression when they all seem to reduce
    Neel> to the same thing?  (This seemed to me to be the thing that
    Neel> was confusing you. Let me know if I misunderstood.)

Exactly.

    Neel> The answer is that different operators have different
    Neel> precedences, and having separate rules is a way of ensuring
    Neel> that the parser will emit a syntax tree that properly
    Neel> observes the operator precedences that are expected.  A
    Neel> simple way of getting an intuition about this is to write
    Neel> down some arithmetic expressions and then manually reduce
    Neel> them using the rules of the grammar.

Ah. So, if I am to understand you correctly, this syntax is being used
such that precedences are expressed in a neutral fassion. That
wouldn't be necessary using Yacc since it offers %left and friends to
describe operator precedences. Is that accurate?

Thanks for the tip; I'm downloading the source to have a look at the
grammar.



From wang at rjka.com  Fri Jan 14 16:12:47 2000
From: wang at rjka.com (Huayin Wang)
Date: Fri, 14 Jan 2000 15:12:47 -0600 (CST)
Subject: class.__getattr__ behavior:  feature or bug? 
Message-ID: <200001142112.PAA27786@ws89.krumm0.com>

>>> class S:
...    def __getattr__(self, name):
...       if name == 'c':
...          self.a = self.a + 1
...          return self.a
...       else: return None
... 
>>> s=S()
>>> s
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: call of non-function (type None)
>>>



From tjreedy at udel.edu  Tue Jan 18 01:21:09 2000
From: tjreedy at udel.edu (Terry Reedy)
Date: Tue, 18 Jan 2000 01:21:09 -0500
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com>
Message-ID: <8610or$nsu$1@news.udel.edu>

"tye4" <tye4 at yahoo.com> wrote in message
news:860ftk$bas at news.or.intel.com...
> Having an 'end' block would help tremendously.
>
> if x == 10:
>     print 'x is 10'
> end if
> print 'Hello World'

You are welcome to use "# endif" or whatever as a training aid.

> Indentation is a confusing and lame way to separate what is inside a block
> and what's outside it.

Prediction: within three months you will have ceased either to believe this
xor to use Python.

TJR






From aahz at netcom.com  Tue Jan 11 14:28:58 2000
From: aahz at netcom.com (Aahz Maruch)
Date: 11 Jan 2000 19:28:58 GMT
Subject: [ANN] Multimethod.py -- multimethods for Python
References: <1264508924-1480880@hypernet.com>
Message-ID: <85g09q$o7e$1@nntp3.atl.mindspring.net>

In article <1264508924-1480880 at hypernet.com>,
Gordon McMillan <gmcm at hypernet.com> wrote:
>Doug Hellmann wrote:
>>
>> How is this different from function overloading, as implemented
>> in C++? I'm not a C++  user, but my impression is the effect
>> would be the same. Is that the point?
>
>Not at all. C++ (and Java) are limited to figuring out one override
>/ overload at a time. If you are dealing with base class pointers,
>you need to bounce the message around for awhile to resolve all of
>them. First, C++ will discover the proper derived class (the override),
>and then the proper overload.  Unfortunately, this means declaring
>scads of methods which don't do anything except let the language
>stumble through the resolution.

Hmmm...  So it's kind of like sparse overloading?  ;-)
--
                      --- Aahz (@netcom.com)

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

Have a *HAPPY* day!!!!!!!!!!


From skaller at maxtal.com.au  Sat Jan  8 09:56:57 2000
From: skaller at maxtal.com.au (skaller)
Date: Sun, 09 Jan 2000 01:56:57 +1100
Subject: control structures (was "Re: Sins")
References: <016a01bf57d3$cecb2900$01ffffc0@worldnet.att.net><14452.45436.641890.197509@beluga.mojam.com><005801bf5868$5d2dd0f0$0100a8c0@rochester.rr.com> <14452.60226.853060.116123@beluga.mojam.com> <03cd01bf5897$6d61dfa0$01ffffc0@worldnet.att.net>
Message-ID: <38775039.DA48C3D5@maxtal.com.au>

Emile van Sebille wrote:
> 
> FYI,
> 
> I get a significant improvement running Darrell's sample
> with string based exceptions (python -X...):
> 
> Time: 7.36000001431
> Time: 0.5
> 
> w/-X
> Time: 2.86000001431
> Time: 0.490000009537
> 
> but running Skip's there is no difference.
> 
> I-don't-know-what-it-means-ly y'rs

Strings are ready to throw. Classes must be instantiated,
which means searching for an __init__ method, etc -- 
extra overhead. And the 'matching' to find a handler
is more complex with class based exceptions, than a simple
string comparison.

-- 
John (Max) Skaller, mailto:skaller at maxtal.com.au
10/1 Toxteth Rd Glebe NSW 2037 Australia
homepage: http://www.maxtal.com.au/~skaller
voice: 61-2-9660-0850



From gerrit.holl at pobox.com  Fri Jan 28 07:25:09 2000
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Fri, 28 Jan 2000 13:25:09 +0100
Subject: memory
In-Reply-To: <86q8ee$2g7k$1@noao.edu>; from cheselka@tucana.tuc.noao.edu on Thu, Jan 27, 2000 at 08:05:34PM +0000
References: <86q8ee$2g7k$1@noao.edu>
Message-ID: <20000128132509.A565@stopcontact.palga.uucp>

Matt Cheselka wrote on 948999934:
> Hi,
> 
> Are there any issues with freeing memory in python?  Say for example I do a
> 
> 	d = array.array('i')
> 	for i in range (2000):
> 	    d.append(0)
> 
> Do I need to free 'd' before I quit the program?

Not necissary, but you _can_ do it with the del statement:
del d
removed the object and frees the memory.

regards,
Gerrit.

-- 
Please correct any bad English you encounter in my email message!
-----BEGIN GEEK CODE BLOCK----- http://www.geekcode.com
Version: 3.12
GCS dpu s-:-- a14 C++++>$ UL++ P--- L+++ E--- W++ N o? K? w--- !O !M !V PS+ PE?
Y? PGP-- t- 5? X? R- tv- b+(++) DI D+ G++ !e !r !y
-----END GEEK CODE BLOCK-----



From sholden at bellatlantic.net  Tue Jan  4 16:57:44 2000
From: sholden at bellatlantic.net (Steve Holden)
Date: Tue, 04 Jan 2000 21:57:44 GMT
Subject: calling functions from Tk
References: <842ma7$u2u$1@nnrp1.deja.com> <3864E324.11308615@callware.com>
Message-ID: <38726CED.F31D069D@bellatlantic.net>

Yeah, that's what we need: PythonDWIM!

<anticipating-I'll-get-a-SCSI-socket-in-my-neck-first>-ly yrs
 Steve

Ivan Van Laningham wrote:
> 
[a helpful reply to scoobysnax5336 at my-deja.com snipped]
> 
> <python-does-what-you-tell-it-not-what-you-mean>-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
> Author:  Teach Yourself Python in 24 Hours
> ----------------------------------------------


From mwh21 at cam.ac.uk  Thu Jan 20 16:00:25 2000
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 20 Jan 2000 21:00:25 +0000
Subject: Now it is getting wacky was( Re: A modest Error (I mean Proposal))
References: <XFMail.000120170754.mikael@isy.liu.se> <p8Kh4.2347$L4.74646@newsread2.prod.itd.earthlink.net>
Message-ID: <m31z7claqe.fsf@atrus.jesus.cam.ac.uk>

"Arinte" <shouldbe at message.com> writes:

> I change my code to look like this
> 
>  def IOCTL(self, command, dargs):
>   command1 = long(0)
>   if type(command)==type(2):
>    poss.write("IOCTL was int")
>    command1 = long(command)
>    poss.write("IOCTL Set to Long");
>   else:
>    if type(command) == type("  "):
>     poss.write("IOCTL was a string")
>     command1 = command
>    else:
>     if type(command)!=type(2L):
>      poss.write("IOCTL fail")
>      return -1
>   self.dump();
>   poss.IOCtl(self.devname,command1,dargs) #make this pass the ioctl command
> as an PossArg with a special name
>   print len("1234567")  #ADDED THIS LINE
>   return 0
> 
> Just to test len.  And now my program works???? If I remove this line or
> even just remove the print it doesn't work???
> Now what ??? If anything.  Another not here that critical print statement
> doesn't print the len of the string either...

Is it possible that you were importing an old pyc from somewhere? This
shouldn't happen, but it could explain what you were seeing
(particularly if the edits that took the py file away from what the
pyc said moved the lines around so the traceback didn't print the
right bits of the file).

Cheers,
M.


From mgushee at havenrock.com  Sat Jan  8 11:21:48 2000
From: mgushee at havenrock.com (Matt Gushee)
Date: 08 Jan 2000 11:21:48 -0500
Subject: Question about dictionaries (Question about is ?)
References: <3.0.5.32.20000105215635.0091d420@mail.dicksonstreet.com>
Message-ID: <m3wvpkv8ir.fsf@havenrock.com>

Felix Thibault <felixt at dicksonstreet.com> writes:

> >>> text = 'spam-a-lam-a-spama'
> >>> names = {'spam': text[:4]}
> >>> dict1 = {text[:4]: {text[-5:-1]: 'eggs'}}
> >>> dict2 = {names[text[:4]]: {names[text[-5:-1]]: 'eggs'}}
> 
> with the difference being:
> 
> >>> dict1.keys()[0] is dict1['spam'].keys()[0]
> 0
> >>> dict2.keys()[0] is dict2['spam'].keys()[0]
> 1
> 
> ...but I'm new to programming as well as to Python, so I'm not sure what
> this means-

Holy Dangerous Loops of Logic, Batman! Are you the reincarnation of
Nikola Tesla or what?

I'm not sure what it means either, but what seems to be happening is
that dict1['spam'].keys()[0] returns a reference to text[-5:-1] --
i.e.:

        'spam-a-lam-a-spama'
                    ^^^^
whereas dict2['spam'].keys()[0] returns a reference to text[:4], --
i.e.:

        'spam-a-lam-a-spama'
         ^^^^
... and the two just happen to have the same string value. 

>>> dict1.keys()[0] == dict1['spam'].keys()[0]
1

> does dict2
> take up less memory,

I don't know enough about the internals of Python to be sure, but I
can't imagine it would. dict1 references two different objects with
the value 'spam'; dict2 references a single spammy object twice; but
they both contain the same number of object references.

> or something, or is this just a Stupid Programming
> Trick?

Hah! Your next challenge is to find a use for this interesting
technique. If you can do that, it won't be just a Stupid Programming
Trick ;-)

-- 
Matt Gushee
Portland, Maine, USA
mgushee at havenrock.com
http://www.havenrock.com/


From movits at lockstar.com  Mon Jan 24 15:35:35 2000
From: movits at lockstar.com (Mordy Ovits)
Date: 24 Jan 2000 15:35:35 EST
Subject: OFFTOPIC: *-ly y'rs,
References: <Pine.BSF.4.10.10001241209150.67783-100000@bunny.monsterbymistake.com>
Message-ID: <388CB742.4C9678EF@lockstar.com>

Agent Drek wrote:
> 
> I know this is a stupid question but at least I'm not complaining
> about delimiters :) here goes ...
> 
> I have only ever seen these sign-offs on this list:
> 
> *-ly y'rs
> 
> Where did they come from and are they patented :)
> How can one join the signoff-ly club?
> 
> these-are-super-fun-ly y'rs,
> 
>     derek
> 

It's a highly infectious meme that was spread by the pernicious Tim
Peters. :-)

I'm-not-infected-ly yours,
Mordy
-- 
o Mordy Ovits
o Cryptographic Engineer
o LockStar, Inc.
--
There are two kinds of fool; One who says "This is old
and therefore bad," and one who says "This is new and
therefore better." -- John Brunner


From kamikaze at kuoi.asui.uidaho.edu  Mon Jan  3 03:07:46 2000
From: kamikaze at kuoi.asui.uidaho.edu (Mark 'Kamikaze' Hughes)
Date: 3 Jan 2000 08:07:46 GMT
Subject: Python not a Very High-Level Language?
References: <84i4bn$rqh$1@clematis.singnet.com.sg> <84ic6m$ao9$1@news1.xs4all.nl>
Message-ID: <84plci$ojt$1@news.fsr.net>

31 Dec 1999 13:48:06 GMT: in <84ic6m$ao9$1 at news1.xs4all.nl>,
boud at rempt.xs4all.nl spake:
>Ajith Prasad <aprasad at magix.com.sg> wrote:
>> http://www.oreilly.com/news/vhll_1299.html is an article by Greg Wilson
>But even so, he doesn't mention good stuff like the ways you can nowadays
>combine code and documentation (as in Qt), or work with diagrams. Class
>browsers already exist (even though I find most of them more confusing
>than helpful). It's only the presentation that's lacking, I think. But
>for that you don't need in intermediary language, although that would
>come in handy to make one environment work for more than one language.

  There's a much simpler way to achieve his purported goal: Java's
javadoc comments allow you to embed HTML, including image tags,
hotlinks, links to other javadocs with @see, formal descriptions of the
parameters and return types of your methods, and a great number of other
features...  And they're just plaintext comments, so you can still edit
your code with vi.

  Some form of this would definitely be a Good Thing for Python to
adopt.  I already use the javadoc format in my Python code, even though
I can't actually use the javadoc tool to make web pages from it.

  If you WANT a really high-tech editor, you could presumably make one
that interactively interprets javadocs as you're reading the code, but I
suspect that most programmers still use a plaintext editor on files of
source code for one excellent reason: It's the fastest way to get our
thoughts from our brains to the computer.

  Smalltalk attempted to make us all more efficient by using an
overarching IDE that did all the fancy stuff possible at the time
(mostly that damned annoying class browser), and it was, IMO, a complete
and utter pain in the ass when you actually wanted to write code.

  Someone else mentioned a folding editor - the old GFA Basic (it was a
structured variant with editor-ENFORCED indentation-as-control but also
keywords) editor/IDE (originally for the Atari ST, but it was ported to
PCs and Amiga) had function folding that operated on a single keypress
in that function...  This made reading and studying big programs MUCH
easier, because you could see the overview of all functions available,
then look inside any one you wanted quickly.  There have been
threatening noises in the function folding direction from vim lately,
too.

  The author's garbage-collection comments are too low-level a detail
for my current almost-intermediate-level Python knowledge, but if true,
that's something that should definitely be fixed, but it's hardly an
issue of high- or low-level.

-- <a href="http://kuoi.asui.uidaho.edu/~kamikaze/"> Mark Hughes </a>


From pollarda at lextek.com  Sat Jan 29 14:39:32 2000
From: pollarda at lextek.com (Art Pollard)
Date: Sat, 29 Jan 2000 11:39:32 -0800
Subject: Which Python book?
References: <85035b$1sb$1@bignews.shef.ac.uk> <850pt8$qv9$1@nnrp1.deja.com> <852g8h$75@gap.cco.caltech.edu> <slrn879km7.n3d.grant@grante.comtrol.com> <86fl0m$gmj$1@inputplus.demon.co.uk> <TVPi4.2485$K5.22193@newsfeed.slurp.net> <l40v8skdl4mdo6j98bq2pfaja7kr430adu@4ax.com>
Message-ID: <8752bh$anb$1@jaffa.itower.net>

Kevin McKinley <kevinmck at infi.net> wrote in message
news:l40v8skdl4mdo6j98bq2pfaja7kr430adu at 4ax.com...
> I'll be starting Python soon.
>
> Which tutorial are you referring to?

I don't think that I referred to a tutorial (except for a non-existant one.)

What I recommend (though I am just learning Python) is the following:

1) Quick Python
2) Python Essential Reference

Quick Python has IMHO a great explanation as to how Python works
and will help explain the syntax etc. which you need to get up and going.

Python Essential Reference is also a wonderful book and its explanations
are more terse but well written.  It is just what the name implies a
reference
and a good one at that.  Python's syntax from what I have seen so far is
pretty simple so once you get that down, I think the main issues are
learning
the tools which Python provides and their parameters which PER provides
along with brief but informative explanations.

The one thing I would like to see somewhere is a tutorial on using IDLE and
how to debug Python programs using it.  I haven't found this anywhere but I
sure would like to see it.

-Art




From tismer at tismer.com  Mon Jan 24 00:11:22 2000
From: tismer at tismer.com (Christian Tismer)
Date: Mon, 24 Jan 2000 00:11:22 -0500
Subject: Stackless Python 1.0 + Continuations 0.6
References: <mt2.0-27365-948450827@news.informatik.uni-bonn.de> <86f8uu$3ds$1@kopp.stud.ntnu.no>
Message-ID: <388BDEFA.8D57A66F@tismer.com>


Magnus Lie Hetland wrote:
> 
> Cool!
> 
> :)
> 
> Without *really* having read all the documentation, I venture a question...
> By using S.P. and continuations etc. -
> would it be possible to stop and pickle a thread? This might be very
> interesting for Mobile Agent Technology
> etc. (I.e. a running program can migrate over a network and resume its
> execution etc...)

This can of course be done. A little more structure will be
necessary to register a particular interpreter for pickling.
Continuation objects would need to learn how to save their
state as well. If there is interest, I will look into it.

Not so clear at the moment is how I would deal with real threads,
so this issue might be a little harder to solve.

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
D?ppelstr. 31                :    *Starship* http://starship.python.net
12163 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From knotwell at my-deja.com  Tue Jan 11 19:06:50 2000
From: knotwell at my-deja.com (knotwell at my-deja.com)
Date: Wed, 12 Jan 2000 00:06:50 GMT
Subject: a small problem
Message-ID: <85ggif$u8t$1@nnrp1.deja.com>

Hello all--

Is the following a bug?

>>> q = [(1,'dd'),(2,'ddd'),(3,'ddddd')]

>>> reduce(lambda x,y: x[0] + y[0],q)

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 1, in <lambda>
AttributeError: __getitem__

Meanwhile, map(lambda x: x[0],q) --> [0,1,2] as one would expect.
Similarly, filter appears to work as well.  If this isn't a bug, why
should reduce behave differently than filter and map?

Thanks.

--Brad


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


From frree at money.com  Tue Jan 18 00:12:04 2000
From: frree at money.com (frree at money.com)
Date: Mon, 17 Jan 2000 23:12:04 -0600
Subject: make money while surfing!!!             68457
Message-ID: <170100231204@money.com>

Everyone likes making easy money.  This place pays you 50 cents for every hour you browse the web!!!
Get even more if you refer people!  I just got a check for $36.00!!!

Sign up here:
http://www.alladvantage.com/go.asp?refid=IJE218







Poj8].zs.K


From nolan_d at bigfoot.com  Mon Jan  3 10:12:43 2000
From: nolan_d at bigfoot.com (Nolan Darilek)
Date: Mon, 03 Jan 2000 09:12:43 -0600
Subject: Python language spec
In-Reply-To: <20000103094852.B5669@xs4all.nl> (message from Thomas Wouters on
	Mon, 3 Jan 2000 09:48:52 +0100)
References: <E124x1i-00053N-00@ethereal> <20000103094852.B5669@xs4all.nl>
Message-ID: <E1259A3-0006qt-00@ethereal>

>>>>> "Thomas" == Thomas Wouters <thomas at xs4all.net> writes:

    Thomas> Not to spoil your fun at playing with yacc & parsing
    Thomas> python, but you dont really need to tackle yacc/parsing
    Thomas> yourself to embed an interpreter. Are you sure you want
    Thomas> Muq to take on only a subset of Python expressions, and
    Thomas> not a subset of python builtin modules/functions ? If the
    Thomas> latter, you dont need to do any parsing, you can build a
    Thomas> modified __builtins__ namespace and execute user code in
    Thomas> that.

I'm not building an interpreter; I'm building a compiler which
compiles a Python subset into Muq bytecodes. The resulting functions
should, in all likelihood, be usable in Forth, C, Lisp, and any other
future languages which the server may support. Believe me, I wish I
could just get away with embedding but ah well, this is definitely a
learning experience. :)

    Thomas> Of course, if you really want to tackle python code
    Thomas> directly, for instance to avoid common problemns like
    Thomas> 'a.apped(b);b.append(a)' and 'while 1: pass', the patch
    Thomas> can't be more than a small hint. If you do that, though,
    Thomas> keep me posted, i've been wanting to work around those
    Thomas> problems for a long time, for embedding purposes, and have
    Thomas> only been lacking time to do so ;P

Yuck, haven't considered that. I'll have to see how the server
architecture handles those types of things; hopefully jobs have
time/tick restrictions to prevent those types of conditions. :)
Anyhow, thanks for the pointer to the TinyMUSH patch, I'll give it a
glance.




From ullrich at math.okstate.edu  Mon Jan 31 14:58:58 2000
From: ullrich at math.okstate.edu (David C. Ullrich)
Date: Mon, 31 Jan 2000 13:58:58 -0600
Subject: Dynamic class construction?
References: <001301bf6c16$81e08ac0$3acbd9c2@peridot.optichrome.com>
Message-ID: <3895E982.C71DFB0B@math.okstate.edu>

Adrian Eyre wrote:

> >> There's already a module which does this...
>
> > I sorta thought there might be a module that does this, but I
> > couldn't find it.
>
> You didn't look very hard :)

    Actually the error I made was assuming that when I did
a full install of 1.5.2 the docs that appeared would be the
same as the official docs. This was stupid, since I've seen
before that it's not so. But I _did_ look as hard as appropriate,
in what was _supposed_ to be a right place.

    Evidently the contents of www.python.org/doc/current
are subject to change even when python hasn't changed?
Makes perfect sense, but it raises the question of whether
there's a way to tell whether one's local copy of the
docs is complete and current. Noting that it's 1.5.2
isn't enough.

>
> http://www.python.org/doc/current/lib/module-new.html
>
> > Um, I still can't find it. It appears to be a builtin module -
> > new.__doc__ says
>
> > '     Functions to create new objects used by the interpreter.
> >
> > You need to know a great deal about the interpreter to use this!',
> > which is helpful,
>
> I thought so :)
>
> > It's not in the "module index", I don't find any interesting
> > files contaning the text "classobj", etc.)
>
> What? This one?
>
> http://www.python.org/doc/current/modindex.html
>
> Ah. Be aware this module was only introduced in 1.5.2 IIRC.

    No, I wasn't referring to that modindex.html, I was referring
to the modindex.html that came when I installed 1.5.2.
You know, the modindex.html that (i) comes with "Release 1.5.2"
of the "Python Library Reference" (the quotes indicating cut&paste
quotes from the files) and (ii) skips from netrc to nnttplib.
_That_ one...

    It's obviously a good thing for the docs to be revised more
often than the program itself. But then the docs should have
separate version numbers. (If the docs _do_ have a separate
version number I've never found it, all I've ever seen is that
"1.5.2".)

    Hmm, I see it _does_ say "Python 1.5.2 Documentation
(revised)" on the site. Evidently the "revised" is supposed to
be a hint in this direction. But I don't see how a person is
supposed to know whether he already _has_ the "revised"
docs, or whether there's been a revision since the revised
ones he got, or what.

    Thanks.

> --------------------------------------------
> Adrian Eyre <a.eyre at optichrome.com>
> http://www.optichrome.com



From cpatti at black-racer.atg.com  Tue Jan 25 12:19:25 2000
From: cpatti at black-racer.atg.com (chris patti)
Date: 25 Jan 2000 12:19:25 -0500
Subject: Quick Python v.s. Learning Python
References: <387152EB.10DD0E17@Lextek.com>
Message-ID: <yu166wit6g2.fsf@black-racer.atg.com>

Art Pollard <Pollarda at Lextek.com> writes:

> I just purchased Programming Python.  As I am 
> totally new to Python, I learned from this book
> that I want to learn more.  Unfortunely, I don't 
> think this is the best book for a Python newbie.
> (Though from what I've seen, it makes a good 
> language reference.)
> 
> So right now, I am looking for another learning
> book.  The two which popout are:
> 
> Quick Python (In my local bookstore)
> Learning Python (Not in stock but obtainable)
> 
> Does anyone have any preferences / recommendations???
> 
> I have extensive C/C++ experience but would like a good
> book to help me break through the wall and reach that
> Ahhh Haaaa experience when the language becomes clear.
> (I still remember the Ahhhh Haaa from many years ago 
> when I figured out C's pointers... What a day. :-)
> 
> -Art

Well, a bunch of folks here have already chimed in for the Quick Python
Book.

For me and the folks I work with, we found _Learning Python_ to be 
vastly superior.

The Quick Python book is rife with typographical errors, some serious
enough to keep even the basic examples from being used.

_Learning Python_ was written by training professionals and it shows - 
the examples are all well laid out and the information is organized in
a way that makes it very easy to assimilate.

Just my $.02 

-Chris


From support at dnet.net.id  Sat Jan  8 19:18:58 2000
From: support at dnet.net.id (Support)
Date: Sun, 9 Jan 2000 07:18:58 +0700
Subject: Newbie : postscript printing
Message-ID: <sM0e4.1394$15.5706@news>

How to make a program to print my data (from database) to
postscript printer (under linux) ?.
does python provide a tools to print a document ?

Thanks.




From jima at aspectdv.com  Tue Jan 25 20:42:17 2000
From: jima at aspectdv.com (Jim Althoff)
Date: Tue, 25 Jan 2000 17:42:17 -0800
Subject: tabs do WHAT?
In-Reply-To: <000301bf6706$99fca7c0$632d153f@tim>
References: <86h5ep$iq$1@mach.vub.ac.be>
Message-ID: <4.2.0.58.20000125114146.00c1a740@mail.aspectdv.com>

In case this thread gives the impression that
it is Guido against the rest of the world ...

I LOVE "the whitespace stuff"!  I'm thrilled every time
I type code and don't have to put in nasty
syntactic delimiters (maybe I get my thrills
too easily  :-)  ).

By the way, here's an interesting experiment to
try if you have been using JPython together
with using lots of the Java APIs (a Swing-based
GUI application, for example).  Take a couple
hundred lines of your JPython code and
translate it into pure Java.  Even though much
of your code is already straightforward Java
API calls anyway, you will be amazed at how
much of your effort reduces to
a boring and tedious exercise in
adding in LOTS of cluttering syntax:
braces, semi-colons, and paren.s as far
as the eye can see.  After spending lots of
pleasant time running the compiler over
and over and over, you will actually succeed in
putting all of the delimiters in the right places!
What could be more fulfilling!  This simple exercise
will surely give you a wonderful appreciation for
"the whitespace stuff".

Jim


At 02:34 AM 1/25/00 -0500, Tim Peters wrote:
>[Warren Postma]
> > If Guido reads this, please note my personal feature request
> > for Python 2.0:  Make the "whitespace" stuff optional, for
> > people who like weird syntactical features only. I'll love]
> > you forever. Thanks.
>
>[Thomas Hamelryck]
> > This will not happen.
>
>Good bet.
>
> > People here (including Guido Himself) get extremely picky
> > when the use of indentation is criticised.
>
>Guido hasn't responded one way or the other to indentation criticism in
>years, and I doubt you'll find one of his old responses being "extremely
>picky" either (unbending, yes, but conviction isn't pickiness).
>
> > They do not realize that this feature is an important obstacle
> > for many people
>
>Neither is Guido an idiot <0.9 wink>.  Of *course* he knows it's a
>show-stopper for some people.  But so are many other things.  Every language
>designer knows this, and the successful ones figure out early there's no
>future in trying to make everyone happy.  There is a future in keeping a
>language's philosophy consistent for as long as possible, and the use of
>indentation to denote block structure wasn't an arbitrary choice in Guido's
>view of the world.  He realizes it's an important obstacle; do "they"
>realize it's an important manifestation of Python's philosophy?
>
> > (and since these people have stopped using python they are
> > obviously not posting here).
>
>Then what keeps this thread going year after year <wink>?
>
> > Indentation is great, but should not be part of the syntax
> > of a language.
>
>like-sex-is-great-but-should-not-be-part-of-life-ly y'rs  - tim
>
>
>
>--
>http://www.python.org/mailman/listinfo/python-list





From wlfraed at ix.netcom.com  Tue Jan 18 22:54:42 2000
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Tue, 18 Jan 2000 19:54:42 -0800
Subject: Python Rocks!
References: <860ftk$bas@news.or.intel.com> <YZRg4.1292$Ve7.28796@iad-read.news.verio.net> <862jqd$l9s@news.or.intel.com> <zb5h4.1605$Ve7.38526@iad-read.news.verio.net> <8631bt$1nu@news.or.intel.com>
Message-ID: <pzGFON0M2KHa8c0YY=BUjxhT79wF@4ax.com>

On Tue, 18 Jan 2000 16:41:06 -0800, "tye4" <tye4 at yahoo.com> declaimed
the following in comp.lang.python:


> 
> So you don't think this is a problem?? Hello!, read all the messages in this
> thread. Seems to me a lot of programmers have a problem with this.
> Thanks to this quirk, I now have to worry how an editor saves tabs to files.
> I have offered you many reasons why we need 'end' block support..
> You have not offered a single reason why this is not a problem..
>
	Sounds more like a problem with the selection of editing tools
and not the language itself... For my usage -- I stick with default tabs
and don't insert spaces; as long as I'm not editing something written by
someone else using a different convention I have no problem.

	Even DEC/VMS FORTRAN77 get bit by the difference between tabs
and spaces... It counts a tab as one character when determining line
length -- allowing for lines that extend beyond "screen" 72 where the
"same" line with spaces gets rejected for having code in a "comment"
zone.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <


From eddy at vortigen.demon.co.uk  Wed Jan 19 18:40:11 2000
From: eddy at vortigen.demon.co.uk (Edward Welbourne)
Date: Wed, 19 Jan 2000 23:40:11 +0000 (GMT)
Subject: metaclass doc
In-Reply-To: <200001191604.LAA18944@eric.cnri.reston.va.us> (message from
	Guido van Rossum on Wed, 19 Jan 2000 11:04:54 -0500)
References: <E12Ax4C-0002AB-00@lsls4p> <200001191604.LAA18944@eric.cnri.reston.va.us>
Message-ID: <m12B4hv-00007oC@vortigen.demon.co.uk>

G>>> Have you read my (very old) metaclasses article?
G>>> ... syntax ... slightly different ... the rest is the same.
E>> Non-trivial differences exist ...
G> Like what?

  * a generator creates the object before the suite gets executed; so it
     * only gets, as keywords, any things the interpreter feels like
       feeding it - e.g. __name__, __file__ - that are known at the
       start of the generator statement (it has, of course, the option
       of discarding those, filtering them or adding to them); and the
       name of the new object isn't anything special, except in so far
       as the interpreter *has* provided __name__
     * gets to hang functionality methods off it that influence what
       happens when you do, e.g., the assignments in the suite
     * gets to initialise any attributes it wants to on the object,
       which the suite will be able to read as local variables (it may
       also be able to modify them, if the functionality methods provide
       for this)
     * and the namespace in which the suite gets executed borrows from
       the objects used as bases, so sees their attributes as (readable)
       variables (whether it can modify or - more likely - hide these is
       as for stuff set by the generator)

  * the interpreter is no longer obliged to *know about* the names
    __class__, __bases__, __dict__ and __getattr__ (nor about the added
    names, __lookups__, __meth__, __data__), let alone pay any attention
    to them when deciding how to handle [e.g. call] an object [but it
    may need to know about a new name, __dir__, to implement dir() on
    objects for which __dict__.keys() may be inadequate; but I am
    inclined to argue for that anyway - such objects already exist]

  * a generator provides one object as the generated object and another
    object [usually borrowing attributes from and modifying attributes
    on the generated object] in whose namespace to run the suite,
    allowing folk to build things with (typically) greater liberty in
    what can be done *during* the suite as opposed to what can be done
    to the object *thereafter*: e.g., setting a type constraint on it
    which can't, subsequently, be subverted by simply setting a more
    liberal type-constraint in its place.

I grant the business of having two objects may be avoidable, but only at
the expense of having the interpreter see existing namespaces as
  * *four* distinct types for which it has to do
  * attribute lookup in *six* different ways (built-in attribute,
    dictionary, borrow, inherit, getattr, implicit import, any others ?)
    with each type supporting only some of these
  * calling in *three* incompatible ways (it's a class and we have a
    burried way of calling it that isn't the value of it.__call__; it's
    an instance, so it.__call__; it's a module or package so refuse)

where the generator idiom provides for *one* type on which attribute
lookup is *always* done by calling the function packaged by that type
(with one argument, the attribute name); and calling a namespace-object
is *always* done by looking up the object's __call__ attribute and
calling that (if its type is function, you now know how to; otherwise,
recurse).  That the function called (to do lookup) differs (internally)
according to the flavour (which used to be type) of the namespace
involved is a matter the generator deals with and the interpreter
doesn't have to know - or care - about.  [For __call__, read (in less
severe degree) any other magic name.]

This doesn't rule out talking me into settling for metaclasses (but know
that, now I've understood how *good* they are, I *want* safe tunnels) in
place of generators, it's just that I *really* don't like a whole lot of
discriminatory practices that are going on, like the fact tha defining
__call__(*args, **what) in a module doesn't make the module callable:

-- Begin: trial.py
def __call__(*args, **what):
    print args, what
-- End: trial.py

Python 1.5.2 (#5, Oct  4 1999, 13:36:16)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import trial
>>> trial(1, 2, 3, hello=4, goodbye=5)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
TypeError: call of non-function (type module)

I am *offended* by that - I take it *personally* (you don't have to).
Someone drummed intolerance of unfair discrimination into me *young*.

Consider the implications of defining __getattr__(key), __delattr__(key)
and __setattr__(key, value) in a module and having folk unable to mess
with its contents other than in so far as the last two permit.  Some of
the folk on types-sig might care *a lot* about that.

--

I am puzzled by one thing about metaclasses; I can't quite express it
clearly (and you might have to read init for call and rearrange
somewhat), but this is roughly its form:

def A (B): pass

Now, B.__class__ is present, and callable, so suppose it has a .__call__
So if we try to look up B.__call__, don't we get 
lambda *args, **what: B.__class__.__call__(B, *args, **what)
or, rather, *why* don't we ?  (Or: why *isn't* this badly broken ?)

Or, if you have a way out of that, if B.__class__ provides B with
__call__ in the regular way, (and B hasn't done the same for its
instances) B's instances are going to see B.__call__ inherited from
B.__class__, so why don't they then proceed to inherit *this* [with,
once more, `potentially unwelcome' results]

I'm guessing the answer is that inheritance of methods from one's class
contains a special case (forgive me, but I read that as *prejudice*) of
some sort [e.g. the attribute of the class must be an unbound method ?]
that prevents this.  I'm wondering what inconvenience this might
introduce [e.g. when what I *want* is a method which will get invoked
with args (metaclass, class, instance, ...), though only a
meta-meta-class would want to provide such a thing; but drop a layer of
meta (and the first arg) to get a more plausible case ...]  Of course,
the special case may be in `how do I call this' rather than in `how do I
inherit from my class'.  Can you see why I don't like it, either way ?

This is the sort of ugly question that forced me to the conclusion that
a class should have distinct __meth__ and __data__ sub-objects.

After that, I became calm and happy, albeit the entirety of the rest of
the generator proposal got forced on me by it.

I didn't invent generators because I *want* them: I did so because
they're what classes *demand* if namespaces are to be treated ... justly.

	Eddy.
--
Guido: thanks for being good at asking The Right Questions, too.



From Ronny.Dewaele at village.uunet.be  Sun Jan  9 16:35:04 2000
From: Ronny.Dewaele at village.uunet.be (Ronny Dewaele)
Date: Sun, 09 Jan 2000 22:35:04 +0100
Subject: Newbie : postscript printing
References: <sM0e4.1394$15.5706@news>
Message-ID: <3878FF08.9AFD2CF6@village.uunet.be>

Of course, there are several ways to get the output in a file :
- redirect the output of your program (allowing the use of print to
generate the output)
      e.g. python myapp.py > outfile
- create the file in the python application :
      e.g. f=open('outfile','w')
  and write to it using write commands (e.g. f.writeline(outputstring) )
Ronny

Support wrote:
> 
> How to make a program to print my data (from database) to
> postscript printer (under linux) ?.
> does python provide a tools to print a document ?
> 
> Thanks.


From python at rose164.wuh.wustl.edu  Sun Jan 30 09:43:17 2000
From: python at rose164.wuh.wustl.edu (David Fisher)
Date: Sun, 30 Jan 2000 14:43:17 GMT
Subject: [Newbie Problems] AttributeError from CGI script?
References: <5650A1190E4FD111BC7E0000F8034D26A0F2A1@huina.oceanic.com> <slrn898d3h.86h.steve_w@localhost.localdomain>
Message-ID: <20000130.14431716@mis.configured.host>

Ah, well that's easy then.  In version 1.4, the string module doesn't 
have a replace function.  After importing string, a dir(string) will 
give you a list of the module's attributes.  Replace will not be in 
that list, at least until the mythical upgrade <wink>, hence, the 
AttributeError.
Here's the definition of replace chopped from string.py (version 
1.5.2):

def replace(str, old, new, maxsplit=0):
	return joinfields(splitfields(str, old, maxsplit), new)

joinfields and splitfields are defined in 1.4, so this should work.

On a side note, this function, while being defined in string.py is not 
actually used.  Along with most/all of the other functions in 
string.py, it is replaced at the end of the module by another function 
named replace from the strop module, which is written in C.  Now why 
this is done is a complete mystery to me.

David Fisher

>>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<

On 1/30/00, 6:34:14 AM, steve_w at localhost.localdomain () wrote 
regarding Re: [Newbie Problems] AttributeError from CGI script?:


> I checked and I don't appear to be assigning anything to a variable
> call string.  Thanks for the suggestion though.  The server I'm
> working from is using Python 1.4 on Solaris 2.6.  They promised
> me they'd upgrade a few weeks ago so it'd match my documentation
> and local version but so far no dice.

> To make matters worse I'm now getting some different error which seems
> to be from the CGI Wrapper software.

> Steve Wall